|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
The Effects Of Animated Tex tual Instruction On Learners' Written Production Of German Modal Verb Sentences Elizabeth A. Caplan A dissertation submitted to the Faculty of the University of South Florida in partial fulfillment of the requirements for the degree of Doctor of Philosophy In Second Language Acquisition and Instructional Technology Carine Feyten, Co-chair William Kealy, Co-chair Robert Dedrick Jeffra Flaitz James White July 12, 2002 Tampa, Florida Keywords: Animation, Dual Coding, German Grammar, Computer Assisted Language Learning Copyright 2002, Elizabeth A. Caplan
ii Dedication To those who thought bilingualism was impossibl e to achieve past the threshold of childhood.
iii Acknowledgments I am grateful to my husband for his encouragem ent and understanding. I thank my father for his support in all things. I thank my mother for her awesome inspiration and the unseen help she has given me. I thank my teachers and colleagues for their exciting commitment to open-minded inquiry.
ivTable Of ContentsList of TablesviiiList of FiguresixAbstract xChapter I Introduction1Overview Of Relevant Research 2Explicit vs. Implicit Grammar Instruction 2Computer-Based Instruction 3Animated Grammar 4Importance of the Study 5German Language Education 5Process Taught by the Treatment 6 Research Questions 8Organization of the Study 9Glossary of Terms 10Chapter II Literature Review 13Introduction 1 3Theoretical Background and Pertinent Research Findings 13The Role of Grammar in Foreign Language Teaching 13Premises of Grammar and Approaches to Teaching It 14Structural vs. Natural Approach 15Form-Focused Instruction 18Implicit vs. Explicit Grammar Instruction 18Computer-enhanced Grammar Instruction 19Computer-Based Instruction 21Dual Coding Theory: Mental Representations 21Logogens and Imagens 23The Shape of the Letter E" 24
vDual Coding Theory and SLA 26Dual Coding and Total Physical Response 28Animation Research 29Substantive Issues 36Functions of Animation 36Cognitive Theory of Multimedia Learning 40Multimedia Principle 42Spatial Contiguity Principle 42Temporal Contiguity Principle 42Coherence Principle 42Modality Principle 43Redundancy Principle 43Personalization Principle 44Congruency 44Mental Modal Building vs. External Imagery 45Processing Speed and Prior Knowledge 45Conjoint Retention Theory 46Dual Task Objectives 47Semantic Matching 48Spatial Conventions 49Focus Shift 49Redundant Narration 50What the Present Study Will Add to the Field 52Animation and CBI 52German L2 Pedagogy 53Summary 54Chapter III Method 55Introduction 55Research Questions 55Participant Information 55
viInstrumentation 63Content Validity of Instruments 63Pretest 63 "Will "6 4" Mchte "6Posttest 65Qualitative Instruments 66Instructional Treatments 67Content 67Development 68Imagery Cueing 68Spatial Conventions Study 69Narration 70Pacing 70Content 71Meaning 71Conjugation 71Word Order 72 Data Collection Procedures 72 Scoring 73 Chapter IV Results 75 Research Question 1 77 Effect Sizes 78 Research Question 2 79 Effect Sizes 81 Conjugation 81 Word Order 81 Posttest Survey 82 Chapter V Discussion 84 Limitations 85
vii Sample Size 85 Content Area 85 Ceiling Effect 85 Multimedia Principle 86Coherence Principle 87Redundancy Principle 87Implications 89 Recommendations 89References 92 Appendices 107 Appendix A: Note To Participants 108 Appendix B: Pretest Survey Responses 109 Appendix C: Relevant Websites 110 Appendix D: Pretest 111 Appendix E: Posttest 113 Appendix F: Qualitative Instruments 115 Appendix G: Spatial Conventions Study 116 Appendix H: Instruction Screens With Narrative Script 119 Appendix I: Score Sheets 126 Appendix J: Posttest Survey Responses 128 Appendix K: Comment Rating Scale 130 About The Author End Page
viii List of Tables Table 1 Number of Participants by Section 57 Table 2 Sample Sizes for Class Sections by Presentation Format 57 Table 3 Experience With German by Presentation Format 58 Table 4 Time Passed Since Last Course of German Study of 1-2 Semesters 59 Table 5 Time Spent Living in Germany 60 Table 6 Self-Rated Ability for German 61 Table 7 Experience With Second Languages Other Than German 61 Table 8 Experience With Instructed Second Language Learning 62 Table 9Internal Consistency Reliabi lity (Cronbach Alpha) for Pretest Scores by Group 64 Table 10 Internal Consistency Reliabi lity for Posttest Scores by Group 66 Table 11 Pearson Product Moment Correlations of Pretest Subscales 75 Table 12Descriptive Statistics for Pretest Total Scores By Section and By Method 76 Table 13Descriptive Statistics for Po sttest Total Scores By Section and By Method 78 Table 14 Pearson Product Moment Correlations of Posttest Subscales 79 Table 15 Mean Achievement Posttest Scores of Treatment Groups 80 Table 16F-values and Significance Levels for Series of 3 ( Section) x 2 (Method) Posttest ANOVAs 82 Table 17 Mean Ratings of Participant Comments 83
ixList Of FiguresFigure 1 Mental Representations: A Dual Coding Approach 22Figure 2 Conjoint Reten tion Theory for Map Features 47Figure 3 Modal Verb Chart Showing Pa tterns and Irregular Conjugations 72
x The Effects of Animated Textual In struction on Learners' Written Production of German Modal Verb Sentences ELIZABETH A. CAPLAN (ABSTRACT) This study investigated the eff ects of animation for a techno logy-assisted German grammar presentation on modal verbs. The premise wa s that many intangible concepts of dynamic grammar involve syntactic components that possess visuospatial characteristics. It was further speculated that these characteristics could be more effectively represented by animated versus static instructional presentations. The supposition that animation would lend pedagog ical advantage was supported by dual coding theory (Paivio, 1971, 1990), which pos its two functionally separate representational systems, the verbal and the nonverbal, with dynamic mental im agery residing solely in the nonverbal system. The strength of dually coded information is that it is represented in both subsystems and, due to referential associations that cross between th e two, is more easily re tained and recalled. When verbal explanations are accompanied by illustrati ons depicting their content, it can provide external support for the learner's me ntal simulations of that content. Under two treatment conditions, 44 university students of beginning German (GER 101) received large-screen multimedia instruction concerning the meanings and conjugated forms of German modal auxiliary verbs, and the grammatical rules which govern sentence structure. The independent variable was the type of visualiz ation: static or animated text. The dependent variables were participants total test scores as well as their indi vidual scores on each of two task types: conjugation and word orde r. In addition, a posttest survey asked participants for their opinions of the instructional treatments. Participants in both treatment groups achieved high scores on the posttest with no significant difference between them; however, the posttest survey showed that the groups did differ significantly in their opinions of the treatments, with those in the animated group reporting more positive reactions to the pres entation. Detailed planning and lengthy preparation of both treatments may explain the high scores for both groups, and the elementary nature of the content may also account for the resulting ceiling effect. Animation should be studied further, especially with respect to more complex tasks, as well as in concert with other aspects of multimedia, such as interactivity, user-contro l, practice, and feedback.
1 Chapter I Introduction This study investigated the effects of animation for a technology assisted German grammar presentation on students' understanding of modal verbs. The premise was that many intangible concepts of dynamic grammar invol ve syntactic components that possess visuo spatial characteristics. It was further specul ated that these characteristics could be more effectively represented by animated versus st atic instructional presentations. The research hypothesis was that students woul d learn the foreign language grammar concepts better with animation. The value of the animation was empi rically assessed in or der to determine its effectiveness within the instructional design. The rationale for using animation to depict concepts of German grammar lies in the fact that certain aspects, such as conjugation and word order rules, can be described as dynamic, because they involve changes in form or pos ition. The supposition that animation would lend pedagogical advantage was supported by dual coding theory (Paivio, 1971, 1990), which posits two functionally separate representational syst ems, the verbal and the nonverbal, with dynamic mental imagery residing solely in the nonve rbal system. The strength of dually coded information is that it is represented in both subs ystems and, due to referential associations that cross between the two, more easily retained and recalled. When verbal explanations are accompanied by illustrations depicting their cont ent, it can provide external support for the learner's mental simulations of that content. The factor of efficiency is important to mo st of the state university systems across the United States. Publicly funded systems of higher education are standardized in terms of the length of courses that are considered comparable across universities. This means that virtually all United States public universities allot the sa me amount of time (14-16 weeks) in order to complete equal expectations for each comparable course, such as Beginning German I. The use of multimedia enhanced grammar instruction is an attempt to provide conceptual explanations that are understandable for indivi duals by incorporating a variety of perceptual learning styles. It
2 is also a means to render instruction consiste nt, replicable and, as informed by research, efficacious. Much of the research into individual aspects of multimedia enhanced instruction is aimed at discovery of effectiveness and effici ency: What can be effective within given time constraints? In foreign language teaching, the dynamic properties of the classroom, live bodies and spirited interaction, should be exploited to facilitate conversation and communicative exchange and to promote verbal language acquisition. Ti me is short, and opportunities for speaking practice are few and precious. Research has shown in which conversational ability is facilitated through the negotiation of meaning, the give and ta ke that speakers partic ipate as they exchange concepts and try to express themselves (Ga ss & Varonis, 1985; Kramsch, 1983; Long & Porter, 1985; Pica, 1994; Young, 1983). Negotiation of meani ng takes time; more importantly, it takes at least two. It is not something that can be e xpected of learners to accomplish on their own. Therefore, the classroom hour should be devoted to the type of activit ies that require dual or multiple learner participation and the instru ctor's actual presence (Savignon, 1991; VanPatten, 1991). Other aspects of language lear ning are equally important, but may be served in other, more individualized, instructional settings. Gram mar is one of those aspects, and the multimedia computer environment is one of those setti ngs. The research presented here is classroom centered; it took place w ithin a typical classroom of language learners. However, the questions investigated were part of a la rger inquiry into how grammar instruction can be effectively presented in a multimedia format, which can ultimately be made available for use outside of the classroom. Overview of Relevant Research Explicit Vs. Implicit Grammar Instruction Some second language acquisition (SLA) theoreti cians argue persuasively for an implicit approach for inductive learning of gramma r principles (Krashen, 1983, 1985; VanPatten, 1987, 1996; Williams, 1999). In a similar manner that one has acquired the rules of one's own native language, the structures of th e target language may be glean ed through exposure, usage and habit, and, as the argument goes, researchers beli eve that retention (i.e., overall acquisition) is enhanced by this method. With an inductive appr oach, grammar is not explicitly explained,
3 instead the rules are induced th rough carefully graded exposure to and practice with examples [ ] (Gollin, 1998). Researchers McCarthy and Ca rter agree that grammar can be learned inductively, but they refer only to the gramm ar of the spoken language (Carter & McCarthy, 1994; McCarthy & Carter, 1995). More often than not, however, researchers argue for an ex plicit approach to grammar instruction (DeKeyser, 1994, 1995; Nagata, 1997; Seliger, 1975; Sheen, 1996), and they note that, although first language acqui sition is primarily an inductive matter, many of the more advanced features of a language are learned expl icitly -at school, for example -even by native speakers (Hall, 1998). Grammar rules and structural principles ma y be induced, but the inductive process is slow and acquisition deepens onl y over the extension of time. On the other hand, grammar explanations delivered as classhour content (a) take time for presentation, (b) learners mental imagery mechanisms vary, (c) teaching strategies needed to address the va riety of learning styles are numerous, and (d) for many, the subjec t matter has a reputation for being dull. Computer Based Instruction Computer based instruction (CBI) has been shown to be above all efficient (Aust, Kelly, & Roby, 1993; Curtin & Shinall, 1984; Morse, 1999; Ramirez & Revard, 1998) and its assets address the above four perils in a most efficient way. Its greatest potential is simply its ability to pull the grammar out of the classroom, leav ing the entire class time for communicative interaction (VanPatten, 1993). CBI gives grammar a new home; a home where all visitors have custom access. Custom in terms of schedule (computer instruction modules can be presented at flexible times and places through various, convenient delivery modes); custom in terms of pace (students can control the speed of instructi on and the timing of practice pace) (Matoon, 1991); custom in terms of responsive feedback (the co mputer can measure and manipulate the learners response time to facilitate incr eases in recognition and recall ra tes) (Hativa, Sarig, & Lesgold, 1985; Rieber & Parmley, 1992; Tennyson, 1985). Grammars "new home" also has custom access in terms of student learning styles such as: auditory input pref erence (Viteli, 1989), textual/verbal preference (Leutner & Plass, 1998), learner controlled time on task (Okolo & Haye s, 1996), and visual representation or mental modeling (Justice, 1999). These ar e essentially the various form s of language input that are
4 available through the avenues of multimedia: sound, text, picture and animation. The various modalities available through multimedia can be offered sequentially, simultaneously, or the learners can make choices according to their own preferences. Animated Grammar A multimedia grammar presentation may be desi gned as if it were a theatrical production. In a theater, the spectators are manipulated as a single entity. Each enters as an individual with separate concerns and different amounts of complex thoughts occupying their minds. One reason that they come into a theater in the first place is to clear their minds and to allow new input (i.e., the story, characterizations and concepts) to enter their minds and to occupy their thoughts nearly exclusively. To apply a theatrical approach to instruc tion speaks to the importance of attention upon retention (Gregg & Farnham, 1975; Hwang, 1999; McGallagher, 1975; Schleppegrell & Oxford, 1988; Tomlin & Villa, 1994; VanPatten, 1990; Willia ms, 1999). The theatrical directive always concerns the audiences focus. Every pr oduction value (movement, dialogue, light, sound or special effect, etc.) controls the at tention of the spectator. Apply this directive to the features of a multimedia grammar presentation and one gets what, at first, seems to be just a lot of bells and whistles or gratuitous attention grabbers. Animation, whether it is moving text and pi ctures or simply interesting transitions between visual tableaus, can be an effectiv e attention grabber, which lays the necessary foundation for learning (Gagne, 1968, 1970). While different learning st yles and sensory preferences determine where and how an individua ls attention will be most attuned or attended, anyones attention can be "caught," directed and manipulated at leas t for an instant. Since time is an infinite series of single instances, a pe rson's attention may be caught and focused, recaught and refocused, in essence, held, for a relatively substantial period of time. Each element of a multimedia presentation adds hours and expense to the production process, so the gratuitous use of such extras is prohibitive. Yet, foregoing elements necessary to enhance instruction should also be precluded. The details of a pedagogic multimedia presentation (text, graphics, sound, etc.) must be ex amined individually as well as in concert with each other within the targeted in structional context. In as far as possible, examination of the pedagogic value of each element should take pla ce before resources are committed to its use in
5the production. Some elements may have more inst ructional value in some disciplines than they do in others. Sound, for example, may be superfl uous in math instructi on yet crucial in music theory. Animation has been shown to have pe dagogic value in math and physics instruction (Poohkay & Szabo, 1995; Rieber, 1996, 1988, 1989a, 1989b) but it could have more, less or even negative value in the context of fo reign language grammar instruction.Importance Of The StudyGerman Language Education German is a highly structured language in that it has many rules that have few or no exceptions. It has an inflected system of verbs, which means that a given verb takes a different form depending on which pronoun with which it is paired. The clearest English equivalent is found with the verb to be (German: sein ). Paired with the pronoun I , the verb takes the form am, but together with he the verb must be changed, or conjugated to the form is. In German, all verbs undergo such a change. Conjuga tion is one of the invariant rules of German grammar. The rules concerning placement of th e verb are mutable according to prescribed circumstances. That means that verb placement rules are also invariant, but they change only in response to other rule-constrain ed changes in word order. German SLA requires language learning effort s common to all foreign languages, such as vocabulary learning and development of listening comprehension skills, but it also requires a degree of mastery over the transformation rules of German sentence structure. The traditional display of grammatical information is typically st atic, textual explanations with examples printed on a page, to be read silently by each individual learner. Ma ny aspects of grammar involve a process or a transformation whereby specific types of words (e.g., verbs) undergo a change, either in form or in placement within the sentence. Such a change or process can be seen as a series of steps which can be imparted clearly in silent, textual form, as textbooks do. However, the production of many grammatical transformati ons involves a single, unified process. A textbook presents a series of nonse quential steps that a learner, through practice, is expected to be able to merge into a smooth dynamic proce ss. Animation can be used to present the information as a single, unified process right from the start. The merging of steps into a smooth transformation may be visually portrayed. Because it stimulates mental visualization of the
6dynamic process, the effect of textual animation is expected to be signifi cant in helping learners to perform grammatical transformations accurately. Process taught by the treatment. The instructional treatment s of the present study cover the four-step process of word or der transformation triggered by th e addition of a modal auxiliary or helping verb. A modal verb helps anot her verb by expressing its mood. Any verb can combine with a modal to express: obligation (must: mssen ), permission (may: drfen ), volition (desire: mchten ), intention (want: wollen ), ability (can: knnen ) or subjunctive obligation (should: sollen ). The addition of a modal verb to an English sentence follows the usual rule for verb placement (after the subject), and then it causes displacement of the main verb by one place, e.g., I go to town often I must go to town often. In a German sentence, the modal placement also follows the primary verb placement rule (second position), but the main verb is displaced out of the sentence and replaced at the very end of the clause, e.g., Ich gehe oft in die Stadt Ich muss oft in die Stadt gehen. In both cases, English and Germa n, the main verb reverts to its infinitive form, only the word order transformati on presents a strong contrast between languages. The steps involved in the tr ansformation are as follows:1. Conjugate the modal infinitive to agr ee with the subject or subject pronoun. 2. Finite modal verb assumes the main verb position (second place for statements, first place for questions).3. Dependent infinitive is displaced and pushed to the final position of th e clause (last place). 4. Dependent verb is restored to its original infinitive form. Using the following sentence as an exampl e, the four steps amount to a process of transformation: Mein Onkel kommt nach Hamburg. (My uncle comes to Hamburg.) Step 1. Add conjugated form of modal verb knnen kann Step 2. Modal assumes main verb position (second) Mein Onkel kann Step 3) Displaced dependent verb is pushed to last place and Step 4. restored to the infinitive form Mein Onkel kann nach Hamburg kommen (My uncle can come to Hamburg).
7 A fifth step occurs when the verb has a separable prefix. When the infinitive stem is pushed into last place in the sentence, its position is always di rectly adjacent to its ow n prefix. Therefore, the stem is reunited with its separated pref ix and becomes a single word once again. E.g., ankommen = to arrive. Mein Onkel kommt heute morgen an Mein Onkel kann heute morgen ankommen (My uncle can arrive this morning). These steps are only superficially divided into five. They could readil y be seen as a single seamless action. Each step is represented by a verb: 1) Conjugate2) Position3) Disp lace and Push4) Restore5) Reunite This gives the procedure the charac teristics of a physical process. As such, it is a process that is congruent to animated illustration. Because the transformation of sentence structure is a dynamic process, it is content that call s for graphically illustrated physic al motion, or nonverbal as well as verbal encoding. The general concept of modal verbs is not di fficult for a learner with English as a native language or as a high-level L2. English also ha s modal verbs, and many of them are partial cognates with their German counterparts. Howeve r, there are some differences in meaning of modals between the two languages and these differences are not often addressed in the lessons of most textbooks on the topic. In a survey of the seven leading college-level textbooks of beginning German ( Deutsch Heute, Deutsch Na klar, Deut sch Plus, Kontakte, Neue Horizonte, Wie Geht's, & Vorsprung ), the better part of all lessons on German modal verbs concerned their irregular conjugations. In all of the textbooks review ed for this research, th e topic of word order was never given more than a slight paragraph, a nd rarely with any mention of movement, change or transformation. None of them offered a descript ion of a process, which could cue imagery of a transformation event, with the exception of Wie Geht's which depicted motion through the use of an arrow pointing from the inflected verb toward the end of the sentence. The lack of emphasis on modal verb syntax prompts instructor s either to present s upplementary material, or to rely on the learner's ability to di scover the word order rules inductively. This study began with the creation of a pr esentation designed to supplement the textbook used (for years) by the unive rsity where the study took place. This book was one of the few to mention word order, but the explanatory text made extensive use of metalinguistic terminology. "The modal verb is always the inflected verb in the sentence. The complementary infinitive
8(which is the second part of the predicat e) comes at the end of the sentence" ( Deutsch Heute 2000, p. 81). Since no existing supplementary material was found, that emphasized the word order aspect of modal verbs, a presentation was created to animate the visual samples of modal verbs in order to depict, rather than just expl ain, the transformational process involved in their use. The animated presentation was time-consumi ng and expensive to produce and to present. In addition, while the presentation was well received and student test scores seemed to reflect benefits from the instruction, it was not clear whether the animati on was the beneficial feature, or merely the computer-enhanced presentation. Ther efore, this study was undertaken to determine the extent to which computer animation supports learning for the word order rules of German modal verbs.Under two treatment conditions, 44 university students of beginning German (GER 1120) received instruction concerning the meanings a nd conjugated forms of German modal auxiliary verbs, and the grammatical rules that govern sent ence structure. Both conditions utilized large screen projections of multimedia presentations with audio and color slides (PowerPoint). The independent variable was the type of visualization: static or animated presentation of sample sentences in text form. The research questions that follow sought to meas ure the effects of the different instructional treatments upon participants' ability to pr operly rewrite German indicative sentences adding modal auxiliary verbs.Research Questions1. What is the effect of animation of sample sentences depicting a grammatical process, explained in a concurrent na rrative (audio) presen tation, on test scores measuring a language learners ability to rewrite German indicat ive sentences adding modal auxiliary verbs, compared to static presentation of the same sample sentences? 2. Is the effect of animating the sample sentence s different for each of the two discrete tasks (conjugation and word order) invo lved in the rewriting process? To illustrate the premise of question 1, imagine the words of the sample sentences in a textbook chapter, concerning word order, coming to life and moving around in accordance with the silent, textual explanations. Does the animation of words aid significantly in understanding the morphosyntactic process? More precisely, doe s this "assisted understanding" result in the learner's increased production of grammatical sentence transformations involving that process?
9 The second question seeks to discover someth ing about the particular tasks for which animation can facilitate learning. The dependent variables were participants total test scores as well as their individual scores on each of two criterion measures: conjugation and word order. The original aim of this research was to determine whether the use of animation added instructional value commensurate with its expense. Over the course of the study's design and implementation, the expense of using animati on has decreased considerably, while access to technology in the classroom has become nearly common-place. It is no longer a question of whether to use animation, but rather, how to use it for sp ecific objectives. The consensus among media researchers is that animation may or may not promote learning, depending on how it [is] used. For these reasons the search for media effects has been calle d off. In its place is a search for the conditions under which various media, such as animation, affect the learning process. (Mayer & Moreno, 2002, p. 88)Educational researchers concerned with visual display "agree that effective learning with visuo-spatial text adjuncts is not dependent on th e professional appearance of visuals, but on the relation between these displays and the task demands and on learner's prior knowledge and cognitive abilities" (Schnotz, 2002). Second language learning involv es various skills, and there are a variety of ways to achieve progress for a variety of learners. This research sought to determine whether animation had a measurable benefit for the specific tasks involved in German grammar transformations.The findings should be most useful for designers of multimedia for German grammar instruction, but they may also be applied to the use of animation for multimedia grammar instruction for other foreign languages, especi ally when the point of grammar involves a consistent change of position or form. The ke y inquiry is: What happens when you put sample text in motion in order to depict grammatical dynamics?Organization of the StudyThis chapter introduces the i ssues underlying this researc h. The purpose of the study as well as the specifics of the research questions have been delineated. There follows a glossary of definitions of terms that are used in the literat ure review of chapter tw o. Chapter two presents a review of research litera ture relevant to this study. Three main areas are covered in this review: Theoretical background related to second la nguage grammar instruction and technologyenhanced instruction, research into the dual coding theory of mental representation and
10 educational research into the us e of animation in instruction. Ch apter three describes the design of this research study and provides information about the participants, measures, instruments, and the procedures for data collection and an alysis. Chapter four pr ovides the results and analysis of the outcomes, and chapter five is a summary and discussion of the results. Chapter five also contains the implications of the study and recommendations for further research. Glossary of Terms Animation: The technique by means of which movement is given to a drawing or series of drawings. (A cartoon is a humorous or satirical drawing. Animated cartoons, such as "The Simpsons" are often called simply "cartoons" in English, with the concept of animation inherent in the term and omitted as an adjective.) Associative structure: A within-system network of links and references. The functional connections of logogens to each other (in the verb al subsystem) or of imagens to each other (in the nonverbal subsystem). (see Dual coding) Auditory input: Objects, events or ideas that are expe rienced through the aural sensory organ, the ears. Narration, music, lyrics and sound e ffects are all sources of auditory input. CBI: Computer-based instruction. Distribution of instructional material using microprocessorbased resources. CALL: Computer assisted language learning. Any aspect of langu age acquisition (see SLA) that takes place in a computer environment (PC, Internet, computer lab network, etc.). Conjoint retention theory: Conjoint retention theory expl ains how referential connections between the verbal and non-verbal system not only maintain memory of information but also operate for both recall and reconstruction efforts. Dual channeling: The presentation of information using more than one modality, such as audio plus visual. Dual coding: Dual coding theory posits two functionally separate, re presentational systems, the verbal and the nonverbal, with dynamic mental im agery residing solely in the nonverbal system and linguistic representation occu rring in the verbal system. Imagens: Representations from which mental im ages are generated. Imagens correspond to natural objects, holistic parts of objects, and na tural groupings of objects. Imagery: Mental images of consciously experien ced internal objects, events or ideas.
11 Latency: The time that elapses between a stimulus and the response to it [syn: reaction time]. Logogens : A hypothetical verbal represen tation that registers percep tual word information and makes a word response available when enough relevant information has accumulated. Matching pacing: Forcing a learners reading or viewi ng pace to synchronize with narrations or to control the speed of input processing. Mental imaging: A dynamic process of generating a mental image from perceptual stimuli, such as a verbal description or an experience. Mental model: Cognitive representation of the mental world. Mental rehearsal: Verbal processes experienced as me ntal words or inner speech (silently "talking" to oneself). Mental rotation: Mental manipulation of referent ob jects for spatial transformations and observations of perceptual changes including ment al rotations on any plane, changes in imaged size, distortions of shape, and changes in the relative position of two or more objects. Mental translation: Mental manipulation of a single aspect of a referent object keeping all other aspects constant. Mental world: Cognitive representation of the real world. Modality: A particular sense or avenue of input (sight, sound, smell, feel and taste) [syn: channel]. Negotiation of Meaning: The give and take that speakers enga ge in as they exchange ideas and information. Nonverbal symbolic subsystem: One of dual coding theorys two subsystems for representing knowledge and experience. The nonve rbal system consists of im agery of objects and events. Environmental sounds, smells, tastes and the f eel of objects and events stimulate nonverbal memories that are symbolically represented in mental imagery. Referential availability: The probability that cues to one subsystem will be able to make associations to memories in the other subsyste m, or that logogens will activate imagens and vise versa. The capacity for stimuli to simultaneous ly activate memories in both subsystems is a function of the availability of referential connections.
12 Referential connection: A between-system structural conne ction or access route that permits activation of memory in one system to trigger activity in the other representational system. An example of a simple referential connection is the tie between the image of an object and its name. SLA: Second language acquisition. This concept in corporates foreign la nguage learning, recall, retention, and systematic and pragmatic perf ormance skills (listening, speaking, reading and writing). Semantic matching : When a picture has elements that sh are meaning with textual elements, the elements may adhere to each other in memor y, maintaining a meaningful connection between them. Sensorimotor: Combining the functions of the sensory and motor activities; pertaining to the physical properties of sensory intake system s: vision, hearing, taste, smell and feel. Syntax: A fundamental principle for encoding and decoding meaning through the placement and relationship of sentence elements. Verbal symbolic subsystem: The other of dual coding theory s two representational systems. Words, both visual (text) and auditory (lyr ic and narration) are represented by symbolic associations (phonemes, letters, words, sentences writing patterns, etc.) that make up the system of verbal representation of the world. Visual input: Objects, events or ideas that are expe rience through the visual sensory organ, the eyes. Text, pictures, colors, graphs, and their an imated transformations are all sources of visual input.
13Chapter 2Literature ReviewIntroductionThis chapter presents the rese arch literature that is rele vant to the questions under study. The overarching field of the present study is Ge rman language acquisition as it is facilitated through computer-enhanced classroom instructi on for principles of grammar. This review discusses views of grammar and its role in fore ign language teaching approaches. Some issues of foreign language teaching are framed within th e environment of computer assisted language learning (CALL). CALL has been informed by re search in the field of computer based instruction (CBI), which, in turn, is framed here within the th eory of mental representation called dual coding. The particular feature of CBI that is examined in the pr esent study is animation and its effects on learning outcomes. This chapter, therefore, also discusse s research relevant to the use of animation in educational multimedia. Theoretical Background and Pertinent Research Findings This study began primarily as a question of how to teach German grammar in a computer-based environment. This review begi ns with a discussion of the changing views of grammar and of different approaches to teaching it. Antecedent to the question of how to teach grammar is the question of whethe r or not it should be taught. The Role of Grammar in Foreign Language Teaching The issue of how to teach grammatical rule s and principles within the formal class curriculum has generated quite a lo t of interest in the field of foreign language education. Most researchers agree that the question is not whet her grammar has a role in language learning, but rather whether the function of grammar is a pe ripheral component of foreign language teaching, or the primary core of the in structional curriculum. As a peripheral component, grammar forms an unanalyzed structure that the learner eventually c onforms to by means of "feel", much like a
14 native speaker. For first language use, most peopl e never think of the term "grammar", yet they make grammaticality judgements every time they communicate. Arguments against grammar-centered language instruction rely heavily on evidence from first language acquisition research, reasoning that effective L2 learning should emulate L1 acquisition as far as possible. Other arguments counter that la nguage acquisition, whether L1 or L2, with no focus on form, may likely reach only si mplified registers, that are judged (at least socially) inferior, along with the speaker (Hall, 1998; Lally, 1998). Premises of Grammar and Approaches to Teaching It Approaches to the teaching of grammar have changed in indirect response to changing notions of what grammar is and how it is define d. Fodor, Bever, and Garrett (1974) have defined grammar as "any system which is able to specify the set of structural descriptions characteristic of the sentences of a language" (p.21). Throughout the decades of 1930-1950, the term "structuralism" was employed, pertaining to a ta xonomic description of language features. The approach to teaching grammar during this time was typically called "grammar translation". It was a top-down approach that began with a body of existing language and analyzed relationships among its constituent parts. Following the seminal work of Noam Chom sky (1957), views of grammar evolved to encompass the term "generative". Generative gr ammar "allows the notion of transformational derivation" (Fodor, Bever, & Garrett, 1974, p. 103) which can represent language as a sequence of transformations, from which complex sentence s are derived from simpler ones. This concept inspired different approaches to teaching grammar. As a view of grammar as a useful to ol for generating future language use, transformational-generative grammar (TGG) is compatible with propositions in favor of instructed grammar in foreign language teaching. Taxonomic grammar is based on corpus statistics, where syntactic rules are derived from the evidence provided by analysis of existing utterances. Bound to corpus descriptions, stru cturalism provides no explanation of how people form sentences that they have never used before (Chomsky, 1957). TGG is not constrained to existing evidence. Instead, it is a view of language as boundless in the possibilities of utterances. Under TGG, the structure of language can be approa ched as a set of interchangeable parts, which join together in consistent, reliable patterns, a nd these patterns are studie d more closely than the
15 parts themselves. Chomsky's view of language as a vast set of building bl ocks with a limited set of rules for putting them together into an infin ite number of new utterances, allows an approach to language learning, which is consistent with the bottom-up approach to grammar instruction, termed "structuralist ap proach" (Ellis, 1994). Structural vs. natural approach. The word "structuralist" is confusing here because the approach to foreign language curriculum does not rest on the structuralist view of grammar. Structuralism as a view of grammar leads to the grammar translation method of foreign language teaching in which a given corpus of target language is analyzed and translated into the learner's native language. The structural approach begins with the rules and features of grammar, and treats them as building blocks fo r creating learner-invented utterances. To the linguistic adult, grammar has power as a "device" (Chomsky, 1957) for generating novel sentences. The "natural" acquisition of a gram mar device takes time and massive exemplary input, while overt explanations of grammatical principles provide at least a framework for immediate use of that device. Ther efore, as an approach to curricu lum, the structural approach is supported by the transformational-gene rative approach to grammar. The structural approach to German L2 curriculum for college students embraces conceptual learning, especially for syntactical prin ciples. Structuralists r ecognize two facts which influence this choice. The first is the time cons traint; usually one academic year or roughly nine months. This is the time allotted to pr ovide enough instruction for a solid foundation of continued language learning; yet few students take courses beyond th e introductory le vel, so this single school-year is the boundary of forei gn language education for most Americans participating in post-secondary education. "In one year of colle ge German, an American student cannot afford to learn the language by trial and e rror and from context only; he has to resort to grammar if he wants to master German quickly" (Lohnes & Strothmann, 1973, p. xv). The second fact that influences the structuralist approach is the r ecognition that college-age students are "linguistic adults" whose fi rst language intervenes, to a cer tain extent, with the natural acquisition of second language structures and prin ciples. The structuralist argues that "once we have absorbed one linguistic code, we cannot learn a totally different code without some contrastive analytical knowledge of that new code (p. xv). It is not that an adult's native language presents a barrier to second language learning, but that their prior knowledge of
16language, in general, may present a need to reco ncile new concepts about language with already known facts. In the approach taken here, grammar is taught in order to facilitate language learning for cognitive adults. Facilitate means "to ease and speed the process of" and "cognitive adult" refers to someone who is thinking about language lear ning, exerting particular efforts, and making conscious decisions to acquire ne w linguistic abilities. This is an approach deemed superior by structuralists, who also recognize the time constrai nts inherent in American college-level foreign language teaching. The structuralist approach to German L2 instruction has contributed greatly to the development of curricular sequencing in German textbooks. That is, German textbooks generally sequence the grammatical structures in a like manner according to a logical hierarchy, according to the difficulty of structures and of their frequency in common usage. For example, present tense verbs are taught befo re past tense forms, past tense forms before reflexive verbs, reflexive before passive voice, etc. Each suc cessive structure adds a new element (a form change, an auxiliary verb, a reflexive pronoun, etc.). Pragmatic concerns also determine subordinate schemes in the structural approac h. For example, the conversational past is taught before the simple past because of its utilizati on in conversation, but the simple past forms of sein (to be), werden (to become), and haben (to have) are taught at the ea rliest possible time for their utility and frequency of use.The structural approach to German L2 has been dominant throughout the last three decades in the U.S. (Lally, 1998a), although there ha ve been efforts to move in a new direction. "Grammar and the formal learning of grammatical structures is very much considered a thing of the past" (Pienemann, 1995, p. 8). More recently, textbooks are edging toward the more popular "natural approach" in which co mmunicative context precedes gram matical curriculum. A natural approach textbook, such as Kontakte (Terrell, Tchirner, & Nikolai 2000) offers units that are based on communicative situations, rather than disc rete points of grammar. Instead of a chapter about reflexive verbs, there is a chapter about health. The precedence of s ituational context over grammatical sequencing is act ually just an illusion. The fact is that successful1 natural approach 1 Kontakte is deemed "successful" here, based on the fact that it is one of the most widely adopted textbook in college-level beginning German courses, according to an Internet surveyconducted by the researcher during the year s 1999-2002 (unpublished). See also the American Association of Teacher of German (AATG) electronic bulletin boa rd thread "Kontakte".
17textbooks have maintained nearly the same gr ammatical sequencing beneath their practical contexts. The chapter on health for instance, may concern sports and medici ne, injury and illness. In order to speak of these things, it is necessary to use reflexive verbs. Injuries and ailments occur after other events, so th e prior chapter may have been about professions, hobbies and activities using the past tense forms as the structural component. The concentration seems to be about communicative situations, but the underlying structur e is still a matter of grammar. In the introduction to the textbook, Terr ell and his colleagues (2000b) e xplain their approach as an outcropping of Krashen's five interrelated hypo theses, and they acknowledge his distinction between "acquired" and "learned" knowledge. Acqui red knowledge is utili zed unconsciously and automatically, and learned knowledge is "knowle dge about language and the way it functions" (p.21). Learned knowledge is meta linguistic thought, while acqui red knowledge is instinctive activity. Krashen's hypotheses have been of monumental importance to the field of SLA, and they have had profound influence on the curriculum of foreign language teaching, but they have not been accepted without contest. The main issue has been the dichotomy of the distinction, acquired versus learned. Krashen suggests that what is learne d does not become acquired, a concept difficult to prove and even more difficult to disprove. The distinction is very useful to FL educators no matter how contestable. Acquisiti on of L2 ability is the aim of foreign language education, whether it is achieved as result of learning or in a ddition to learning. Either way, learning has not been deemed detr imental to eventual acquisition. Terrell and his co-authors ( 2000b) base their approach on Krashen's hypotheses, defining acquisition as "the mental linking of meaning to form [which takes] place during meaningful communication" (p. 18) Neverthele ss, they also believe that th e study of gramma r "will facilitate [the] processing [of] input, and make it easier to po sit the meaning-form connections that are the basis of acquisition" (p.18). They use a grammatical syllabus en compassed within a semantically based approach, which they call "topical-situati onal". This approach to grammar is consistent with form-focused instruction (FFI).
18Form-focused instruction FFI is another name for instructed grammar, but it is not to be confused with the concept of grammar translation. The goal of FFI is grammatical accuracy in communication, not expertise in gr ammar rules. It is a concept that acknowledges the durability of naturally acquired language abil ity, but it also declares cons iderable gains in the amount of acquisition that takes place within the (form-focused) classroom (Lyster, 1996).Analysing a broad range of SLA research concerning FFI, Housen (2001) has drawn the following conclusions:1. FFI has a positive and durable effect on foreign language learning when compared to exclusive communicative exposure to the language.2. Effects are: faster learning, higher leve ls of proficiency, more accurate language use (but not more fluent language use).3. Effects are mainly felt in language task s that allow for planning and editing (formal language use, written language use), less in unplanned, oral language tasks.4. FFI mainly affects the development of explicit knowledge (with possible indirect effect on implicit knowledge).5. Effects are strongest with mo re explicit types of FFI.(Slide 38). Implicit vs. explicit grammar instruction. FFI has its own dichotomy concerning the way grammar is instructed. In explicit instructi on, the rules governing gram matical structures are presented as formal statements, whereas implic it instruction "exposes the language learner to grammatical and lexical principles through na tural language experien ce" (Lally, 1998b, p. 154). The central tensions of the implicit-versus-explicit controvers y are whether to teach grammar and how it should be taught. The controversy has been between extremes, and according to Lally, has formed an unproductive binary cons truction. Lally agrees with those (Ellis, 1990; Garrett, 1986; Lyster, 1996) who feel there is a need for a "re-thinking of grammar within language teaching," but she states rather clearly that no answers will be found at either extreme position (Lally, 1998b, p.158). Only a hybrid soluti on, such as those suggested by Schmidt (consciousness-raising, 1990), Adair-Hauck, Donato, and Cumo (1994, whole language), LeBlanc and Lally (1998, student-mediated ex plicit instruction), or Van Patten (1992,
19processing-instruction) will move the professi on of foreign language teaching to more productive conversations and ultimately benefit students' language acquisition. All of these researchers have proposed and investigated approaches that merg e form-focused instruction with a communicative context. For Lally, it is not so much that the hybr id methods will solve the issue, but rather that their middle-ground position is the necessary starting point for productive dialog.Sharing this view, Hall (1998) sums up the tw o positions and then agrees with both of them. "Students who have been through comm unicative teaching at schools do not acquire a mastery of German grammar, which is essential for degree level studies. [However,] the object of communicative language teaching in schools is that pupils learn to co mmunicate in the foreign language; the one-sided insistence on grammatical accuracy belongs to an outmoded form of teaching" (p.42). Lally and Hall both see the mi ddle as fertile ground to find a solution. Hall sees that middle ground in the realm of technology. He takes advantage of the many facets of CALL to combine the explicit and implicit approach es with exploratory grammar learning for German L2 in a British university. Exploratory teaching is described as "planned inductive learning". Hall does not advocate anything resembling a return to a past time. In fact, no researchers, concerned with German pedagogy at the secondary level, describe a methodology of the past that is worth returning to, but neither are there many who seem satisfied with learners' progress under current methods. Hall writes of technology as a tool for teaching grammar, describing such applications as commercial software progr ams, grammar checkers, online grammar books, interactive webexercises, gl ossed texts and tandem learning (email and chatrooms). While computer programs and Internet websites pr ovide extraordinary oppor tunities for learning German grammar, Hall remarks that such refe rences may be "useless to learners who do not understand the grammatical terminology used in th em" (p. 45), but he sees advantages in the future of CALL to help learners overcome their grammar deficits. Computer-enhanced Grammar Instruction The computer has been shown to be an invalu able tool to help re searchers discover more about the way students approach language learning, in terms of the strategies they use most often (Hulstijn, 1993; Nagata, 1997) and of those proven to be most effective (Hill & Hannafin,1997; Hulstijn, 2000). There has also been research involving CALL aimed at determining learners' awareness levels, of the computer or web en vironment itself (Hill & Hannafin, 1997) and their
20 metalinguistic awareness of lear ning strategies (Purpura, 1997). In addition, researchers Davis and Lyman-Hagar (1997) have used CALL to inves tigate ways of training learners in the use of those strategies determined to be most effective. The computer is a unique tool for investig ating various teaching conditions. There are four main instructional conditionsincidental, implicit, enhanced and explicitwhich have often been discussed in SLA/ FLE literature (Hall 1998; Mi tchell & Redmond, 1993; Nagata, 1995, 1997; Robinson, 1997; ). CALL researchers ha ve not only been able to produce true versions of these conditions, but they have also been able to measure the effectiveness for various types of tasks and stud ent learning styles (Viteli, 1989). The incidental condition, for example, assumes that the learner can pick up the salient aspects of the material incidentally, that is, without any special focus on those features. The implied condition presents the specific features to be learned but offe rs no explanation, assuming the learners will form their own rules and integrate them naturally into their inter language development. Interlanguage (Selinker, 1972) is the internal language that learners possess as they devel op their foreign language skills. It includes all of the hypotheses of rules that learners typically hold in the course of their language acquisition. (Interlanguage therefore is never fixed, but ev olves along with the learner's progress.) It has been described as a "system of implicit L2 knowledge th at the learner develops and systematically amends over time" (Ellis, 1994). The implicit condition of instruction, therefore, aims to influence this system and to bring about accurate amendments. The enhanced condition is similar to the imp lied in that there is no explanation, but the salient features are punctuated and stressed by various means such as color coding, italics, differing fonts or sizes, animation, etc. In th e explicit conditi on, the pertinent information is simply stated in the instructi onal portion of the pr esentation. Most studies of CALL and teaching conditions have favored an explicit form-focused instruction coupled with training in the use of metalinguistic strategies for metalinguistic aw areness (Nagata, 1997; Purpura, 1997; Robinson, 1997; Spada & Lightbown, 1990; Yang, 1995). Recently, Chapelle (2002) has ma de the distinction between CALL and other uses of the computer for FLE, namely testing and SLA res earch. She uses the term CASLA, spelled out as computer applications in second language acquisition to broadly encompass the three often
21blended areas: CALL, computer-assisted langua ge testing (CALT), and computer-assisted second language acquisition research (CASLR).Computer-based instruction. To investigate the best practices for CALL, research must look backwards as well as forwards. A look over the shoulder of CALL research will see the broader field of CBI. Such research covers a vast diversity of discipline s and learning objectives, but, the type of CBI research most relevant to the present study is that concerning the use of multimedia for providing dual channeled input. The re search in this arena investigates aspects of multimedia for their effects on cognitive proces ses. CBI is designed according to theoretical principles, but it has also been designed in orde r to test those principl es. This review covers research studies that have examined the effect of multimedia in CBI as it applies to the theoretical princi ple of dual coding. Dual Coding Theory: Mental Representations Dual coding theory is perhaps the most influential ph ilosophical foundation for this study and for the pedagogical justification for exploiting the affordances of multimedia. Formulated by cognitive psychologist, A. Paivio, dual coding theory postulates a division between two dominant cognitive processing systems, the ve rbal and the nonverbal. Dual coding theory addresses mental representations and mental imaging. a dynamic process that includes motor as well as sensory components (Paivio, 1971, 1990) Researchers supporti ng the theory claim that information is easier to retain in memo ry when it is dual coded (ChanLin & Chan, 1996; Mayer, 1992, 1997). Paivio (1990) divides mental representations into two classes, the concrete picture-like and the abstract language-like. Photographs, draw ings, maps and diagrams offer more concrete representation of knowledge elements than linguist ic descriptions of those elements. A threedimensional colored motion picture with a sound track could be indistinguishable perceptually from real world events[whereas verbal descri ptions are] functionally abstract and rather arbitrary in their relation to th e represented world (1990, p. 17). The most general assumption in dual coding theory is that ther e are two classes of phenomena handled cognitively by separate subsys tems, one specialized for the representation and processing of information concerning nonverbal objects and events, the other specialized for dealing with language (Paivio, 1990, p. 53).
22Paivio used a graphic representation to illu strate the elements of dual coding theory in both verbal (words) and nonverbal (p ictorial) form. In figure 1, Paiv io refers to the the nonverbal (symbolic) subsystem as the imagery system. He further defines imagery as a work space in which cognitive processes can operate. It is part of the conscious memory process, which others refer to as working memory (1990, p.74). The verbal (symbolic) subsystem also engages working memory, but in this case, the conscious is occupied with words and other linguistic elements in a sort of mental rehe arsal of speech or writing patterns.Paivio often uses the term symbolic to name both subsystems because the elements of the minds mental model are symbols that refe r to and then invoke the imagery of knowledge and/or experience. The two symbolic systems, verbal and nonverbal, each receive input through the visual as well as the auditory sensorimot or modalities, but the input has different values within each subsystem. Visually, the verbal subs ystem is activated by visu al words (text, caption, annotation, etc.) and auditorally, by words as well (narration, announcement, aside, etc.) Whether auditory or visual, words are consider ed verbal stimuli. The nonverbal subsystem is also activated in the visual modality by visual obj ects (pictures, colors, animations, etc.) and in the auditory modality by environmental sounds (music, sound effects, etc.). The nonverbal symbolic system also has the added modalities of taste, smell and feel to activate memories and VERBAL STIMULINONVERBAL STIMULI SENSORY SYSTEMS VERBAL RESPONSES NONVERBAL RESPONSES Logogens ImagensAssociative structureAssociative structureREPRESENTATIONAL CONNECTIONS REFERENTIAL CONNECTIONSFigure 1. Mental Representations: A Dual Coding Approach (adapted from Paivio. 1990, p.67). Imagery System
23 stimulate recall. According to Paivio, visual objects (as opposed to text) and environmental sounds are nonverbal stimuli, or imagens which lay memories (nonverbal knowledge units) in the associative structure of the nonverbal subsystem. Logogens and imagens. Imagens refer to representations from which mental images are generated. Imagens correspond to na tural objects, holistic parts of objects, and natural groupings of objects. Logogens a term described by Morton (1979), are verbal representations while imagens represent knowledge in nonverbal form. Both can also be viewed as cues, because they are knowledge representations, whic h trigger the instan ce of other knowledge representations. Words activate logogens, which re gister perceptual word information and make a word response available which, accordin g to dual coding theory, accounts for word recognition (Paivio, 1990, p. 76). [O]bjects or their pictures activate imagens, motor movements activate motor patterns, and so on (Paivio, 1990). All stimuli are either logogens or imagens. Logogens operate in the verbal symbolic subsystem of mental representati on. Any stimulus that triggers a memory in the verbal system is a logogen and any verbally sy mbolic memory activated by that logogen is a logogen, too, for it will, in tur n, activate other memories. The memory activated by logogens is also referred to by Pavio as a memory trace, de fined as any kind of psychological record or representation of past episod ic experience (Paivio, 1990, p. 24). Logogens that activate other logogens do so with in the associative structure of the verbal system. Logogens also activate imagens. Word s evoke imagery and experiential memories as well. A logogen activates an imagen in th e nonverbal system by crossing a referential connection between the two systems. When ne w knowledge is learned, or encoded, using both representational systems, then referential connections between those systems are formed. Paivio calls these connections referen tial availability which capita lizes on the powerful mnemonic properties of nonverbal stimuli and imagery (Paivio, 1990, p. 253). The language system is peculiar in that it deals directly with li nguistic input and output (in the form of speech or writing) while at the same time serving a symbolic function with respect to nonverbal objects, events, and behavi ors (Paivio, 1990, p. 53). Since language is used to describe objects, events, feelings, and other nonverbal phenomena, logogens are based upon and initially bound to imagens th rough referential connections.
24 A primary precept of dual coding theory is that the two systems operate separately. Although logogens are initially based on imagens wo rds refer to things, the verbal system can and does eventually function inde pendently. That means that l ogogens often activate only more logogens and that language activi ties can take place entirely with in the associa tive structure of the verbal mental world. This independence precept assumes that, in an advanced verbal symbolic system, referential connections to nonverbal representation are not automatic. This powerful referential avai lability is only present when nonverbal stimuli are connected with verbal cues. To formulate interconnections between subs ystems, the nonverbal system would have to be stimulated in order to produce additional me ntal imagery memories within the associative structure of the nonverbal system. This should be done in tandem with input to the verbal system. The aim with respect to dual coding theory is to meld experiential (nonverbal) memories with the verbal knowledge units at the tim e of instruction, in order to lay a network of referential connections between subsystems from wh ich to draw upon during recall efforts. Visual enhancement in the form of colors, f onts and animation can be used to associate words with imagens or to give experiential me mory to text based learning. Animating written text can give a nonverbal referent to the verbal knowledge unit, th us strengthening the referential connections between the verb al and nonverbal subsystems. One aspect of this study that sets it apart from previous animation research is that there are no graphics included in the instructiona l presentation, only the textual information is animated in the experimental treatment. The essent ial aim of animating the text is to change the verbal information into illustration. Animating the text turns it into a picture, which creates a referential connection between verbal memory and non-verbal imagery. If significant benefit is achieved through the use of animation, it may be attributed to the creation of a dually-coded memory trace and a broader web of recall connections. The shape of the letter E. To illustrate his analysis of the human representational system, Paivio asks his readers to think of the letter E. Both auditory mention and textual representation of the letter E wi ll likely evoke a mental connec tion to the associative structure within the verbal symbolic system. Recall of the letter E may activate memory nodes along the hierarchical lines of the verbal structure as the letter E stands ready for associative building, as
25 a phoneme, an orthographic agent, and in some cas es (e.g., Italian), as a word. But if commanded to reverse the letter mentally to its mirror-image position, rotate it 90 and then number the corners in counterclockwise or der, the recall request would st imulate a nonverbal response and the E would become a referent object to be tr ansformed within the mental imagery mechanism. In order to maintain a mental image and to tr ack the perceptual changes of the object, now , associative nodes and pathways in the nonverbal system would be activated. This object could lose all referential connect ions to the verbal system and become a shape rather than a cipher, with new symbolic associations. Now imagine the letter E embedded within a word, say, men. Associations activated within the verbal symbolic sy stem form a mental representa tion of the letter constrained by linguistic knowledge structures, wh ile associations within the nonverbal system form a mental image that represents a pictorial concept of men as a concrete object present in the real world. The mental image of the word men has refere ntial connections that cross back and forth between the two symbolic subsystems. Men is a word and a thing. The mental imagery mechanisms stand ready to manipulate the word men linguistically and to manipulate the conceptual thing men within the nonverbal mental world. Unde r command to mentally rotate the word 180 the E in its new position may spur new referents associated with the shape , but it would also be likely to re tain associative connections to verbal referents. In fact, the nonverbal imagery mechanism that helps to manipulate the object MEN into the object NW may form a new referential connection in to the verbal struct ure, which stimulates activation of the logogen, new. In other words, the mental manipulation of wo rds as objects may help to form referential connections that cross back and forth between the two subsystems, forming a sturdy integrative knowledge web. Words, mentally manipulated as nonverbal images, still retain their semantic connections to the verbal symbolic system wh ile their manipulation, movement, transformation or other nonverbal process forms an experien tial memory trace accessible to the nonverbal system. In the above example, mental transformati on of the logogen/imagen men into new would also be likely to activate new associat ive connections (logogens or imagens) in both subsystems. The wordplay could evoke new logoge ns of orthographic representation even while
26fresh associative connections are stimulated by th e juxtaposition of the social concepts new and men.While most associative connections, that is, paths of knowledge structures, are constrained by fixed dictum, such as laws of physics and grammatical ru les, which govern those knowledge domains, the nonverbal representativ e systems are rule-f ree and unconstrained. Die Gedanken sind frei (thoughts are free). Mental imagery is unbounded and need not be otherwise. Research, however, has begun to show that ment al imagery typically conforms to particular patterns governed by knowledge domains (Shepard, 1978). The way we think about a particular phenomenon, the imagery which represents it, may be pivotal to our eventual knowledge of it. This does not mean that in order to really know a thing or to have expertise in a knowledge domain one must share the common imag ery processes conformant to all experts in the domain. One may need to speak this way to be heard and to w rite this way to be understood, but it is not necessary to imagine this way to unders tand or to think this way to know. It does suggest, however, that the manipulat ion of a novice learners mental imagery is not automatically a constricting dictate to conform ment ally, but an effort to stimulate the referential connections that typically suppo rt specific knowledge structures. Dual Coding Theory and SLA Paivios (1971) dual coding theory has inspir ed several empirical st udies that endeavored to test his hypotheses that the memory trace is modality-specific and that referential encoding results in a dual or multimodal trace (p.xx). Thr ough review of dual coding research conducted between 1971 and 1989, Paivio has found support fo r these hypotheses but he has amended his theory to account for an imbalance in the mnemonic value of the two encoding systems, perhaps by a 2:1 ratio favoring the image c ode (Paivio, 1990, p. 77). Many researchers agree that a picture is worth 10,000 words (Mayer & Gallini, 1990; Mayer & Sims, 1994; Rieber, 1997). Paivio concludes further that activation of both encoding systems can have additive effects on recall (Paivio, 1990, p.77). Referentia l connections between the two subsystems strengthen the chances of memory retrieval. In a discussion concerning first language acq uisition, Paivio postula tes that the nonverbal representational system precedes development of the verbal system and that initial verbal representations refer to primarily concrete objects al ready represented in the nonverbal symbolic
27 system. This means that the verbal system is dependent upon and overlaid atop the nonverbal memory structure. An important postulate of dual coding theory, however, is that the two systems function independently. The verbal system begins as a series of referent connections leading to nonverbal representatio ns, but it can eventually f unction as a separate system. In the field of second language learning, Paivio himself, has tested his theories relative to French vocabulary learning (Pai vio & Desrocher, 1979). His conc lusion that the associative learning experience requires use of imagery as well as intraverba l connections has been reiterated by subsequent re search in the SLA field. Yoshii and Flaitz (2002) found that word annotat ions utilizing both text and pictures were more effective for helping adult learners retain vocabulary than either te xt-only or picture-only annotations. Regardless of prof iciency level, the dual sour ce annotations, available in a computer-based hypertext, helped ESL learners score higher on the three tasks of picture recognition, word recognition, and definition-supply tests, than learners who accessed only one mode, picture or text, as an annotation aid. Ehlers-Zavala (1999) used a dual coding approach to study foreign language reading by an English as a foreign language (EFL) high sc hool population. Given either a text-only or a text-plus-illustrations condition, 129 EFL learners then answ ered a comprehension questionnaire, provided free-reports and rated st ory segments for the strengths of imagery and affect evoked by the story during reading. Ehlers-Z avala considered the research results to support dual coding because participants who read the story with pi ctures performed better than those who did not in that they reported a higher number of images in their free-reports and more vivid images in their story ratings (1999). This dissertation studys evidence that the text+picture condition offers superior results when compared with a text-only condition s upports dual coding theory, and in this case, the study refers to the field of second language learning. Studying French-speaking studen ts recall (measured by transl ation tests) of Arabic L2 vocabulary items, Hammouds (1982) disserta tion uses dual coding ideas to develop an associative-field technique in wh ich students were exposed to lexical items in imagery as well as in verbal, associative contexts. This research al so suggests that associative imagery facilitates vocabulary learning (as observed by translation recall tasks).
28 Dual Coding and Total Physical Response. While Paivio and his associates (Clark, Pressley and Desrochers) have studied dual coding theory in second language (L2) learning only for the goal of vocabulary learni ng, he postulates an important ap plication of dual coding theory for L2 learning that applies to the nonverbal-situational cognitive and behavioral contexts of all language skills including syntax (Paivio, 1990, pp. 254-6). Paivio sees Ashers (1969) Total Physical Response strategy (TPR) as an exemplar y application of dual coding theory. In having learners act out responses to suggestions and commands, the L2 is studied in the context of nonverbal behaviors (Asher, 1969, 1972; Paivio, 1990 ). TPR elicits a motor reaction which is a concrete response to commands such as wal k, run, jump, but assimilation into the associative structure of the verbal system requires an abstra ct response to the mental imagery of those verbs. The verb run can evoke the im age of a boy running, as opposed to a horse or an engine, a stocking, a campaign, a nose, etc. Dual coding theory leads to a strong emphasis on the role of situational contexts and imagery in second language learning. In particular, the theory suggests that language-learning st rategies based on the systematic use of referent objects, pictures, activities, and mental imagery would be especially effective in promoting learning (Paivio, 1990, p. 257). Paivio also considered the implications of dual coding theory fo r acquisition of second language structural concepts. He felt that gr ammar learning should be facilitated by the use of appropriate nonverbal referent s ituations, pictures, or imagery and that learning strategies should be based on the systema tic use of referent objects, pictures, activities and mental imagery (1990, p. 257). [T]he development of grammatical skills involves formation of referential interconnections, not only between repres entations corresponding to objects and their names, but also between the abstr act and dynamic attributes of objects (e.g., relations, transformations, actions) and their corresponding verbal descriptions (Paivio, 1990, p. 256). Although Paivio has not, as yet, considered any aspects of technology whatsoever, the medium of instructional presentation under study here, multimedia, has properties that connect to every aspect underscored in his hypothesis: refere nt objects, pictures, activities, and mental imagery. Referent objects are the fundamental units of instructional design architecture. To
29 program events in a computer ap plication, all elements, including text in various parcels, are treated as objects. As objects, hunks of text as well as piece s of pictures, can be made to move, change or be manipulated so that the le arner may interact with them and cause them to interact with each other. The users activities, in this regard, prompt a sensorimotor experience that should help to form nonverbal associations Multimedia affords learner-activated as well as automated animation of objects (hunks and pieces of text and picture). The entire multimedia module, integrating actions and objects helps to activate mental imagery. Animation Research Paivio has yet to consider dual coding theory in the light of instructional technology; however, many supporters of his theory have test ed the tenets in a CBI environment. Research into the cognitive effects of animation is ty pically based on Paivios theoretical precepts (ChanLin & Chan, 1996; Lai, 2000; Mayer, 1994, 1997; Mayer & Anderson, 1991, 1992; Mayer & Gallini, 1990; Mayer & Moreno, 1998; Mayer & Sims, 1994; Rieber, 1988,1989b, 1989c, 1990, 1991, 1995, 1996, 1996b; Rieber & Hannafin, 1988; Rieber & Parmley, 1992; Schnotz & Grzondziel, 1996). If knowledge has to be acquired about a dynamic subject matter, computer-based learning provides the possibil ity not only to use static, but also animated pictures. In the case of static pictures, the learne r has to imagine the dynamic of the subject matter that is, he/she has to stimulat e the process internally and observe the stimulation with the mental eye. Animat ed pictures, on the contrary, can display the dynamics of a subject matter dir ectly. (Schnotz & Grzondziel, 1996, p. 3) Studies on the use of animation in CBI have been meager to date, and their results are spread thin across diverse disciplines and di sparate designs. However, the one thing that animation studies have in common is that they all show some kind of quantitative differences due to the use of animation in instruction. Some have asserted significant benefits affecting attitudes, enjoyment, motivation and frustr ation levels (ChanLin & Chan, 1996; Hsieh, 1992; Poohkay & Szabo, 1995; Rieber, 1988, 1989b, 1989c, 1996; Szabo & Poohkay, 1996; Williamson, 1992; Wong, 1994), while others have tried to measure animations effect upon immediate recall and problem-solving skil ls (ChanLin & Chan, 1996; Caraballo, 1985; Childress, 1995; Hays, 1996; Lee, 1997; Lee, 1996; Mayer, 1992; Poohkay & Szabo, 1995; Towers, 1994). Delayed recall, or retention, has b een the dependent variable of focus for only a few researchers (Mayer, 1992; Hays, 1996; Cara ballo, 1985) and their conclusions are mixed,
30 with all results dependent upon additional, divers e variables such as the absence or presence of explanatory text, of synchronous or successive narration, or of spa tial ability and prior knowledge. With the exception of Schnotz and Grz ondziel (1996), Rieber has been the only researcher to examine the effect of animation upon latency or response times. Rieber also measured the effect of animati on across several posttest learning objectives in a mixed ANOVA design. Mayton (1991) followed Riebers design principle with his study testing the effect of presentation strategy (static graphic, imagery cued and animated) on five different learning objectives concerning the structure and function of the human heart. The most extensive research concerning th e pedagogic effect of animation has been undertaken by Rieber (1988, 1989a, 1989b, 1989c, 1989d, 1990a, 1990b, 1991, 1995, 1996a, 1996b). Within the discipline of physics, Rieber has studied the effect of animation on students understanding of such concepts as Newtonian laws of motion and properties of velocity. Performing nearly identical experiments with 119 4th and 5th grade children and 141 college seniors in a computer education class, Rieb er attempted to tease out the developmental differences that might interact with the eff ect of animation on learning. In every experiment concerning animation, however, he al so tested an additional variable that he found crucial to the learning outcome. Most Rieber st udies, indeed with the excepti on of Packard (1994), every study that has examined the effect of animation, has used a three-condition experiment to do so. The conditions of static graphics, and animated graphi cs were also compared with the condition of no graphics (text only). In one case, (1988), Rieber also tested the conditions of text and no text along with the graphics treatments. His anal ysis techniques always involved a mixed ANOVA design because he also tested such within s ubject variables as lear ning objectives, latency, attitude, and transfer (near and far). Rieber was not the only researcher to test the effect of anim ation upon attitude (Poohkay & Szabo, 1995; ChanLin & Chan, 1996; Chen, 1995; Curtin & Shinall, 1984; Lai, 2000; Rehaag & Szabo, 1995), however, he has been the only re searcher, as yet, to examine the effect of practice combined with the abovementioned cond itions. He tested three types of practice with adults, behavioral, cognitive and no practice, and with children, he set up a very similar experiment, but the practice was either relevant or nonrelevant to the concepts instructed, Newtonian laws of motion. He found that practice rather than animation was the effective
31 element. He continues to use animation in his CBI programs, but he is mindful of the appropriateness of the animation to the content. He has even viewed animation as somewhat detrimental to adults, who are able to form th eir own mental models of dynamic subject matter. He still believes, however, that animation can benefit children, as well as those, whose mental imagery machines are weak, i. e., little imagination (L. P. Rieber, personal communication, October 6, 1998). His last study of animation did not concern instruction, but rather, feedback. the reply that the computer gives to the learne r in response to a righ t or wrong answer (1996). With regard to feedback, Rieber concluded that animated, graphical feedback can provide greater tacit knowledge than textual fee dback, but he also reported th at explicit understanding was not affected by visual presentation condition. Rieber has always founded his research on Paivios dual codi ng theory in which he sees support for the use of visuals, whether static or animated (1989b). In tes ting the animation effect so intensively, Rieber sought to clarify the discrepancy he f ound between favorable reviews of the use of animation and the la ck of supportive research findings His research efforts no longer concentrate on the effect of anim ation in instruction, but he does continue to endorse dual coding theory and the concepts of asso ciative and referential encoding. As an education researcher, he supports the dual coding theory through his research findings (1990a, 1990b, 1991, 1994, 1997) and in citing the commonly held belief that the two distinct forms of input, verbal and visual, creat e at least two distinct ports of information retrieval, virtually doub ling the chances for recall. Rieber attributes the learners ability to activate meaningful representational and associ ative processing to the dual coding aspect of combined visual/verbal input, especially with resp ect to feedback in computer simulation settings (Rieber, 1996). Riebers studies (1988-1997) have tested the use of animation as an instructional strategy, using animation to provide immediate or realtime feedback to illustrate principles underlying natural phenomena, specifically Newtonian mechan ics or laws of motion. Instruments developed for his research include computer programs that ca lculate the learners inpu t and apply that input directly to the motion, position, direction, speed, etc. of an obj ect portrayed by the computer. Riebers work has centered on content area from th e fields of physics, math and science and the principles instructed are t hose that involve motion or ot her change of physical state.
32 While he supports the use of animation in multimedia presentations, Rieber has become less enthusiastic about the pedagogica l value that can be attributed solely to anima tion in relation to other aspects of a multimedia instructional program. He credits animation with the capability to represent information and relationships in ways that closely resemble natural phenomena but he discredits animations ability to make di screte features and re lationships salient. He recommends animation as enhancement to explicit instruction, yet warns of the possible reverse effect of its use in place of overt verbal explanation (Rieber, 1996, p. 7). In their study based on dual coding theory, Schnotz and Grzondziel (1996) ask, what is the influence of animated pictures on the process and the result of knowledge acquisition? The knowledge under investigation was German L1 illustrated text comprehension, on the subject of the earths rotation relative to time phenomena. They characterize text comprehension as a cognitive process, in which a reader transfor ms the presented linguistic information into a propositional representation, which in turn guid es the construction of the respective mental model (p.7). The propositional representation corresponds to dual coding theorys verbal symbolic structure and the term mental model refers to Palmers (1978) representation of the mental world. Like Palmer, they describe the mental model as the representa tion of subject matter by analogy and like Mayer (1994), they view a dynamic mental model as one that can be mentally played or run in mental simulati ons to represent the dynamic of the subject matter (Schnotz & Grzondziel, 1996, p. 8). In their analysis of mental representations, they define graphic pi ctures as analogies or two-dimensional external models and they postula te that understanding a picture is a process of establishing an analogy between the picture and a corresponding mental model (Schnotz & Grzondziel, 1996). But they claim that a static picture can only present information about the invariant structures of a phenomenon whereas animated pictures can also display the transformation of the structures, if the subject matter is a dynami c one (p. 8). They investigated the use of animation on text comprehension to disc over if it could help the learner to construct dynamic mental models which can be run as ment al simulations. They believe that animated pictures offer external support help to lighten the cognitive load of mental model construction but they also recognize that animated pictures provide more information to be processed by the
33 learner and they reflect, therefore, that the act ual effects of animating pictures, in terms of cognitive demand, remain unknown (p. 10). Under four conditions, two of which tested the effect of animated versus static, graphic pictures, 80 German university students read a 22 paragraph, illustrated, hypertextual explanation concerning the division of the earth into time zones and date zones. The other two conditions were concerned with the type of illustration, or visualization type as the researchers called the variable. In one condition, the illustrations, flat and 2-dime nsional, were called carpet pictures because, like a carpet, they presented th e illustration of the earth as a flat scene with edges that end and begin again on the other side A second condition presen ted circle pictures, where the earth was represented as a globe, without discernible edges. E ach visualization type, carpet pictures and circle-, was tested in bot h static and animated form in a 2x2 factorial design. A pretest instrument asked participants to explain concepts referr ing to time phenomena on the earth. As a learning instrument, part icipants were also given questions about circumnavigation and time changes as a guide to use during reading. Through hyperlinked architecture, students had free access to illustrations and to te xtual explanations and their frequency of access to each was automatically reco rded. In this way, the researchers also studied how student look-up behaviors correspond with visu alization and with graphic presentation type. They considered the frequency of participants access to text, a measure of learner-processingintensity and they regarded this rate as a measure of cognitive effort. Learners displaying more frequent access to te xt were held to have exhibited a greater cognitive effort than those who gathered information more fre quently from pictures. The high frequency of access to text under th e static, circle picture conditi on was taken to mean that the learners showed greater cognitive effort under this condition. This condition, using naturalistic but static pictures, was also th e condition under which the lowest scores were seen for posttest accuracy. For the natural phenomenon of the earths shape and movement, the carpet pictures, while less realistic, always pres ent the earth from an equal pe rspective. Schnotz and Grzondziel posit that a flat view of the earth requires merely a mental translation in order to visualize it in different aspects, whereas a ci rcle picture, or globular view of the earth requires a mental
34 rotation, demanding significantl y greater cognitive effort unl ess the picture is animated. Animated pictures reduce the c ognitive load of knowle dge acquisition as they provide external support for mental simulations. The effect of animation in their experiment was to equalize the two graphic visualization forms to balance the demands for cognitive effort across the two picture-type conditions. This variable, visualization type, confounds the effect due to animation because it is clear that the results depend upon the type of illustration being animated. For engineering design graphics, Asoodeh (1994) found that external stimuli helped students in developing the mental process essential for visualiz ing objects in an engineering system. Compared to the control group, who receiv ed static graphical depictions, the group that received instruction via animation reportedly performed better on a test of mental rotation. With the redundancy of narrated text, Lai (2000) found a significant advantage for the animation of graphics to illustrate to college students the principles of computer programming languages. The researcher claimed that dynamic representation co uld lead to deeper processing [and] help the learner build a runnable mental model of the system (p. 201). Research concerning the use of animation fo r instructional purposes has been conducted in the field of computer algorithms (B yrne, Catrambone, & Stasko, 1996, 1999; Hansen, Schrimpsher, & Narayanan, 1998; Kehoe & St asko, 1996; Stasko, 1996; Stasko & Badre, 1993). These collaborative researchers generally conclude that only small, unrel iable benefits can be attributed to the use of animation, yet they all believe that anima tion may still assist the instructional process of teachi ng computer algorithms. They belie ve that, whether due to faster processing (Kehoe & Stasko, 1996), or as an ai d to mental model building (Byrne, Catrambone, & Stasko, 1996), or simply as an intrinsic motivator (Hansen, Schrimpsher, & Narayanan, 1998; Stasko & Badre, 1993), the particular benefits of animation should be each formally studied, especially with respect to specific tasks and to prior knowledge levels. Animation has been used to extraordin ary effect. Computer generated three dimensionally animated models of patient blood vessels have enabled doctors to make crucial decisions that were previously only possible vi a surgery (Marsh, 1998). Animations in computer simulations have also given dentists a tongues eye view of their patien ts bite collisions. In these cases, animation has provided essential info rmation in a pre-coalesced form. That is, all of
35 the separate pieces of data gathered through s cans and other measurement devices, have already been fit together to form a big picture. In the specific case of EAI Corporations program to give doctors a non-invasive way to measure the di mensions of aneurysms of the abdominal aorta, the program not only delivers an accurate three dimensional model, but it also makes a decision regarding treatment of the problem (1998). In studying the role of animation in inte rpreting representations, Enyedy (1997) used think aloud protocols to see if animation helped young learners to grasp abstract mathematical relationships. The discussions of 13 pairs of middl e school students, concerning the basic concepts of probability, were found to be qualitatively rich er when CBI used animation rather than static graphic representation. In order to support a constructivist pedagogi cal approach, Enyedy sees animation as a way to provide students with ri ch experiences with mathemati cal phenomena. Complementary to the constructivist ideal of self-constructed knowledge, Enyedy s study is also founded on the dual theory of distributed cognition. Distributed cognition describes the ba lance of interaction and experience on the one hand with the repr esentation of knowledge on the other. The representation of knowledge incl udes symbolic notation, metalanguage, graphic representation and mental modeling. Constructivism decrees that the former precedes the latter; experiential knowledge should precede representational knowledge. Enyedys content area was the Probability Inquiry Environment (PIE), a process of inquiry based on a cycle of conjecture, observati on and analysis. The lear ner's task was to judge the fairness of a game of chance. With the goa l of judging, as opposed to playing, the researcher was able to frame the activity as an objective inve stigation and to focus st udents attention on the frequency of outcomes and away from individual e nd results. In effect, the game represented an event tree and a histogram at the bottom of th e game screen represented the event space. The qualitative methods employed think-al oud protocols, audi o and video taping, computer recording of interactions and transcri ptions including gesticular descriptions. It was this last data collection that proved the most revealing. Analysis of gestures enabled the researcher to observe where the students atte ntion was focused and to code their problem solving approaches as either the process perspective or the outcome perspective Gestures along a horizontal plane (at the bottom of the screen) could be coded as outcome perspective because
36 the student was pointing to the event space, while vertical gestures along th e event tree indicated that the focus was on the internal nodes of the tr ee, and that student atte ntion was occupied with reasoning about the process which leads to outco mes rather than with quantifying the results. Enyedy (1997) never quantified the results of this study nor did he make any claims about the superiority of perspective or of the use of animation in teaching mathematical probability. He did however, claim that the anim ated condition was more conducive to a process oriented approach. The vertical movement along the event tree drew atte ntion away from the bottom line of final results. The visual grammar of the event tree supports a process interpretation that narrates the events ove r time and space (p.14). Enyedy postulates that animation helps learners to create a mental simu lation of the process and run that simulation attending to and reasoning about the intermediate events (p.14) He also shows his preference for animation in his interpretation of Tversky (1995 ), with the summation that static images may successfully convey conceptual information, as they are easily interpreted as objects or concepts, but since they are not usually in terpreted as actions or processe s, they do not provide an equal framework with which to generate an event sp ace for new situations. This statement can be interpreted to suggest that animation may be nece ssary to facilitate transfer; the use of existing knowledge in novel applications, at least in certain content areas namely, processes, actions or changes of state, or as Rieber views it, content areas that are congruent to motion or transformation. Other animati on researchers have referr ed to this as process dynamics (Mayton, 1990, 1991; Paivio, 1971, 1990; Schnotz & Grzondz iel, 1996). Mayton equates knowledge of dynamic processes with higher order concep ts involving multivariate changes over time (1990). Paivio declares mental manipulation in imagery to be a dynamic process (1971) and Schnotz and Grzondziel stipulate that animation eas es cognitive load if the subject matter is a dynamic process (1996). Substantive Issues Functions of Animation In a chapter of the AECT handbook of res earch for educational communications and technology entitled, Visual Message Design and Learning , Anglin, Towers, and Levie (2001) analyzed the role of static and dynamic illustra tions. Reviewing 132 primary research studies of static and dynamic graphics as well as a number of meta-analyses of st udies concerning pictures
37 and knowledge acquisition, the handbook states, the re is significant ev idence that generally memory for pictures is better than memory for words. This consistent finding is referred to as the picture superiority effect" (p. 761). In describing Knowltons (1996) semiotic approach to picture analysis, the handbook draws the distinction between digi tal signs and iconic signs. Digital signs bear no resemblance to their referents" (p. 759). Words and numbers are examples of digital signs since they are arbitrary in their appearance, while pictures, ma ps, blueprints, charts a nd diagrams are iconic signs that must in some way resemble the thing that they represent. All previous animation studies, save one (S peelman, 1997), have investigated the use of dynamic pictures, graphs, maps or other iconic sign s that resemble the content they represent. In a review of 42 studies which included at least one animation treatment, the ACET handbook (2001) reports no study that used dynamic text as opposed to animated pictures, graphs or diagrams, nor did the review find any animation studies in fields othe r than science, physics, geometry, mathematics, statistics, or electronics None of the studies of dynamic visuals reported animating the text in addition to (or instead of) the iconic display (the picture). However, Speelmans (1997) experimental treatment presente d line-at-a-time text, as compared to all-atonce textual information. The controlled presenta tion of visual information may be seen as equivalent to an animated condition, because it fulf ills an instructional f unction as an attentional guide (Park & Hopkins, 1993). In summation of the research review, the ACET handbook (2001) presents yet another conclusion that the use of animated graphics doe s not facilitate learning (p. 768); yet, because of the limited number of studies, the inconsistent functionality, relevance, and congruence of the animations tested, and due to a number of met hodological flaws, the verdict is still out on the effect of animated treatments on student lear ning (p. 768). Like many researchers who study animation, the writers of the ACET handbook refuse to abandon the idea that animation is a useful and desirable tool. Therefore, they proffer directives for future re search investigating the effect of dynamic visual displays on learning (p 768). This study follows the directives: (a) that research should be based on a functional framew ork; and (b) that shou ld instruction include content for which external visual information is needed and which requires the illustration of motion (p. 768). In the case of ACETs third dir ective, (c) control for the effect of static
38 graphics, this study cannot strictly comply, since there are no graphics in use. The control is textonly because only the text is set into motion. Nevert heless, the spirit of the third directive is met, if not the letter. Visually, the two treatments ar e identical, with the factor of motion as the sole manipulative. The instructional treatment under study here animated only words and incorporated no illustrations. It is only the text, or digital si gn, that was animated. The words which mean the boy eats a piece of cake were made to move, but there was no boy or piece of cake pictured. It is only the text that was animate d, in order to control focus and direct attention. Learner focus is drawn to the verb and its syntac tical position, not to an image of a boy eating cake. The sentence the boy eats a piece of cake could very well evoke mental imagery, as well as mind wandering what kind of cake?, how old is the boy?, etc. Animating the text reclaims attention and returns it to the co ntent of syntax and away from boys and cake. Attention-getting is no small matter. It is requisite to facilit ate detection and the reception of information (Sharwood-Smith, 1991). Since attention is a lim ited resource, it should not be squandered on extraneous information (Tomlin & Villa, 1994). Illustration is not beneficial if it serves as a distraction. A pictorial representation of the subject of the sentence, a boy eat ing cake, would have a counte r-productive function. It would distract attention from the content area. The senten ce is only a sample, so its subject matter is not germane to the instruction, and it s illustration would likely be de trimental to efficient learning. Such a picture could serve as a type of mnem onic device, but recall of the content would be circuitous at best. The memory of the image of a boy eating cake will not readily transform into the utterance I want to eat a piece of cak e, whereas the mental image of the words, syntactically set into motion, might. In sum, the handbook quotes Rieber's (1994) st ance that visuals are effective some of the time under some conditions" (p. 132). Therefor e, there is a need to delineate the types of illustrations and their functions in order to determine what kind of picture has what kind of effect, why, how, for whom, and under what speci al circumstances. The use of pictures is bounded by guidelines and limitations, the prim ary guideline being the framework of functionality. Each picture must have a determined function.
39 This study incorporates no pi ctures, but since the words are animated, they begin to function as pictures. Therefore, consideration must be given to the functional roles they perform as if they were illustrations. Their function is the illustration of the transformative properties of German grammar, fulfilling an e xplicative role in the case of verb syntax and conjugation. An additional function is to turn th e sample text into a picture. However, what turns text to picture also imposes the guidelines for dynamic visuals. The handbook offers Riebers (1990) conc lusions that animation may be used to assist in attentiongetting, presentation or in practice opportunitie s. Other guidelines offered by the handbook are Park and Hopkins (1993) five instru ctional roles of animated visuals: 1. As an attention guide, the animated visu als can serve to guide and direct the learner's focus. 2. As an aid for illustration, dynamic visuals can be used as an effective aid to represent the structural and functional relationships among components in a domain of knowledge. 3. As a representation of domain knowledge, movement and action can be used to effectively represent specific conceptual details. 4. As a device model for forming a mental image, graphical animation can be used to represent system structures and functions that are not directly observable (e.g., blood flowing through the heart). 5. As a visual analogy or reas oning anchor for understanding abstract and symbolic concepts or processes, animation can make such c oncepts (e.g., velocity) become more concrete and palpable. (p. 767) In this study, animation of the text, to rende r it illustrative should accomplish most of those instructional roles: (1) it should garner and focus attention; (2) illustrate the syntactical relationships between parts of speech; (4) assist in the form ation of a mental image that represents the unobservable grammar transformation; and (5) serve to make the abstract concept of linguistic metamorphosis more concrete and somehow, observable.
40 The Cognitive Theory of Multimedia Learning Mayer and Moreno (2002) summarize a decade of research findings in a collection of seven principles for the design of animation in instructional multimedia presentations. The goal of their meta-analysis was to focus on the role of animation in multimedia instructional messages, and to define the way that it can be used to promote understanding. They calculated effect sizes for 30 experiments conducted by Ma yer and his colleagues over a ten-year period. All of the experiments tested college student s' achievement via problem-solving transfer questions. To Mayer and Moreno, higher transf er score achievements constitute deeper meaningful learning, because tran sfer involves the active proces sing of new material, that is: selecting, organizing, and integrat ing it into the existing knowle dge system. The instructional treatments covered the content areas of weathe r, mechanics, anatomy, mathematics, and biology. The researchers created four short, narrated, anim ated instructional messages that explained: (1) how lightning storms form, (2) how pumps wor k, (3) how car brakes work, and (4) how human lungs work. They also created tw o interactive computer games. One for children to learn basic math functions by playing with a bunny (Moreno & Mayer, 2000), and a botany game for teen and college students learning to design plants th at can survive in various environments (Moreno et al., 2000). Their seven principles for the design of animation in instructional multimedia provide the foundation to create presentations that are cons istent with a cognitive theory of multimedia learning. The cognitive theory of multimedia learning has three theoretical foundations: (1) dualchannel assumption; (2) limited capacity assu mption; and (3) activ e processing. The dualchannel assumption is in concert with dual c oding theory. It assumes there are two separate channels for the intake of information, one for the visual/pictorial and a separate channel for auditory/verbal representations. The limited cap acity assumption is aligned with the bottleneck view (Broadbent, 1958) of the attentional syst em; "only a few pieces of information can be actively processed at any one time in each channel" (Mayer & Moreno, 2002, p. 91). The third theoretical foundation assumes that meaningful learning is achieved th rough active processing of relevant material. Active pr ocessing refers to learner e ngagement through selection and organization of new information, and its inte gration into existing knowledge systems. A bottleneck, or limited view of the attentional system means that only a small part of the available
41 incoming information is ever attended to, therefor e the rest of it is unavailable for processing. This limitation to learning is offset by the fi rst assumption of dual-channel separation. The separation of the two modalities, visual and audi tory, allows attention to twice the amount of input, because the two channels ea ch have their own allotment of working memory. Visual input, such as animation, is more competitive with othe r visual input, such as text, than it is with auditory input, such as narration. An alternate view (Moray, 1959, 1969; Treisma n, 1960) holds that the attentional system serves to either organize or hold in chaotic limbo all that is ever seen, h eard or otherwise taken in by the eyes, ears or other senses. In other word s, all input is availabl e for processing, but only some of it can be organized and integrated in to the knowledge system at a single time. This means that some input may be remembered in an illogical fashion, that is, not associated with any other piece of knowledge in memory. Input see n, read or heard, but not learned, may have no associative connections, but it could still have referential node s ready to become connections. Pictorial representations often behave this way, which may explain the clarity of research showing that illustration-enhanced instruction aids learning significantl y (Plass, Chun, Mayer, & Leutner, 1998; Yoshii, 2000; Yoshii & Flaitz, 2002) Non-verbal propositions don't require a web of connections in order to remain in memor y. A picture can "float" in the "back of the mind" and only later, spring to the fore front of consciousness, availa ble for active processing. Although this alternative view of atte ntion competes with Mayer and Moreno's second assumption above, it only reinforces the other tw o assumptions. Active processing can still occur using stored memories, especially if they are pictorial memo ries that were dually-encoded in synchronization with relevant verbal input. The following seven design principles are cons istent with cognitive theory of multimedia instruction, under either view of the attentional system. Th ey are the summation of Mayer's collaborative animation research, and they serve as a guide to the effective use of animation for instructional purposes. 1. Multimedia principle 2. Spatial contiguity principle 3. Temporal contiguity principle 4. Coherence principle
42 5. Modality principle 6. Redundancy principle 7. Personalization principle Multimedia principle. Learners achieve deeper learni ng from instructional texts, when words are accompanied by pictures. The addition of a "pictoral explanati on (i.e., animation) [ substantially supports] learne rs' problem-solving transfer pe rformance" (Mayer & Moreno, 2002, p. 93). Recognizing that not all animations are e qual, Mayer's six other principles form the guidelines for making animations instructionally e ffective. In the present experiment, the static version of the instruction exemp lified the multimedia principle, even though it had no pictures or animation. The instruction, or "tex tual explanation" was delivered on an audio channel, while the visual channel offered an entire ly different text. Instead of me rely reading what was already heard, the viewers were presented sample sentences, meant to illu strate the explanation. Offering dual channel input and verbal e xplanation accompanied by visual i llustration, the static version also employs the multimedia principle as Mayer et al envision it. Spatial contiguity principle. Animations should be placed in close proximity to the texts that they are meant to illust rate. Moreno and Mayer's (1999) study revealed a moderate median effect (0.48) for the placement of text and its accompanying animation, concluding that when the pertinent text is found far from the relevant pict ure, e.g., at the bottom of the screen "learners must waste limited cognitive capacity in searching for the portion of the animation that corresponds to the presented te xt" (Mayer, et al., 2002, p. 95). Temporal contiguity principle Analysis of eight experime nts conducted in four studies (Mayer & Anderson, 1991, 1992; Mayer & Sims, 1994; Mayer et al., 1999) yielded a strong and consistent effect for the aspect of simultane ous presentation of animation with corresponding textual narration (median effect size = 1.30). For this study, the audio pres entation of instruction was carefully synchronized to the animation, in agreement with the temporal contiguity principle. Coherence principle Exclusion of extraneous words, sounds, music, decoration, etc., will minimize cognitive overload. In their analysis of five experiments (Moreno & Mayer, 2000; Mayer et al., 2001), Mayer and Moreno (2002 ) found strong and consistent support for the effect they term, the coherence principle. When explanations of how lightning works were
43 accompanied by background music, environmental sounds or interesting video, participant performance on problem-solving transfer tests su ffered substantially (med ian effect size = 0.90). "The learner may attend to the irrelevant mate rial and therefore have less cognitive resource available for building mental connections betw een relevant portions of the narration and animation" (p.95). The coherence principle was adhe red to in the making of the presentations for this study. Redundant visual text and artistic features, such as colors, fonts, geometric shapes and gratuitous attention grabbers we re systematically excluded from the programs, in order to create maximum coherence between the visual sample sentences and the auditory explanations. Modality principle. Connections between words and their corresponding pictures are built better when they are presented in different modalities or channels. Viewing visual displays while reading visually presente d text is challenging to the c ognitive capacity. Since the eyes cannot be in two places at once, there must be an attention shift from the written word back and forth to the onscreen i llustration. Text that is presented au dially may be attended to while the eyes are free to view supportive, but different information. The modality principle of the cognitive theory of instructional multimedia opera tes within the limits of the attentional system to stretch working memory across two receptive channels (eyes and ears), presumably increasing cognitive capacity. A meta-analysis of six experi ments in three studies (Moreno & Mayer, 1998; Mayer & Moreno, 1999; Moreno et al., 2000) show ed support for the presumption (median effect size = 1.17). In presentations explaini ng how brakes work, how plants grow, and how lightning forms, animations were found to be more effective when accompanied by narration rather than printed (on-screen) words. Here agai n, both treatments of this experiment follow this principle by offering a verbal explanation in an auditory channel and presenting a visual sample supportive of the explanation. Redundancy principle. This principle springs from two experiments (Mayer & Moreno, 2002) in which a presentation, explaining the form ation of lightning, utili zed on-screen printed words in addition to audial narration and animated depicti on. The redundancy principle states that "less is more". The redundant visual text resulted in less, not mo re learning. According to the same rationale as the modality principle, th e extra text adds an unnecessary burden to the visual processing mechanism. It hinders rather th an helps the viewer to integrate the illustration with the explanation. In accordan ce with Mayer's redundancy princi ple, the verbal explanations
44 are given in verbal form only, while the visual te xt is comprised of sample sentences, which are different, but illustrative of the narrated text. Personalization Principle The seventh principle of the cognitive theory of instructional multimedia design is the preference for a conversational style of address over a more formal style. Constructed with first and second pe rson pronouns, "I" and "you," such a convention is thought to involve the learner more personally. This principle holds that learners "work harder to understand an explanation when they are pers onally involved in a c onversation" (Mayer & Moreno, 2002, p. 97). This program did not follow th is principle, because the narration never used any form of address to the listener. Congruency In articles describing his research on an imated computer instruction for Newtonian physics, Rieber has defined the purpose of anim ation to be dependent on the congruence of the content with the need for motion. Mayers (1997) work agrees with this in so far as the motion of a pump or a piston in mechanical science is a fiel d content that calls for animation to describe it to the greatest effect. Language, an intangible element, is not typica lly thought of in terms of a physical process requiring motion to describe its different states. However, many structures of language can be seen as transformations or movements from one state to another. Thus, motion or animation is indeed called for to best describe some features of linguistic subject matter. Such a circumstance is called for primarily in the area of word order or syntax. When the grammatical structure of a language dictates that a certain type of word be placed in the sentence according to a strict or rigid rule, then the concepts of position and change of position become key elements of the content of instruction. Change of position may be represented as a motion from one position to the other, in effect, a transformation. The acquis ition of language involves mastery of the ability to create novel utterances based upon the gramma tical rules that govern the language involved. When a speaker wishes to communicate a novel idea by addition of an element, e.g., a time expression or a word of negation, which dict ates a change in word order, a grammar transformation must take place. Some, but not a ll, grammar transformations are well represented by motion. In order to be congruent, animated input must show an appropriate relationship between mental representation and grammatical form (Hwang, 1999).
45 Mental Model Building vs. External Imagery Animations that depict the motions of grammar transformation actually impose a mental model upon the viewer. Learners are no l onger free to develop their own images of how things work, unaffected by th e instructional presentation. In the theater, directors will usually avoi d seeing any performance of a play before envisioning their own production, played out in their head, as it were. Some would argue that effective language acquisition woul d entail the learners constructi ng their own mental models of grammar processes. Self-constructed models w ould be more readily recalled. That argument would be strong indeed if ther e were a guarantee that learners mental images were accurate (Shih & Alessi, 1994). Schnotz and Grzondziel (1996) praise the us efulness of animation to support mental model construction and to ease the cognitive efforts involved in mental simulation, but they also warn of the danger with animated pictures [in] that individuals are pr evented from performing relevant cognitive processes by themselves and, thus, process the subject matter more superficially than with static pictures (p.10). Blaming oversimplification, Lai (2000) conte nds that dynamic visuals could lead to misconceptions. Lais learner-controlled CBI also allowed observation that animation of visual representation resulted in faster, i.e., sh orter time spent viewing the instruction. Processing Speed and Prior Knowledge It is suspected that animation not only ease s cognitive efforts but also speeds up the pace of cognitive processing as well. In accord with Lai (2000), Schnotz and Grzondziel (1996) report that a major feature of animation is faster processing of new subject matter. By helping learners to comprehend a concept holistically, portraying relevant information simultaneously, animation also tends to shorten th e time needed to proce ss the information. Pilot testing in preparation for this study has shown that in a user-controlled CBI, learners moved considerably faster through the pr ogram under the animated condition. Increased speed in the processing of new information may very well be more detrimental to mental model building than it is helpful for c ognitive effort. Pilot testing for this study has also shown that, under the animated c ondition, students seemed to feel more confident about their understanding of the subject matter, as eviden ced by faster processing speeds, but it was not
46 clear if that confidence would b ear out in terms of production accuracy over time. They get it better now, but will they know it better later? The ideal of graphically represented informati on is that everything is presented at once as opposed to the linear presentation of facts endemic to prosaic te xtual presentation. In the case of a textual graphic, such as figure 1, all of th e text is presented at once, everything could conceivably be read at once, so all of the information is available simultaneously. Paivio conveys this by means of the overlapping circles at the right of the illu stration. All associations in the nonverbal system are available simultaneous ly, while information in the verbal system is processed (and presumably retrieved) in a lin ear, hierarchical fashion one thought leads to another. This simultaneous availability of so mu ch information is what makes the nonverbal structure so powerful as an organ of memory and recall. Knowledge that is symbolically represented in the nonverbal system supports memory associations that are helpful for reason and recall. Research supporting dual coding theory has shown that knowledge that is presented in simultaneously available graphic form helps to stimulate the imagery that creates nonverbal associative memories, but researchers also warn th at this simultaneous availability of information can be too much. Learners with low prior knowledge must invest extra mental effort to digest and encode graphical information. Students with limited dom ain knowledge sometimes view graphics as an excess complexity and incomprehensible informa tion if the connections are not transparent to them (ChanLin & Chan, 1996, p. 5). Conjoint Retention Theory Paivio's encoding theory concerns the pro cessing of incoming information. It suggests ways that instructional material can be presente d in order to maximize the intake of input. A tangent theory by Kulhavy, Lee and Caterino ( 1985), concerns the retention of processed information and the recall event for dually en coded (inter-connected) data. "The conjoint retention hypothesis is essentia lly a rendition of the dual codi ng approach (cf., Paivio, 1971), in the sense that linguistic/verbal and perceptual/s patial representation in the cognitive milieu are
47 treated as separable coding proce sses" (p. 29). The difference is that conjoint retention theory focuses more on the recall activity rather than the initial encoding process. Dual task objectives. Kulhavy and his colleagues (Abe l & Kulhavy, 1986; Kulhavy, Lee, & Caterino, 1985; Kulhavy, Stock, Peterson, Prid emore, & Klein, 1992; Rittschof & Kulhavy, 1998; Verdi & Kulhavy, 2002; Verdi, Johnson, Stock, Kulhavy, & Whitman-Ahern, 1997) have studied the learning of textual information using geographic maps. Maps are, at once, pictoral text adjuncts, as well as supplie rs of additional (mostly spatial) information. An example of text content is a story about events in a town and its various buildi ngs and map sites (e.g., church, townhall, etc.). The map contains features representative of the town, therefore it serves as a structural support of the text info rmation, but it is also a thing to be learned in and of itself. In Kulhavy's work, the learners' tasks involve recall of both text features (e vents) and map features (spatial relationships). It is a dual task and his experiments have shown significant support for dual-coding and conjoint retention theory for recall. One task was to reconstruct map information, and a second was to recall facts about the events in the story. Instead of the two disparate tasks causing twice the cognitive load, the dual encoding at the time of processing makes them both easier, because each task eases the other. The mapping of the verbal memories of events onto the visual map features at the time of learning provided twice the memory storage capab ility (dual-coding theo ry), and as the mental image of the map features helped to recall story events, the verbal memories helped to reconstruct and reinforce the mental image. Instead of suffering, both tasks benefit from one another (conjoint retention theory). The text and the map have features in co mmon that allow an overlay as depicted in Figure 2. Imagine that the circles marked "verbal information" were holes, and that the small Verbal information Map features Figure 2. Conjoint Retention Theory for Map Features (adapted from Kulhavy, Stock, & Kealy,1993)
48 Semantic matching. The key phrase above is "semantically matched." The verbal information can only have adherence to images that are meaningfully representative of the verbal proposition. Meaningful representation can be created either through convention (Winn & Solomon,1993) or by relationship to reality (the de gree of realism). The animations created for this study have a conventional re lationship to the propositions e xplained. It is a convention to expect that a sequence of words will be perceived as having a left-to-right order, so that the word farthest left is in the "first position" and the one farthest ri ght is "last." Another convention dictates that a movement toward the right signi fies a movement toward "the end." The spatial arrangements of linguistic elemen ts plays a large role in the es tablishment of meaning (Winn & Solomon, 1993). Spatial conventions An intent to examine the effects of animation raises the important issue of what to animate, as well as how. Linguistic elements, letters, words, phrases, etc, can be isolated and made to move, but the choice of wh at to isolate, and the way it is transformed, can not be arbitrary if the animations are to be meaningful in a determined way. Animation can be used to "suggest cognitive correspondences between mental spac es and real ones" (Morrison, Tversky, & Betrancourt, 2000, p. 2). In fact, in a study of the effect of spatial arrangements on viewers' interpretation of English and nonsense sentences, Winn and Solomon (1993) found that consistent conventions of spatial arrangement of objects, may "act w ith at least some of the force of grammatical rules" (p. 30), th erefore such arrangements requi re specific prescriptions that reinforce the semantic content. To test the conve ntional expectations that influenced the design of the animations for this experiment, a small case study was performed. The details are described in the methods section of this document (chapter 3). The purpose of the case study was to determine the extent to which the animations were consistent with vi ewers' beliefs about the meanings of those movements. circles at each intersection of the grid, marked "map features," were knobs that fit snuggly into those holes. The map and the text, when semantically matched, would adhere to each other and support recall of both. This adhere nce creates conjoint retention.
49Focus Shift In a study of how graphic styles ( none, static, animated ) influence CBI of a molecular biotechnology concept (Basic Recombinant DNA), ChanLin and Chan (1996) discovered through interviews and observations during think-aloud protocols, that learners shift focus switch their attention back and forth between animations and text. They noticed that some students began by utilizing the textual informa tion available. These they termed verbally oriented. Those who started by studying the grap hic presentations were considered visually oriented. They contend that focus shift be tween the two presentational formats requires more effort for verbally oriented learners es pecially if the graphics are animated. Instructional graphics have been de fined by Poohkay and Szabo (1995) as any representation of an object concept or process, as perceived through the eye, which does not rely on the use of text or numbers. Animation refers to the use of a series of graphics which change over time (p. 1). The graphic animations to be tested in this study ar e textual animations or moving words. This means that they could fo rm associative references in each symbolic subsystem while creating referen tial connections between them. Op erating in both a verbal as well as nonverbal system, textual animations dont require any focus shift. They are designed to be read even while they move. The observers c onscious awareness is occupied with reading, or the verbal intake of stimuli, while the readers eyes are manipulated to move in a new symbolic pattern. A nonverbal image is planted along with the verbal message. This differs from Poohkay and Szabos view of graphic representation becau se, like Paivios letter E, text becomes a graphic when it is animated. It also departs from Riebers view that animated subject matter be congruent to movement or that the concepts instructed are those that have transformational properties in the concrete world. Text doesnt ex ist in the concrete, tangible world, only representations of linguistic symbols exist, such as printed letters and recorded words, and these things have no natural properties that could be said to move or transform in a determinate process. In this sense, text is not congruent to animation, yet an imating text is conducive to learning according to dual coding theory. Animating text could force linguistic thought back into nonverbal imagery, forging the cross-system refere ntial connections that would st rengthen retention and recall.
50Speelman (1997) is the only researcher thus fa r to examine the effects of animated text or line-at-a-time animation. With a posttest-onl y control-group design, she found a significant difference in the short-term retention scores of graduate and postg raduate instructional technology students, which she attri buted to the effect of animati ng the text in the experimental instructional treatment compared to static textual presentation.ChanLin and Chan (1996) note that, for verbally oriented students, the animations in their study were distractions, perceive d as interrupters to students attention, requiring added mental effort. They observed that:Some students constantly pointed at th e verbal explanations when viewing animation. This behavior reflected the limited working memory. With limited free memory, students needed frequent access to text and animation so that they could keep these two information sources in working memory. (p. 9)Their recommendation, in view of this focu s shift, is to match the pace of visual information with the presentation of verbal in formation. Matched pacing or synchronicity of information sources is key to effective referent ial processing. ChanLin and Chan (1996) feel this is especially so for less experienced learners. Matched pacing is a majo r consideration in the design of the present experiment. Redundant Narration Matched pacing is achieved through redundant audio narrati on of visual information. Narration is considered redundant because it offers the same content as the visual display, but in keeping with the redundancy principle of multimedia theory, the narration is not identical to the visual information. That is, the audio is not simply an oral reading of the textual explanation, but it is equal in the amount of inst ruction that is offered, and is therefore, redundant. As Mayer and Andersons (1991) research shows, text and gra phics presented in tandem enable students to construct better referencing connections (May er & Anderson, 1991; Pack ard et al., 1994). Mayer and Anderson also studied the inte ractive effects of animation w ith audio presentation of verbal information. In the conclusions from their (1992) study, entitled, Animations need narrations: An experimental test of a dual-coding hypothesis , they support the hypoth esis of referential connections between visual with verbal st imuli. The significant achievement of their animation plus concurrent oral narration treatment group also suggests th at learners were aided by the synchronicity of verbal and visual information. The narration sets the pace of the instruction. It
51determines the amount of processing time allotted a nd it diminishes focus shift. Orally presented, verbal stimuli can be processed at the same time as graphical input. A study by Okolo and Hayes (1996) lends merit to this view. The researchers evaluated the use of children's literature presented via one of three conditions: (1 ) an adult reading a book to the child, (2) the child reading a CD-ROM, si lent version of the book with high animation, and (3) a static CD-ROM version of the book without animation. They found th at children showed a preference for the animated computer version, sp ending almost four times as much time on the reading task than students in the adult-reader condition. However, it was in the latter condition that students obtained the highest scores on co mprehension questions (1996). This suggests that narration adds a potent value to instructional pr esentation. The adul t-readers in the abovementioned study were not recorded, however, therefore they offered the additional feature of live verbal interaction. The ch ildren could ask them to repeat words and to explain concepts. Lai (2000) has recognized that audio elem ents can gain and maintain attention throughout a multimedia program (p. 204). Usi ng narration synchronized to the visual presentation, Lai hoped to direct the viewers atte ntion to details of the graphic representation and to influence the pace of student-controlled instruction. While textual animation can control attenti on, narration can control pace. Both elements are capable of being used to direct attention and to affect the duration of the learners focus. Rehaag and Szabo (1995) used redundant audio to narrate a CBI mathematics lesson to 82 high school students. They reported a reduction in the time required to complete practice questions. Childress (1995), Mayer and Sims (1994), a nd Rehaag and Szabo (1995) have also studied the issue of concurrency or synchronicity of narration with visual presentation. Rehaag and Szabo found positive effects, for lower ability students, if the textual explanations were simultaneously narrated (as opposed to not at all). The design for the narrated condition of this study reflects their findings for re dundant audio or dual channeling. Both treatments of this study incorporated recorded narration, whic h determined the pace of the treatment, matching the static presenta tion to the animated. R ecording the narration not only kept it constant across treatme nt groups, but it eliminated the presence of a live instructor, whose gestures and expressions c ould be additionally informative.
52What the Present Study Will Add to the FieldAnimation and CBIResearch studies on the animation effect by inve stigators in the fields of physics, science, mathematics, medicine and microbiology (Enyedy, 1997; ChanLin, 1998a; ChanLin & Chan 1996; Hays, 1996; Large, 1996; Mayer, 1997; Mayer & Anderson, 1991, 1992; Mayer & Sims, 1994; Mayton, 1991; Nicholls, 1996; Packar d et al. 1994, 1996; Rieber, 1988, 1989a, 1989b, 1996a, 1996b; Schnotz & Grzondziel, 1996; Szabo & Poohkay, 1995, 1996; Williamson & Abraham, 1995) have yielded support for Paiv ios dual coding hypothesis, which states, indirectly, that two sources of i nput are better than one. Text and pictures are better than either one alone. Paivios theory postu lates distinctly separate ment al mechanisms for processing verbal (textual) as opposed to visual (graphical) input. Riebers work examined the role of animati on as a unique element affecting the reception of visual input that is distinct from static graphic input. Mayers work investigated the effect of auditory text, a factor that is distinct from writte n text, while still a form of verbal input. Both of them reported beneficial attribut es of dual coding of information. Unique to this study is the content area of foreign language instruction. Previous research has examined the instruction of physical processes, which involve actions and consistent motions: motions that are real in that the processes actually do oc cur in a physical space. Pumps exist, and pistons do go up and down, a nd everyone agrees upon which way is up. The motions of grammar transformation are imaginar y. The directions of the action model are arbitrary and could be unique to the speaker People need not share the same idea of directionality in relation to gr ammar processes. (The finite ve rb doesnt necessarily rise up when displaced by a modal auxiliary.) Therefore th e congruency of the process is not as clear in the field of language as is pos sible with active processes pert aining to science or physics. In terms of animation, this study borrows from re search begun in physics, science, math and medicine, and extends it to the field of foreign language education. From dual coding theory, it draws the hypothesis that, although textual, lingu istic transformation models can be made to represent graphic imagery, which means that grammar could be a content congruent to animation.
53Another unique aspect to this st udy is the use of text rather th an pictures as the object of animation. With the single exception of Speelman (1997), all animation research has dealt with graphics : diagrams or pictures. Speelmans dissertati on study tested the effect of line-at-a-time animation as opposed to static presentation of textual explanations. She found a significant effect for animating the textual presentation of inform ation technology content. The animated condition tested in the present study differs fr om Speelmans in that the textual explanations are not the objects to be animated. Instead, the examples are animated to illustra te the dynamic process. In this sense, this experiment is similar to all of the other studies that have examined the animation of graphic objects with textual explanations held static. Because the co ntent areas examples are textual, animating them transforms them into graphic objects yet the studys focus remains animated text rather than animated graphics. Theref ore, this study is the firs t of its kind; the first to test the effect of animating text to stimul ate nonverbal association (creating a graphic visual out of a linguistic element), and the first to examine the use of animation in the field of computer-enhanced language teaching. The key inquiry: "What happens when you put text in motion?" is different from previous studies on th e effect of animation, which ask: "What happens when you put graphics in motion?" German L2 Pedagogy The experiment conducted here adds to the field of foreign language pedagogy in the computer-based environment by providing evidence that describes the effects of the multimedia element, animation, on specific skills of Germ an language acquisition, namely, performance of conjugation and syntactic rules du ring sentence rewrite tasks. Th e resulting advice should be most useful for designers of multimedia for German grammar instruction, but it may also be applied to the use of animation for multimedia grammar instruction for other foreign languages, such as Dutch, Swedish or English as a second language, especially when the point of grammar involves a consistent change of position or form. The study conducted here was a plan to isolate some particulars of animation for instructi onal purposes, more narrowly, for foreign language education, and more specifically, for German language grammar principles.
54SummaryDual coding theory is a description of how the mind perceives knowledge of the world in a mental representation. This revi ew of the literature has exam ined how dual coding theory has informed educational researchers interested in the use of multimedia, particularly animation, to support learning and strengthen recall abilities. The implications of the theory for second language acquisition have been di scussed and the applications of a dual coding approach for CALL, or foreign language teaching in the com puter environment, have been considered. This chapter has also reviewed literature from other disciplines (physics, biology, medicine, math and statistics) that investigated the effect of animation on learning outcomes. Substantive issues associated with the animation of text for grammar instruction have also been raised in this chapter. Alternate views of th e structure of grammar have been described along with some approaches to teaching Germ an L2 that are based on those views. The question of congruency of subject to an imation has been considered along with the issue of externally imposed imagery as oppos ed to self-constructed mental modeling. This research stands on the precept that novice L2 learners with low prior knowledge of linguistic infrastructure will benefit from externally co mposed imagery of dynamic grammatical principles (ChanLin, 1998b; Mayton, 1991; Paivio, 1990; Schnot z & Grzondziel, 1996). The instructional design conforms to the cognitive theory of multimedia learning (Mayer & Moreno, 2002) and adheres to conjoint retention theory's (K ulhavy et al., 1985; Kulhavy, Stock, & Kealy, 1993; Verdi & Kulhavy, 2002) requirement of semantic connection between information and illustration. The next chapter provides detailed informati on about the participan ts. It describes the development of the instructional treatments a nd the tests that measure prior knowledge and posttreatment achievement. It also describes the data collection procedures that form the design of this research.
55Chapter 3MethodIntroductionThis chapter is organized into five sections including the research questions, participant information, instrumentation, and data collection procedures. This chapte r also describes the development of the multimedia instructional program used in the experiment. Research QuestionsThe following research ques tions were investigated:1. What is the effect of animation of sample sentences depicting a grammatical process, explained in a concurrent na rrative (audio) presen tation, on test scores measuring a language learners ability to rewrite German indicat ive sentences adding modal auxiliary verbs, compared to static presentation of the same sample sentences?2. Is the effect of animating the sample sentence s different for each of the two discrete tasks (conjugation and word order) invo lved in the rewriting process?The first question represents the prim ary focus of the study and addresses whether animation provides benefits for conjugati on and word order performance beyond those achievable through an otherwise identical, static presentation. The second question investigates animation more closely, trying to determine specific circumstances where animation might be helpful. The overall task consists of discrete parts (conjugation and word order), so it was possible to see exactly how animation affected completion of the task. It was expected that participants viewing th e animated version would achieve higher scores on the criterion tasks of conjugation and word order when co mpared to participants in the static presentation group.Participant InformationParticipants were students enrolled in three sections of beginning German I at a state university in the southeastern United States. Ages of participants ranged from 1860 years, with a large proportion in their early 20's. Sixty-six students registered fo r one of three sections of the course designated GER1120. Due to drop-outs and ab sences, there were 44 participants available
56 to take the pretest. Three partic ipants, who had taken the pretest, were subsequently absent at the next class period and consequentl y, did not view the treatments or take the posttest. However, there was one participant who vi ewed the instruction and took the posttest, but who was absent for the pretest portion of the experiment. One ot her student exercised the option offered in the study's written informed consent form ( Appendix A ), and accordingly declined to take the posttest. Of the three class sections involved, secti ons one and two were daytime classes led by a female teacher in her 20s. Her classes met four times a week for 50 minutes per session. Section three was an evening class that met twice a w eek, two hours per session, and had a male teacher, over 60. The two instructors had roughly equal experience of two years each for teaching beginning German at the university level. Neit her teacher had utilized technology in their classrooms prior to the study. Both teachers had been interviewed weekly to monitor whether modal verbs had been inadvertently ta ught or modeled within each section. As a form of control for the variation within each section, the participants were matched into pairs, based on their class ranking. This ranking was calculated using four prior chapter exams covering structures of German grammar, sc heduled as part of the regular course syllabus. The exam schedule was the same for each section. By the time of the data collection, all three class sections had been instructed and tested on the grammar and vocabulary of the first four chapters of their textbook, Deutsch Heute (Moeller, Liedloff, Adolf, Kirmse, & Lalande, 2000). Participants' scores on essays, oral exams and clas s participation were excluded from their class ranking. After ranking the students of each class section, they we re paired by drawing a line under every second student. One member of each matched pair was randomly assigned to either the static or animated conditions and the ot her member was then assigned to the opposite treatment group. There were 41 participants avai lable to view the instructional treatments and to take the posttests, but only 36 represented matched pairs, (i.e., 18 matched pairs). As seen in Table 1, there had been 21 matched pairs available for th e pretest, but three ab sences for the posttest eliminated three of those pairs, leaving three unmatched participants, in addition to the two unmatched participants of the pretest total.
57Table 1Number of Participants by SectionSections Total 1 2 3 Registered in class 66 22 22 22 Drop outs 13 4 3 6 Absent for pretest 9 4 2 3 Available for pretest 44 14 17 13 Absent for posttest 3a 0 3a 0 Declined to participate 1 1 0 0 Available for posttest 41 13 15 13 Total of: Matched pairs available (pretest) 42 14 16 12 Matched pairs available (posttest) 36 12 12 12 a One participant who had been absent for the pretest, was present for the posttest only. All participants were given code numbers to protect their identities. The code numbers provided an indication of cla ss ranking, class section, and trea tment group. Test packets were then prepared, which were marked with the pa rticipants' code numbers but no names. The two coded test pairs (preand post-) were enclosed in a third sheet of paper containing instructions for participation in the study ( Appendix A ). To enable distribution via a roll call, this instruction sheet was marked with the names of each participant according to their coded assignments, but it also instructed them to discard the paper or retain it for themse lves to preserve their anonymity. Since the participants had been assigned according to their ranki ng in their respective courses, and the test packets had to be prepared in advance, participant absence on the day of data collection resulted in unbalanced groups. Table 2 gives the samp le sizes for the three class sections. Table 2 Sample Sizes for Class Secti ons by Presentation Format Sections Presentation Format Total n 1 n % 2 n % 3 n % Static 23 9 39 7 30 7 30 Animated 21 8 38 7 33 6 29
58 A pretest survey was administered to dete rmine participants' prior knowledge of German and of other foreign languages, the time spent st udying languages, the skil l-levels achieved, and the amount of time spent living in Ge rmany or other foreign countries ( Appendix B ). Tables 3 through 10 break the survey information down into greater detail, which de scribe the participants of each group. Table 3 describes the participants' prior e xperience with German language learning. Of the 44 participants, 57% identified themselves as just starting to learn German. A little more than a third of all participants had previously studied German for 1-2 semesters, a little less than a third had lived in Germany for six months or longe r, and 9% said that German had been used in their homes. Table 3 Experience with German by Presentation Format Presentation Format Option Static n % (n=23) Animated n % (n=21) Total n % (n=44) 2 Just starting 13 57 12 57 25 57 0.001 ns Studied 1-2 semesters 7 30 8 38 15 34 0.03 ns German used in the home 2 9 2 10 4 9 0.01 ns Lived in Germany 7 3 6 3 13 30 0.02 ns Note. Percentages total more than 100% because part icipants were able to select more than one option. *p<.05. ** p<.01. ***p<.001. "ns" = not statistically significant (p>.05). Several respondents checked more than one category for this survey item. Three respondents had said that they were just starti ng to learn German, but they also checked the categories of more experienced learners. Thes e included one who had been exposed to German in the home, but was just beginning to study it, and two who had lived in Germany for a matter of months but had not undertaken formal study. Seven respondents had lived in Germany between two and eight years, and had studied th e language formally for two to twenty years. The static and animation groups were iden tical with respect to the percentage of beginners, with 57% of each group identifying themse lves as just starting to learn German. The animated presentation group had an 8% higher percentage of respondents who had studied 1-2
59 semesters, but this difference was represente d by only one person. The groups were similar in terms of the percentage of participants who ha d used German in the home, and who had lived in Germany (3% of each group). Table 4 Time Passed Since Last Course Of German Study of 1-2 Semesters (n=15) Presentation Format Static Animated How long ago? # of responses # of responses 2 Since High school 1 year 2 years 3 years 4 years 6 years 7 years 8 years 9 years 20 years 25 years 1 1 1 1 1 1 1 2 2 1 1 1 1 2.29 ns 0.93 ns 0.004 ns 1.12 ns 0.93 ns 0.004 ns 0.93 ns 0.93 ns 1.12 ns 0.004 ns 0.93 ns Note. *p<.05. ** p<.01. ***p<.001. "ns" = not st atistically significant (p>.05). Of the 15 participants who ha d studied German for 1-2 semesters, the survey asked them to estimate how long it had been since their last time of formal study. Table 4 shows a range between one and nine years, with a single re spondent for nearly each number in between. Two responses merely stated "high school," and one from each group claimed a time span of twenty years since their last German class. There was also one participant who had not studied German for over 25 years. This same participant had so me prior experience learning French and Italian, as can be seen in Appendix B Chi-square analysis revealed no statistically significant difference between the groups. There were 13 participants (3% of each group) who had lived in Germany. Table 5 clarifies the length of time that these 13 were there. Of the seven in the static presentation group, four had spent only one year or le ss, while all six of those in the animated presentation group had spent at least two and up to eight years living in Germany. Of the other three members of the static group who had lived in Germany, one had spent three years, one eight and the last, had
60 lived in Germany for 14 years. The static presentation group had a total of 24 years experience living in Germany, and the animated presentation group had a total of 25. This seems to be a similar amount, but the distribution was differe nt for each group. There was a much broader spread of experience across the static group, while the experience of those in the animated group was more concentrated. Given the sample size, ch i-square analysis did not show a statistically significant difference between the two groups. Table 5 Time Spent Living in Germany Presentation Format Static Animated Total How long? n % (n=23) n % (n=21) n % (n=44) 2 0 time 3-6 months 1 year 2 years 3 years 4 years 5 years 8 years 14 years 16 69.6 3 13.4 1 4.3 0.0 1 4.3 0.0 1 4.3 0.0 1 4.3 15 71.4 0.0 0.0 1 4.8 2 10.0 1 4.8 1 4.8 1 4.8 0.0 31 70.5 3 6.8 1 2.3 1 2.3 3 6.8 1 2.3 2 4.5 1 2.3 1 2.3 -------2.94 ns 0.93 ns 1.12 ns 0.46 ns 1.12 ns 0.004 ns 1.12 ns 0.93 ns Note. Percentages do not to tal 100% due to rounding. *p<.05. ** p<.01. ***p<.001. "ns" = not sta tistically signif icant (p>.05). Table 6 shows that both groups had an equal number who considered themselves to be skilled in German, ranging from "survival le vel" to "near-native" ability. Again, the animated presentation group had less variability, with four out of five falling into the single mid-range category, "fluent, with poor grammar," but chi-squa re analysis did not show this difference to be statistically significant ( p > .05).
61 Table 6 Self-rated Ability for German Presentation Format Skill Level Static n % (n=23) Animated n % (n=21) Total % (n=44) 2 None Survival Level Fluent (w/poor grammar) Near-native 18 2 2 1 78 9 9 4 16 1 4 76 5 19 0 77 7 14 2 ------1.91 ns 1.00 ns 0.93 ns Note. *p<.05. ** p<.01. ***p<.001. "ns" = not st atistically significant (p>.05). Overall, there was only a single participant who had absolutely no prior exposure to any foreign language. All others had ei ther lived in a foreign country for a month or more, or had taken at least one semester of foreign language instruction prior to the time of this study. Table 7 Experience with second languages other than German Presentation Format Static Animated Total % n % n % # of foreign languages (n=23) (n=21) (n=44) 2 0 1 2 3 4 7 8 8 0 0 30 35 35 0 0 3 11 2 4 1 14 52 10 19 5 23 43 23 9 2 1.63 ns 1.37 ns 2.46 ns 2.35 ns 1.12 ns Note. *p<.05. ** p<.01. ***p<.001. "ns" = not st atistically significant (p>.05). Table 7 shows participants' experience w ith languages other than German. All other participants had studied at l east 1-2 semesters, or had some abilities in a second language through natural acquisition. There was only one pa rticipant who reported no experience with second language learning. Anothe r participant reported no expe rience with language learning, but had lived in Germany over several summers. This table shows an imbalance between the two groups, with 30 % of the static group having no experience with foreign languages other than
62 German, compared to 14% of the animated group. This percentage actu ally represents only four people, but as shown in Table 8, the animated group also had four members who had studied three or more foreign languages beside s German, compared to none in the static group. This shows a practical advantage for the animated group in their prior knowledge of other languages and their experience with fore ign language study. Given the samp le size, however, there was no statistically significant differen ce revealed through chi-square an alysis. Table 7 also shows that more than three quarters of a ll participants had e xperience with at leas t one foreign language other than German. Table 8 describes participants' experience w ith instructed second language learning other than German. This table shows how the language s are distributed between the two groups. Chisquare analysis showed no stat istically significant difference between groups. Many participants had studied more than one language. Appendix B shows which languages and combinations of languages were studied by each participant. Table 8 Experience with instructed sec ond language learning other than German Presentation Format Static Animated Total % n % n % (n=44) Studied 1-2 semesters (n=23) (n=21) 2 No L2 instruction Spanish French Italian Latin Russian Greek 8 9 4 4 2 1 0 35 39 17 17 9 4 0 4 9 7 4 3 2 1 10 43 33 19 14 10 5 27 41 25 18 11 7 2 ------0.06 ns 1.49 ns 0.02 ns 0.34 ns 0.46 ns 1.12 ns Note. Percentages total more than 100% beca use some have studied more than one L2. *p<.05. ** p<.01. ***p<.001. "ns" = not sta tistically significant (p>.05). Participants were also asked about their pr ior experience living in countries other than Germany or the U.S. Three members of the static group and two from the animated group reported having lived in a foreign country othe r than Germany. No respondent mentioned any other German-speaking country, such as Switzerla nd, Austria, or Liechtenstein, nor was there mention of time spent in regions where German has a considerable presence, such as Belgium,
63Luxembourg, or Kazakhstan1. Of the five countries that we re mentioned in response to this survey item, Malta, Poland, Romania, Spain, a nd Taiwan, each had only a single response and, except for the respondents who had lived in Malta and Taiwan, all were natives of the countries in which they had lived, and fluent in the offici al languages of those land s. Chi-square analysis showed no statistically significant difference between groups. InstrumentationThis study included a pretest measuring mean ing, conjugation and word order of modal verbs. There were two instructional treatmen ts and then a posttest measuring modal verb conjugation and word order. The pretest differed from the posttest in two ways. It was shorter than the posttest in terms of poi nt totals (40 points possible for th e pretest, compared to 79 points for the posttest). The pretest also tested each cr iterion task in a manner that isolated one task from another, while the posttest fused the skills (conjugation and word orde r) into an integrated task. Content Validity of InstrumentsThe test measures, pretest and posttest, and the treatment programs, static and animated, were evaluated by a panel of five university Germ an instructors. Each instrument was modified by the researcher based on the panel's remarks, and then reevaluated by the same panel of teachers. To create the instruments, seven college level introductory German textbooks ( Deutsch Heute, Deutsch Na klar, Deutsch Plus, Kont akte, Neue Horizonte, Vorsprung, Wie Geht's ) were consulted along with at more than a dozen webs ites on German grammar, including the websites of the Goethe Institut and the AATG ( Appendix C ). Pretest The purpose of this measure was to determin e if equivalency of prior knowledge across the static and animation groups had been achie ved using matching and random assignments. The three areas associated with German modal verb s, their meanings, their irregular conjugations, and their effect on word order, we re tested in isolation from one another to prevent the test itself from offering instructive clues. The pretest had been pilot tested in four class sections of 1 There are 958,000 native speakers of German in Kazakhstan. An additional 100,000 are native speakers of Plattdietsch (www.ethnologue.com).
64 beginning German and the results were viewed by the German teachers of the review panel. The pretest included six matching items to test th e participants prior knowledge of modal verb meanings (MG). There were 24 fill-in-the-bla nk items corresponding to the conjugation task (CM), and 10 sentence-writing items for the word order task (WO) for a total of 40 items. Internal consistency reliabilit y (Cronbach alpha) was assessed fo r the overall pretest and for each of the three subtests. Given in Table 9, the Cronbach alphas fo r the overall scale were .86 and .46 for the meaning subtest, .87 for conjugation, and .77 for the word order task. Table 9 Internal Consistency Reliab ility (Cronbach Alpha) for Pretest Scores by Group (n=44) Variable # of Items Total Alpha Item to Total Animated Alpha Item to Total Static Alpha Item to Total Overall Test 40 .86 (-.09 to .67) .90 (.06 to .65) .82 (-.10 to .72) Meaning 6 .46 (-.06 to .41) .22 (.06 to .48) .63 (-.26 to .56) Conjugation 24 .87 (.16 to .61) .91 (.00 to .77) .79 (.00 to .68) Word Order 10 .77 (.30 to .61) .86 (.48 to .73) .60 (.01 to .53) The moderate value of the reliability coeffi cients spurred further examination into how individual items were functioni ng. Two items of the pretest ( Appendix D ) had negative item to total correlations. This negative value indicates that students who had gr eater overall levels of prior knowledge were more likely to mi ss these items. The two items were will (to want to) and mchte (to desire to). Details about these items are presented in the next section. "Will." For the six items concerning meaning, one item (A4 = will ) had a negative item to total correlation. This item had an answ er that contained a cognate of the 3rd person form of the modal verb wollen i.e., will The incorrect answer "to have a will to" was chosen for the meaning of wollen more often by participants who tende d to score above the average on the pretest. This is probably due to the interference of a va gue prior knowledge of modal verb conjugations along with an inaccurate memory for their meanings. Of the two incorrect answers possible for the item, "to have a will to" was chos en in 87% of the cases where the answer to A4 was judged incorrect. Only 16% of the par ticipants chose the correct meaning of wollen on the pretest. Given the negative value of this item, the scores for the meaning task were calculated
65 using only the other five items of this subtest. The Cronbach alpha with this item removed from the meaning subtest was .54. "Mchte." For the 24 items concerning conjugation, one item (A7 = mchte ) had a negative item to total correlation. Particip ants with a generally higher knowledge of modal verb conjugation were also more likely to pr ovide the conjugated form of the word mgen which is mag The correct answer for this test was mchte not mag Since the infinitive given was mchten participants with no prior knowledge could be expected to provide the correct answer based on the fact that its conjugation follows regular, known rules. The use of the word mag shows that the participant possess ed prior knowledge of the word mgen Unmentioned in the instructional treatments, mgen is the origin of mchten which is actually the subjunctive form of its parent word. The use of the word mag shows considerable prior knowledge, but it is incorrect based on the modal verb application. Gi ven the negative value of this item, the score for the conjugation task was calculated using on ly the other 23 items of this subtest. The Cronbach alpha with this item removed from the conjugation subtest was .87 (.8713 after item removed, .8677 before). Posttest In contrast to the pretest, the posttest test ed the two areas of c onjugation and word order in an integrated fashion. An indicative senten ce was given and the task was to rewrite the sentence with the addition of a given modal ver b. The English equivalent of the correct resulting sentence was given below each item. This format wa s consistent with items of the test banks for four of the above mentioned textbooks, includi ng the book used by study participants in their regular German classes. The posttest was pilo t tested by four university German teachers, in addition to the researcher, in el even class sections of beginni ng German over a three year time span. Items of the test were evaluated and modified after consultation among the teachers. With 79 total items, the posttest ( Appendix E ) was nearly twice as long as the pretest. It also differed from the pretest because the di screte skills of meaning, conjugation, and word order, were integrated into a unified task, name ly, to rewrite indicative sentences, adding a modal verb. In order to anal yze the posttest errors made by pa rticipants in th e pilot study, four university teachers of German looked at the e rrors and classified them using their own descriptions. Upon comparing the descriptions, it was determined that the conjugation task could
66 be further divided into three distinct error types: conjuga tion of the modal verb (CM), conjugation of the infinitive (CI), and conjugation of the separated prefix (CS). The word order task was also subdivided into two subtests: pl acement of the modal verb (WM), and placement of the infinitive (WI). The level of agreement among the four raters was 100%. The main study's posttest consisted of 18 i ndicative German sentences with blank lines beneath each one. Each sentence was followed by a German modal verb in parentheses. Below the blank lines were English sentences, repres enting the equivalents of the German sentences correctly rewritten with the a dditions of the modal verbs specified. The 18 sentences yielded 18 items each for the four subtests, CM, CI, WM, and WI. Seven of the sentences, containing a separable-prefix verb, comprised the subtest CS. Internal consistency reliability (Cronbach alpha) was assessed fo r the posttest overall scale (.90) and for each of the five subtests. Values and score ranges are shown in Table 10 for all participants and for each tr eatment group (animated and static). Table 10 Internal Consistency Re liability for Posttest Scores by Group (n=41) Variable # of Items Total Alpha N = 41 Item to Total Animated Alpha N = 21 Item to Total Static Alpha N = 20 Item to Total Overall Scale 44 .90 (.07 to .72) .87 (-.04 to .66) .92 (.00 to .81) Conjugation Modal (CM) 18 .80 (.03 to .50) .78 (-.09 to .67) .82 (.00 to .65) Conjugation Infinitive (CI) 18 .92 (.44 to .86) .88 (.00 to .88) .93 (.27 to .84) Conjugation Separable Prefix (CS) 7 .88 (.46 to .87) .88 (.42 to .83) .88 (.43 to .91) Word Order Modal (WM) 18 .66 (.00 to .81) .69 (.00 to 1.0) .53 (.00 to 1.0) Word Order Infinitive (WI) 18 .96 (.50 to .96) .47 (-.12 to .59) .53 (.61 to .98) Qualitative Instruments The pretest also posed questions concer ning the participants prior knowledge or experience with German in particular, and language learning in general ( Appendix F ). The
67 posttest invited participants to submit voluntary comments in writing, expressing their opinion of the presentation they viewed ( Appendix F ). Instructional Treatments Content The instructional materials were de veloped by the researcher, a university German instructor with over ten years of te aching experience for introductory German and a specialty of expertise in German grammar. The multimedia program was created as an animated presentation and later, a static version was adap ted from that. Both programs were revised after viewing by students and teachers of German, based on their questions and comments, some of which were recorded on audio tape. Remarks we re also given by reviewers who had no prior experience with German L2 learning. There were two instructional formats, static and animated presented on large screens in two separate classrooms. Both treatments were di vided into three sections based on each of the three task components: meanings of new words, conjugations of modal auxiliary verbs, and word order rules. The program PowerPoint was used to create all parts of the static treatment and two of the three sections of the animated treatmen t. The section concerning word order rules was created using the program Firestarter for the an imated version, because it allowed point-to-point animation, a feature described further in the next section. The treatment programs were judged by a pr ofessor of instruct ional technology with expertise in visual design, using the followi ng criteria suggested by Boling and Soo (2001, p. 443). 1. The interface and terminology are cons istent from screen to screen. 2. The layout of each screen makes good use of space. 3. Legibility and readability are high. 4. The display makes good use of contrast, repetition, alignment, and proximity. 5. Audio and video playback (where applicable) are of good quality. The animations of text in the experimental treatment were also judged by the reviewer to ensure that they adhered to th e following principles of purpose: 1. To attract, direct, or hold attention. 2. To illustrate the content of the text. 3. To convert a textual modality into a graphic modality.
68 Development The instructional presentations were first constructed as ancillary to regular course materials. The mechanics of word order rules were illustrated using animation software (Authorware 3.5) for multimedia presentation, and shown to classrooms on an overhead projector screen. The instruction was presented to nine class sect ions over a period of five years, but only once was an experiment conducted wi th a static-presentation control-group. The instruction was first developed in Authorware be cause of that programs ability to create path animations, which is the crux of the issue for il lustration of the motions of word order rules. Instruction for the other two task areas, meani ng and conjugation, was crea ted in PowerPoint for several reasons: it is easy to use, it has numerous animation features and it is readily available to most teachers. Authorware was then abandoned for the word order portion of the instruction, because it is not readily available. It is also very e xpensive and is quite complex to learn. A program created in Authorware would not be easy to replicate, except for a skilled programmer. PowerPoint, however, was not adequate to the job of animating the concept of word order motion. Its main limitation is that it can only anim ate an object for its entrance onto the computer screen. It cannot animate an objec t that is already visible on the screen. Therefore, nothing can move from one position on-screen to another. It can only move from limited directions offscreen, into a fixed position on-sc reen: Never from here to there, only from off to on. Since the crux of the word order rule, under focus he re, is that a certain object moves from one specified position to another specified position, h ere to there animation is expressly required. The solution was found at the 11th hour (A ugust, 2001), when the software company, Coffeecup, introduced its imitation of Macromedias Flash software, called Firestarter. The program is easy to get through the Internet, is very inexpensive ( $25), and is extr aordinarily easy to learn. More importantly, it enables the necessa ry here to there animation. It imitates Flash by creating .swf files for embedding into web-rea dy html files. The program can then be shown through the WWW or on a network with smooth a nd consistent presentation. The word order portion of the program, created for the animated ve rsion is consistent across platforms and easily accessible from any Internet terminal ( Appendix C ). Imagery cueing The static version of the instruct ional program utilized what Mayton (1990) has referred to as ima gery cueing strategy, similar to what ChanLin and Chan (1996)
69have termed metaphoric description. The visuals are static, but the viewer is cued to imagine the motions or changes that make the process dyn amic. The static modules created for this study employed this strategy by asking the learner, for example, to imagine the verb moving to the end of the sentence. This is also a design strategy intended to provide control between the static and the animated conditions. It provides a measure of fairn ess. It is only fair to assume that a teacher could induce metaphoric imagery to stim ulate dynamic mental modeling, without the aid of technology. Spatial conventions study To test the conventional expe ctations that influenced the design of the animations for this experime nt, a case study was performed. The animation sequence of the instructional program was segmen ted into five separate screens, one for each discrete motion or transformation. These also co rresponded to the five steps of the grammar transformation. The text of the sample senten ce was changed to scrambled letters forming nonsense words that were meant to convey no meaning whatsoever.Five members (one staff, one professor, one student, and two TAs) of a university foreign language department helped with the case study. The participants were asked to view each of the five animations plus a sequence that combined the five into a single animated transformation. For each of the five discrete animations, they were asked to (a) supply the verb which would describe the animation, and (b) choose the most a ppropriate descriptive verb from a printed list. For the combined sequence, they were asked to (c ) provide a verbal description of the event, and (d) rate the accuracy of a verb al description supplied by the re searcher. Participant responses were recorded on a mini-CD. Appendix G shows the results of this test, which confirmed the opinion of Winn and Solomon (1993) that "in the absence of semantic content, the interpretation of the diagrams was largely determined by syntactic rules of English" (p. 29). However, because the participants were involved in language learning in one form or another, each of them utilized grammatical metalanguage, which effectively expanded the c onvention from which they drew their imagery. While the "words" meant nothing, the string of letters was still considered to be a sentence by all viewers. Each string of letters was considered to be a word as delineated by the spaces between strings. In a sequence in which one word appear ed atop another word and then rose to slightly above it, the upper word was judged to be "conjuga ted" from the lower word, by all participants.
70 Since people who confidently use the word, conj ugate, generally know that it refers only to verbs, the same viewers would then know that the word in question is a verb, even without knowing its meaning. Participants also used the word, conjugate to describe the action that took place after the word string moved to the end of the sentence, when th e "verb" reverted to infinitive form. The purpose of this case study was to determine whether the animation design would be consistent with the vi ewers' conventional expectations. Since their descriptions of the overall event were similar to the explanation used by the researcher in th e narrated instruction, it was adduced that the animation design had the be st possible semantic connections to the verbal information. Narration A narrative script was written ( Appendix H ), based on recordings of three live presentations of the silent, inst ructional program. The recordings captured student questions and comments that were used to revise the progr am and to formulate the accompanying narration. The narration was written, performed and recorded by the researcher/progr ammer/content expert, with many of the sample sentences performed by a native speaker. Both PowerPoint and Firestarter accommoda te narration, but they do so using different file formats. PowerPoint uses .wav files, but Firestarter accepts only the .mp3 file format. A digital mini-disc was used to record the narrati on, and the software program, Total Recorder, was used to save the recorded sound files onto a computers hard drive. They were sampled at 16 bit mono and edited with a combination of Total R ecorder and the Windows internal sound recorder program. It took a third program, dbPowerAMP, to convert the .wav files into .mp3 files for use with the Firestarter program. These three progr ams are all readily available and are either freeware or inexpensive shareware. It may seem to have been a compli cated operation, but it is actually an easily replicable procedure, invo lving onhand or inexpens ive programs that are accessible to anyone. The digital mini-disc is not a necessary component, but it did allow spontaneous recording at locatio ns away from the computer. Pacing Each of the (30 total) screens of the presentati on was synchronized to the narration script and the pacing was adjusted a ccording to comments from several viewers ranging from thoroughly novice to experts of Germa n. With regard to speed, viewers universally suggested that too slow was bette r than too fast. After much adjustment, both versions of the program, as used in this study, ran precisely 20 minutes.
71 Content The content of instruction covers the topic of German modal verbs by distinguishing between the three discrete tasks involved in thei r use. To use a modal verb, one must first know its meaning, then how to conjug ate it, and third, where to put it and how that changes the word order. The content, created by an experienced German instructor, was revised to reflect student questions and comments, assemb led from the previous nine different classroom presentations. For the purpose of the study, the in structional programs were also viewed and scrutinized by experts in visu al design, foreign language inst ruction, instructional technology, and German language education, with revi sions made following each review session. Meaning Both treatments, presented the topic of modal verb meanings, utilizing animation to a minor extent. Since there are no moti ons or transitions that can be said to clearly illustrate or exemplify the meanings of modal verbs, there were no animations created for this study that could be said to prompt dual coding a ssociations. The meaning task was tested in the pretest, because it is a prime indication of prio r knowledge for modal verbs, but it was not part of the posttest criteria. Therefore, the two treatments were iden tical with respect to meaning. Conjugation For each modal verb, there are four possible conjugated forms. The singular first person subject, ich (I) and the third person, er, sie, es (he, she, it) have an irregularity in the conj ugation, including an irregular stemvowel change for four of the six modal verbs. Another feature unique to the conjugation of modal ve rbs, for first and third person singular, is the absence of verb endings. The ich -form regularly carries the ending e while the erform carries a t ending in regularly conjugated verbs. In the case of modal verbs, neither form carries an ending, with th e sole exception of the verb mchten, which carries an e ending not only on the ichform, but, atypically, on the er -form as well. The second person singular, du (you, informal) carries the regular ending st along with the same stem-vowel change undergone by the first and third person singula r subjects. All plural forms carry regular, that is, predictable, conjugated forms, as does the formal second person singular, Sie (you, formal). Therefore, modal verb conjugation involves four possible fo rms distributed over six subject pronouns. Figure 3 represents the final sc reen of the conjugation porti on of the instructional module. The shadowed box highlights the irregular stem-vowel changes undergone by four of the modals in the singular, informal forms. The pretest measured prior knowledge on this variable by
72 Figure 3. Modal verb chart showing patt erns and irregular conjugations.requesting the participant to pr ovide conjugations of the four irregular modals for the six pronouns: ich, du, er, wir, ihr, and sie. Word order Word order is the task-type under the sharpest focus in this study, because it is the aspect for which acquisition is most likel y to be influenced by the use of animation in instruction. A pretest item (section C), that re flects word order only, consisted of jumbled sentences which must be ordered by writing each word on an appropriate blank in a punctuated row next to the jumble. Data Collection Procedures Three class sections of Beginning German I (G ER 1120) participated in this experiment. Members of each class were ranked within section based on their scores from four prior chapter exams. The top two students of the rankings were pa rtnered into a pair, the next two into another pair, and so on. For each pair, a coin was tossed and the first member was assigned a "1" or a "2" (representing treatment groups) de pending on the result of the co in toss. The other member of each pair was assigned to the opposite group. Each group was then assigned by another coin toss, to one of the two treatments to receive either animated or static presentation. Participants were each given a letter introducing the researcher, explaining the basi s of the study, and inviting them to be voluntary participan ts in the experiment ( Appendix A). This letter was also read aloud by the researcher, during each section's regular class hour. ich du er mchte mchtest mchte mchten mchtet mchten wir ihr sie will willst will wollen wollt wollen kann kannst kann knnen knnt knnen darf darfst darf drfen drft drfen muss musst muss mssen msst mssen soll sollst soll sollen sollt sollen
73 The two 50-minute daytime sections were give n the letter and the pretest in one class session, and their treatment and posttest on the following day. The 2.5 hour evening class received the letter, pretest, treatments and posttest during a single class session. The two daytime classes took their pretests in their regular classroom and were then separated for treatment and posttest. (The experi mental group was taken to the classroom directly above the regular classroom. Both classrooms then had equal amounts of disturbance and distraction from the building construction occurring during this time.) Participants in the evening class were separated before their pretest, wh ich they took in the same classrooms where they received treatments and posttests. The same tw o classrooms were used for all three sections. The first third of the presentations was iden tical for both treatment groups. It concerned the meanings of modal verbs and utilized va rying amounts of animation, distributed among the six modal verbs. The second and third parts were animated fo r the experimental group, but presented only through static screens for the cont rol group. Static means that all of the text presented on a single screen appears at one time, without transitiona ry embellishment. Transitions between screens were subtle and identical for both vers ions. The presentations ran for 20 minutes. Participants in each group were then given a posttest, which they completed immediately following the treatment programs. All participants completed the tests within the given timeperiods. Scoring Each test was scored by two raters using scoresheets, and no marks were made on any of the test papers. Instead, ID code and scores for each participant were recorded on a separate form ( Appendix I ). The raters did not have a view of each others marks while scoring. Discrepancies in rater scores were first checke d for miscalculations. With 89 tests rated, there were originally 33 discrepancies. Examination of these discrepancies revealed that all but five of them were judgement errors (e.g., marking a corr ect answer "wrong") or errors of calculation (e.g., incorrect totals or subtotal s) on the part of the raters. Th ese 28 discrepancies were easily resolved, and the remaining five were rated by a third-party German teacher in the presence of the first two raters as well as a fourth-party German teacher. The five discrepancies were all found to be a matter of spelling, inconsequential to the type of error being scrutinized. For example, kaufft was judged by rater #1 to be corre ct because the answer possessed the necessary
74 components of initial consonant, proper stem-vowel and the crucial t ending. However, because there are two f's instead of the proper spelling of only one f, the answer was judged incorrect by rater #2. Rater #3 (as well as the f ourth-party teacher) agreed with rater #1 in all but one of the five cases. There was also one issue of handwri ting. For the test in qu estion, the answers that rater #1 marked as wrong were thought to be co rrect by rater #2 who recognized the handwriting and knew the participant's habit of writing o's that look like a's. With a value of 1 for correct or 0 for incorrect, scores for each test item were entered first into an Excel spreadsheet. These values were then merged into an MS Word document, and then imported into the program, SPSS, for statistical an alysis. The next chapter presents the results of this analysis.
75 Chapter 4 Results This goal of this study was to investigat e the effects of animation for a technology assisted German grammar presen tation on students' understanding of modal verbs. The premise was that the features of anima tion effects (fly-ons, highlighting, etc.) would illustrate intangible concepts of grammar better than static textua l representations (sampl e sentences) if those concepts involved motion or a change of state. The resear ch hypothesis was that students' learning of the German language grammar concepts would be superior using animated effects and that their scores on a post test concerning the conjugations and word order of modal verbs would be higher for the animated presentation type. The first step of the data analysis was to determine if there was a difference between the two groups for their total pretest scores, as well as for the three subscale variables: meaning, conjugation and word order. The three subscales of the pretest were moderately correlated. The correlations are given in Table 11. Table 11 Pearson Product Moment Correlations of Pretest Subscales (n = 40) 1 2 3 1 Meaning 1.00 2 Conjugation .38* 1.00 3 Word Order .47** .31 1.0 Note. *p<.05. ** p<.01. ***p<.001. The general inquiry was divided into two questions. The first was whether animation could bring about higher test sc ores overall. Since the particip ants were brought together from three different class sections, their scores were also viewed by section as an independent variable. The overall posttest sc ore was analyzed using a 3 ( section ) x 2 ( method ) univariate analysis of variance (ANOVA). Table 12 provides de scriptive statistics for the total pretest scores by section and method.
76 Table 12 Descriptive Statistics for Pretest To tal Scores by Section and by Method Presentation Type Section Animated (n = 19) Static (n = 21) Effect Size n = 6 n = 7 1 M 56.58 57.52 -0.05 SD 18.44 17.29 Median 53.95 52.63 Range 36.84 to 84.21 39.47 to 92.11 Skewness 0.53 1.51 Kurtosis -1.01 2.76 n = 7 n = 7 2 M 48.87 48.87 0.00 SD 13.15 18.78 Median 50.00 52.63 Range 31.58 to 73.68 28.95 to 81.58 Skewness 0.95 0.70 Kurtosis 2.05 0.02 n = 6 n = 7 3 M 53.50 50.38 0.16 SD 26.97 9.04 Median 53.95 47.37 Range 23.68 to 94.74 39.47 to 65.79 Skewness 0.41 0.76 Kurtosis -0.59 -0.11 Note. Scores represent percentage correct. The second question was concerned with the sp ecific tasks involved. Posttest scores were analyzed using a 3( section ) x 2 ( method ) multivariate analysis of variance (MANOVA) because the main task, rewriting German sentences with the addition of a modal verb, could be divided into five interrelated subtasks. The conjugation task was subdi vided into the variables (1) conjugation of the modal verb (CM), (2) conjugation of the dependent infinitive (CI), and (3) conjugation of the separ able-prefix infinitive (CS); the word order task was subdivided into the variables (4) word order of the modal verb (WM), and (5) word order of the dependent infinitive (WI).
77Therefore, MANOVA was selected because it exam ines each variable as part of a system. MANOVA can assess whether there is a signif icant overall difference between the two group mean scores, and it offers a structured method to specify the comparisons of group differences for each subtask (dependent variables). In ad dition, MANOVA is "robust to modest violations of normality if the violation is created by skewness rather than by outliers" (Tabachnik & Fidell, 1989, p. 378) and the "SPSS MANOVA adjusts for unequal n (p. 404). Research Question 1The first research question was: What is th e effect of animation of sample sentences depicting a grammatical process, explained in a concurrently narrated presentation, on test scores measuring a language learners ability to rewr ite German indicative sentences adding modal auxiliary verbs, compared to static pr esentation of the same sample sentences? Posttest scores were analyzed using a 3 (section ) x 2 ( method) ANOVA. Results indicated no statistical significance for a section by method effect ( p. = .48), method effect (p. = .19), or section effect (p. = .40). Table 13 provides descriptive statistics fo r the three sections by treatment group. Participants in all three class sections had hi gh scores with means ranging from 77% to 87%. Section 1 had a larger range of scores than the other two sections, with a low score of 34, and a high of 90. However, the median score for Section 1 was in the 80s (81.01) as were the medians for the other two sections for both treatment groups. With the small sample size (9) for this cell (see Table 2), the mean wa s greatly influenced by the outlier.
78 Table 13 Descriptive Statistics for Post test Total Scores by Section Presentation Type Section Animated (n = 21) Static (n = 20) Effect Size 1 M 87.13 74.86 0.84 SD 7.52 19.22 Median 86.08 81.01 Range 76-99 34-90 Skewness 0.15 -1.99 Kurtosis 1.17 4.30 2 M 84.95 83.97 0.08 SD 10.33 13.29 Median 86.08 87.34 Range 63-98 60-99 Skewness -1.02 -1.43 Kurtosis 1.63 2.90 3 M 81.86 79.02 0.25 SD 12.88 10.04 Median 83.54 83.54 Range 58-98 63-91 Skewness -1.31 -0.55 Kurtosis 3.21 -1.04 Note. Scores represent percentage correct. Effect Sizes Table 13 also provides effect sizes for th e differences in groups across sections. Effect sizes were calculated (using Cohen's d = M1 M2 / SD pooled) in order to measure the slight advantage shown by the scores of the animated treatment group. According to Cohen (1988), an effect size of 0.20 indicates a small effect, 0.50 indicates a medium effect, and 0.80 indicates a large effect for the treatment intervention with the higher mean score. The effect size was large for the difference between the animated and the static groups in Section 1 (0.84). However, it should be noted that the mean for the static group was pulled down by the extremely low score of 34%.
79Research Question 2The second research question was: Is the effect of animating the sample sentences different for each of the two criterion tasks (c onjugation and word order) involved in the rewriting process? A 3 ( section ) x 2 ( method ) MANOVA was conducted with the conjugation tasks as the dependent variable s, subdivided into CM, CI, and CS. The correlations for these variables are given in Table 14.Table 14Pearson Product Moment Correlations of Posttest Subscales (n = 41) 1 2 3 4 5 1 Conjugation Modal (CM) 1.00 2 Conjugation Infinitive (CI) .38* 1.00 3 Conjugation Separable Prefix (CS) .19 .35* 1.00 4 Word Order Modal (WM) .27 .40** .22 1.00 5 Word Order Infinitive (WI) .15 62*** .19 .11 1.00 Note. *p<.05. ** p<.01. ***p<.001. One assumption is that the covariances of these three dependent variables are equal across groups. Box's M -test indicated no statistically significant differences (Box's M = 46.674, p = .221). Analysis of the means revealed no statistical significance for the section x method interaction (Wilks Lambda = .928, p = .863), or the method effect (Wilks Lambda = .928, p = .476), or the section effect (Wilks Lambda = .622, p =.13). A 3 ( section ) x 2 ( method) MANOVA was also conducted with the word order tasks as the dependent variables, subdivided into WM and WI. The correlations for these two variables was .109 (Table 14). Box's M -test indicated no statistically significant differences in the covariance matrices across groups (Box's M = 5.939, p = .951). Analysis of the means revealed no statistical significance for section x method interaction (Wilks Lambda = .915, p = .544), method effect (Wilks Lambda = .958, p = .483), or section effect (Wilks Lambda = .883, p =0.366). The posttest mean scores, standard devi ations, medians, and score ranges are given in Table 15.
80 Table 15 Mean Achievement Posttest Scores of Treatment Groups (n=41) Variable Total n = 41 Animated n = 21 Static n = 20 Effect Size Total Score M 81.94 84.70 79.05 0.45 SD 12.60 10.13 14.46 Median 84.81 86.08 83.54 Range 34.18 to 98.73 58.23 to 98.73 34.18 to 98.73 Skewness -1.67 -1.09 -1.72 Kurtosis 4.16 1.76 3.90 Conjugation of M 64.50 67.50 61.40 0.28 the Modal SD 21.40 20.36 22.55 Verb Median 61.11 66.67 61.11 (CM) Range 11.11 to 100.00 27.80 to 94.44 11.11 to 100.00 Skewness -0.46 -0.48 -0.42 Kurtosis -0.18 -0.42 0.11 Conjugation of M 82.50 88.70 76.11 0.55 the Dependent SD 23.50 17.00 27.90 Infinitive Verb Median 88.89 94.44 88.89 (CI) Range 0.00 to 100.00 22.22 to 100.00 0.00 to 100.00 Skewness -2.23 -3.33 -1.71 Kurtosis 4.67 12.53 2.40 Conjugation of M 46.00 50.00 42.14 0.20 the Dependent SD 38.40 39.02 38.40 Infinitive Verb Median 42.86 42.86 35.71 with a Separable Range 0.00 to 100.00 0.00 to 100.00 0.00 to 100.00 Prefix. Skewness 0 .05 -0.04 0.15 (CS) Kurtosis -1.65 -1.62 -1.82 Word Order of M 99.00 98.41 99.44 -0.26 the Modal Verb SD 4.00 5.01 2.50 (WM) Median 100.00 100.00 100.00 Range 83.33 to 100.00 83.33 to 100.00 88.90 to 100.00 Skewness -3.60 -2.98 -4.50 Kurtosis 11.70 7.60 20.00 Word Order of M 95.80 98.00 93.61 0.28 the Dependent SD 15.90 4.50 22.40 Infinitive Verb Median 100.00 100.00 100.00 (WI) Range 0.00-100.00 83.33 to 100.00 0.00 to 100.00 Skewness -5.80 -2.34 -4.30 Kurtosis 35.10 5.30 18.60 Note. Effect sizes: .2 = small, .5 = medium, .8 = large (Cohen, 1988). Scores represent percentage correct out of a 79 point test.
81 There were no significant differences between the two groups for scores on any of the subtest variables. The range of scores for each subtest reached a maximum of 100% and a minimum of 0% for three of the subtests: CI, CS and WI. Table 12 also shows the value for skewness, as a measure of how far the particip ants' scores depart from a normal distribution. Having a negative skewness means that the sc ores were "bunched up" toward the higher end, closer to 100% correct. The scores of the animated group had a larger negative skewness than those of the static group, in all cases except one. These four subtests showed a slight advantage for the animated treatment. Only CS, which had skewne ss for the total sample of .05, showed an approximately normal distribution for both groups. Effect Sizes The four subtests, with larger mean scores in the animated group, had effect sizes ranging from small to medium (0.20 to 0.55). The fifth subtest, WM, showed a small effect size (-0.26) favoring the static group. This was also the subtest which showed the highest kurt osis (20.0) and the greatest negative skewness (-4.5). Standard deviations for both groups were considerably higher for the conjugation tasks than for the word or der tasks. The effect size for total score was 0.45 favoring the animated treatment. Conjugation Across the three subtests of the conjugation task animation was shown to bring about slight but non-significant advantages. For subtests CM and CS, the effect sizes were small, but a medium effect size (.55) was found for the CI subtest. This subtest also showed considerably higher kurtosis for the animated group (12.5 compared to 2.4), nearly twice as much negative skewness (-3.33 compared to 1.71) and only half as much variability ( SD = 17.0 compared to SD = 27.9). Word order For both subtests of the word order task (WM and WI), the scores were very high in both groups. The animated group had a mean score of 98.0 ( SD = 4.5) and the mean score of the static group was 93.6 ( SD = 22.4). There was very little variability in word order scores overall for the animated group. Their scores for both subtests ranged from 83.3 to 100.0. The static group had even lower variability for the WM subtest of the word order task with scores ranging from 88.9 to 100.0, but the WI subtest showed high variability ( SD = 22.4) especially when compared to the animated group mean score for WI ( SD = 4.5). The difference in variability between the two groups for WI was the largest of any subtest with a 17.9 point
82 difference in standard deviations The range of scores for the static group indicated that one participant scored 0.0 for this subt est, while no participant from the animated group scored lower than 83.3 on either of the word order subtests. Anal ysis of these subtests revealed no statistically significant differences. Values for WM univa riate analysis for method effects were F (1,39) = 0.69, p = .41, and for WI, F (1,39) = 0.73, p = .40. There was no statistically sign ificant effect for treatment (method), or for section x method interaction. A statistical significance was found for the effect of section for the subtask of CS only. A complete listing of the F -ratios and probability levels from the series of 3 ( section ) x 2 ( method ) ANOVAs for each of the five dependent variables is presented in Table 16. Table 16 F-values and Significance Levels for Series of 3 ( Section) x 2 (Method) Posttest ANOVAs Variable Section x Method p Method p Section p Total Score 0.750 .479 1.780 .191 0.405 .670 Conjugation 0.365 .697 2.065 .160 0.418 .662 CM 0.713 .497 1.028 .318 1.478 .242 CI 0.347 .709 2.511 .122 0.767 .472 CS 0.098 .907 0.193 .663 4.775* .015 Word Order 1.603 .216 0.322 .574 0.631 .538 WM 0.178 .838 0.618 .437 0.930 .404 WI 1.576 .221 0.663 .421 1.064 .356 Note. *p<.05. ** p<.01. ***p<.001. Posttest Survey Appendix J shows the remarks that participants offered on the posttest survey when invited to add their comments about the presen tation. An equal number of participants in each group (9) declined to submit comments. The comments were collected and randomized by a computer macro in MS Word ( Appendix K ). They were shown to a group of 5 people (staff, TAs and professors) who were asked to rate each comment according to a 5-point Likert-type scale
83 with values ranging from very negative (1) to ve ry positive (5). Interrater reliability was assessed at .95 (intraclass correlation coefficient). The mean rating for the comments of the animated group was 4.7 ( SD = 0.36) compared to 3.7 ( SD = 0.96) for the comments of the static group. A ttest showed this difference to be significant at the .05 level ( p = .0031) with a medium effect size. The means and standard deviations are given in Table 17. Table 17 Mean Ratings of Participant Comments (n=23) Group n Mean SD Static 11 3.69 0.96 Animated 12 4.68 0.36 Note. Effect size = 0.56 This difference in the opinions of the two groups toward the inst ructional presentation may be due to the novelty of the animation, although neither group had been receiving German instruction via computer presentation of any kind prior to the study. Therefore, both presentations were new to the participants. Nevertheless, the animated version elicited more positive comments than the static version. The effect size for this difference was calculated at 0.56, which indicates a medium effect, according to Cohen (1988). The next chapter presents a discussion of th e results, including the limitations of the research, the implications for German language pedagogy, and recommendations for further study.
84 Chapter 5 Discussion This study investigated the effects of animation for a technology assisted German grammar presentation on students' understanding of German modal verbs. The premise was that many intangible concepts of dynamic grammar i nvolve syntactic components that possess visuo spatial characteristics. It was further specul ated that these characteristics could be more effectively represented by animated versus st atic instructional presentations. The research hypothesis was that students woul d learn the foreign language grammar concepts better with animation. The value of the animation was empi rically assessed in or der to determine its effectiveness within the instructional design. This study began with the animated version of the instruction. Throughout seven years of teaching German I and II, the researcher had created over 20 animated grammar presentations using PowerPoint, Flash, and Authorware softwa re programs (Caplan, 1996). The intrinsic value alone made the utility and pedagogic value of the programs clear, but the ultimate benefit to learner achievement was not as obvious. The presen tations aided the clarit y of the instruction in several ways: (a) Everything was al ways legible, (b) the presenter never had to turn her back to the learners, (c) no details were forgotten or erroneously presente d, (d) class time was well-spent and efficiently allotted, (e) illustrations we re of high quality, in vibrant hues, and most importantly (f) 100% of the learners appeared to be paying attenti on to the presentation. The last claim lay in the observations of the researcher. Because the presentations required the lights to be off, the participants ha d little opportunity for r eading or other activities at their desks. Bathed in the light reflected from the projection screen, it could be seen that all heads were raised forward and all eyes were focused on the presentation. The abovementioned considerations spoke in favor of the use of multimedia-enhanced instruction. This study was an attempt to determine if learning benefits coul d also be attributed to such enhancement.
85 Limitations Sample Size The greatest limitation to this study was the sm all number of participants. Even with the procedure of matching of part icipants by their rank, the sample was too small to provide adequate power to detect a si gnificant difference of treatment effect. Therefore, it is not conclusive from this study that animation ha s no pedagogical value because a type-II error may have resulted from the low sample size. Content Area This research did not have the goal of finding evidence for the positive effect of animation in general. The aim was to discover whet her animation was beneficial to illustrate the particular feature of German word order in the use of modal verbs. This is a very specific content area, and, as such, the outcome does not exte nd very far beyond the realm of explicit German grammar instruction. Even the area of modal verb word order was limited in that only the present tense indicative was considered. Word order for questions, depende nt clauses, and the perfect tense forms would require additional explanation. However, had the results of this study shown significant benefit for animation's effect on learne rs' understanding of word order, then it would have suggested that animation may also provide be nefits in the case of other word order issues. Ceiling Effect The lack of statistically significant effects of animation may be due to the narrow scope of the task and the ceiling effect of the test items. The content instructed and the tasks that were tested were very elementary and may have been too easy to allow for a measurable effect of treatment. Since all members of both tr eatment groups had scores above 90% for word order it can be concluded that animation was no detriment to learning for this subject matter, but it is inconclusive as to whether anima tion is necessary or significantly beneficial. Student scores did not suffer from the use of animation, and it is re asonable to believe that it would support learning of more complex tasks involving conjugation and wo rd order, especially with modal verbs. Both the impetus and the design of this experiment began with the animated version, which, in the developmental pilot studies, proved to be very effective over standard textbook explanations and oral teacher-presentation for the content of German modal verb word order rules. A developmental pilot study utilized a pretest-postt est design comparing the animated and static
86word order presentations showed considerable variability in posttes t scores favoring the animated presentation. Pretest scores for both groups ranged from 0-5. For the animated group, posttest scores increased to a range of 5-9, while the static presentation group posttest scores, maintained the range of 0-5. The results of the pi lot study formed the impetus to research further the area of animation for German modal verb word order and to investigate whether it would also have a beneficial effect for the aspect of modal verb conjugation. The static version of the instruction was desi gned to mock the animated version in every way except for the motion of elements. In the pr ocess of development and review, both versions were modified to take maximum advantage of visu al design principles and to follow what current research has suggested about educational multimed ia and instructional design. The static version of the presentation produced posttes t scores that were so high, that there was little room for the animation condition to show and advantage. Multimedia Principle An explanation of why the two treatments produced such equally high results was that both treatments adhered to the main tenets of the design principles of the cognitive theory of multimedia instruction. The theory's multimedia principle concerns the use of illustrations as accompaniment to textual information. Learners ach ieve deeper learning from instructional texts when words are accompanied by relevant pictures The addition of a "pictoral explanation (i.e., animation) [... substantially suppo rts] learners' problem-solving tr ansfer performance" (Mayer & Moreno, 2002, p.93). Typically, this means that words are accompanied by pictures. The instruction, or "textual explanation," of each presentation was delivered on an audio channel through recorded narration, while th e visual channel offered an enti rely different text. Instead of merely reading what was already heard, the view ers were presented sample sentences, meant to illustrate the explanation. The static version also employed the multimedia principle because the audio presentation was different from, but s upported by, the visual presentation, even though both were textual in format. The animated ve rsion added motion to the presentation of the samples, but it did not appear to further the multimedia principle of dual-channeled input. Both treatments offered a verbal explanation in an auditory channel while presenting a visual sample supportive of the explanation.
87The coherence principle Both instructional treatments followed the coherence principle, which precludes the use of extraneous words, sounds, music, and other decoration, in order to minimize cognitive overload. Redundant visual text and artistic features, such as colors, fonts, geometric shapes and gratuitous attention grab bers were systematically excluded from both programs to create maximum coherence between the visual sample sentences and the auditory explanations. Although it served to illustrate specific aspects of the subject matter, the animation in the experimental treatment c ould not be said to add coheren ce to any detectably significant advantage. Redundancy principle. The redundancy principle states that "less is more." Redundant visual text may add an unnecessary burden to th e visual processing mechanism. It hinders rather than helps the viewer to integr ate the illustration with the expl anation. The textual explanations were given in verbal form only, while the visual text was comprised of sample sentences, which were different, but illustrative of the narra ted text. Here again, the animations of the experimental treatment did not add anything to strengthen the presenta tion in this regard. Only the animated version can be said to involve dual coding, because the motion of textual objects creates a non-verbal image. Such an image can be stored in the non-verbal memory and used as a referential link to the accompanying verbal information. The static version utilized only a textual format, st oring all of the information in th e verbally represented system of memory. Associative connections are possible be tween the items of information, but no ties are created that refer to items in non-verbal memory. However, because it used both audio and visual modalities, the static version can be said to involve dual channe ling. Since the linguistic subject matter primarily resides in the mental world of verbal representation, it may be that dual channeling of input is of primary importance in cognitive processing for language learning. Dual coding, in this experiment, did not produce any appreciable bene fit beyond what dual channeling could provide. In that respect, animation could not be said to be of a value proportionate to the trouble and expense that it orig inally instigated. However, si nce the costs and the efforts associated with creating animated presentations have decreased, and the learners have expressed increased enjoyment for animated instruction, the intrinsic values appear to at least equal the investments.
88 Implications For teaching, the implications of the results are that verbal instruction of some grammar principles can be very effective if presented in dual channeled format, i.e., more than one avenue, or modality, of input. This conclusion, in and of itself, does not promote the use of multimedia for grammar instruction because it could be clai med that an instructor's verbal explanation, accompanied by the textbook's visual modality, is already a combination of input avenues. Therefore, it does not take a co mputer to offer dual channeling. It can also be argued that sample sentences, written in a textbook or on a bl ackboard, offer non-redundant illustration of information, another principle of the cognitive theory of multimedia. Since multimedia, strictly speaking, only declares multimodality, e.g., hearing and seeing, it does not have technology as its prerequisite. Only animation of visual stimuli requires the use of computer (or film) technology. Since this experiment found no si gnificant advantage for animation in this particular arena, it could be claimed that traditional teaching met hods (teacher and text/blackboard) would be as equally effective as a computerized presentation. This study did not compare traditional te aching methods (e.g., reading the textbook aloud in class) to the use of compute r-presented instruction, so no clai m can be made that technology is a superior teaching tool to trad itional teaching methods. However, the universally high scores of participants in both treatment groups suggest that such presenta tion is certainly not harmful to learners' ability to process the material for immediate use, and the higher appreciation that learners expressed for the animated version lend merit to its use in German language instruction. Perhaps it is less a matter of how the learne r perceives a lesson than how the instructor prepares it. Both treatments of this study invo lved preparation time a nd design development well beyond what is customary for tradi tional preparation. This research suggests that, given a similar amount of time, planning, evaluati on and revision, as well as the s upport of theoretical research, and the commitment of the instructor, many different teaching methods could provide significant benefits to learners. The benefits of computer-assisted multimedia, therefore, are more obvious for the instructor, than they are for the learner. The planning of the presentation means thorough planning of the lesson. It may mean increased visibility, legibility, a nd clarity of textual explanations. It may involve superior quality of illustrations, better anticipation of learners'
89questions, and more efficient use of time. Ca reful development can mean better preparation overall. The conclusion, then, is that quality pr eparation deserves the credit for effective presentation, rather than any singular aspect of presentation. Summing up current research on instructional multimedia, Moore, Burton, and Myers (2001) conclude that: [L]earning may be unaffected by a partic ular media and that learning of any type can be achieved through a variety of paths (media) if the methods of providing information are well designed, have a theoretical base and are well executed. (p. 869).RecommendationsThe sample size used for this study was too sma ll to detect a statistically significant effect for animation in relation to any of the tasks asso ciated with the use of German modal verbs. It was anticipated that animation would benefit le arners' understanding of German word order the most, and of modal verb conjugations at a s econdary level. While there was not enough of a difference, statistically, to sugg est an advantage, a small differe nce in the scores of the two groups for the conjugation task does suggest that further study into the effects of animation for conjugation alone should be undertaken with a consid erably larger sample size. It is, therefore, recommendable to study the use of animated presentation for basic conjugation of German regular verbs. Instruction in regular verb conjugation typically takes place at the start of introductory level German courses. At such a point, the number of participants is not yet affected by dropout attrition, and the prior knowledge of participants is ge nerally more equal, since study habits and other factors will not yet have come into play. Additionally, it is at the outset of German language learning that clarity of conjugati on instruction would be of the most benefit to learners, in order to construc t a knowledge base upon which to build new information (such as new verbs, irregular conjugations, and general rules of word order).It would also be useful to study the use of animation for instruction of more complex aspects of German word order. The tasks of this study were elementa ry, but the potential of animation to depict transformation and motion c ould be used advantage ously for higher level German word order issues such as: (a) Questi ons and commands (the modal comes first), (b) Modals in dependent clauses (the modal goes to the end), (c) compound-complex sentences (the modal goes to the end of the clau se, not the end of the sentence) (d) modals in the past and conditional tenses, (e) modals in double infini tive constructions, and (d) modals in double
90 infinitive dependent clauses (the modal "jumps the line" to the place in front of the double infinitive). Animation may offer instructional de signers a useful tool for illustrating some of these structures, which often present the in structor with difficulty in lesson planning3. The high level of enjoyment reported by the animation group suggests that it has high intrinsic value. Such value may be exploited by instructors, as long as the design of the animation incorporates associations that are semantically relevant to the content. This requirement calls for further study into these relationships. It would be very productive to determine how specific elementals of animation (individual motions and changes) are interpreted by viewers in relation to linguistic elements. The importance of such a step should not be underestimated in the design of educational animation research. For example, a study involving two different treatments of inst ruction, both animated, could rev eal benefits to learning for specific forms of animation relative to specific grammatical features (e.g., comparison of "fadeins" with "zoom-ins" in relation to conjugation principles). Animation, therefore, should not be viewed as a monolithic entity, but as a diverse c onstruct of great variation, which calls for closer investigation of its speci fic semantic properties. Computer-assisted multimedia presentation has many benefits to the teacher in terms of lesson preparation, and it bears no apparent detr iment to the learners' ability to process the instruction. Learners seemed to appreciate the effort and en joy the instructional format. However, it is unproductive to simply duplicate traditional textbook inst ruction in multimedia form, especially when the computer offers aspects that the textbook is unable to deliver. Animation is now readily available to the instruc tional designer, often to the average teacher, but the research attesting to how it should be used is not. Another important aspect of multimedia, not available in a text book, is interactivity. Instructional designers ha ve the ability to create software that not only takes advantage of multimedia features, such as animation, but can al so be responsive to the learner's own inquiries and commands. The original software program cr eated by the researcher, which served as the impetus to this study, had user-control and inte ractivity as overriding properties. The animations 3 See message thread Past Subjunctive, begun May 24, 2002, posted to the AATG (American Association of Teacher of German) electronic mailing list, archived at http://listserv.iupui.e du/archives/aatg.html
91 used were the same as those used in this study, but they occurred in re sponse to the learner's actions. The learner was required to drag the modal verb into the sentence at the correct placement, in order for the sentence to transform into proper configuration. This provided a form of practice along with immediate feedback, in that the animations could only be performed given the correct placement of the modal verb. The ac tions could also be re peated or reversed, in response to the learner's mouse commands, whic h provided a measure of user-control to the instruction. In order to design multimedia instruction for German grammar, these issues of task complexity and interactivity need further study. It should be determined how animation works for other specific structures and tasks, how it wo rks in concert with user-interactivity and usercontrol, and to what extent viewer appreci ation of animation can affect learning. While interactive, animated programming has great pe dagogical potential, it is not possible to speak about animation as an asset in itself. It is an impor tant element, but it is also part of a system of features that multimedia affords. The use of animation requires and deserves further study.
92 References Abel, R. R., & Kulhavy, R. W. (1986). Maps, mode of text presentation, and children's prose learning. American Educational Research Journal, 23 (2), 263-274. Adair-Hauck, B., Donato, R., & Cumo, P. (1994). Using a whole language approach to teaching grammar. In L. Shrum & W. Glisan (Eds.), Teacher's Handbook" Contextualized language instruction (p. 90-105). Boston: Heinle. Anglin, G. J., Towers, R. L., & Levie, W. H. (2001). Visual messa ge design and learning: The role of static and dynamic illustrations In D. H. Jonassen (Ed.), Handbook of research for educational communications and technology, 26, 755-794. Retrieved March 14, 2002, from http://www.aect.org/Intran et/Publications/edtech/ Asher, J. J. (1969). The Total physical response approach to second language learning. Modern Language Journal, 53 (1), 3-17. Asher, J. J. (1972). Children's first language as a model for second language learning. Modern Language Journal, 56, 3. Asoodeh, M. M. (1993). Static visuals vs. com puter animation used in the development of spatial visualization. Texas A&M University. Dissertation Abstract s International, 54 11A, (University Microfilms No. AAG94-10704). Aust, R., Kelly, M. J., & Roby, W. B. (1993). The use of hyper-reference and conventional dictionaries. Educational Technology, Research and Development, 41, (4). Boling, E., & Soo, K. S. (2001). CALL issues: Designing CALL software. In: J. Egbert & E. Hanson-Smith. (Eds ), CALL environments: Researc h, practice and cr itical issues (pp.442458). Alexandria, VA: TESOL. Broadbent, D. E. (1958). Perception and communication London: Pergamon. Byrne, M., Catrambone, R., & Stasko, J. (1999). Evaluating animations as student aids in learning computer algorithms [Online] Available: http://citeseer.nj.nec.co m/byrne99evaluating.html Byrne, M., Catrambone, R., & Stasko, J. (1996). Do algorithm animations aid learning? [Online] Available: http://citeseer.nj.nec.com/byrne96do.html
93 Caplan, E. (1996). Animated German Gr ammar Tutorials [Computer software]. Available: http://www.webgerman.com/Presentations/ Carter, R., & McCarthy, M. (1994). Grammar and the spoken language Paper presented at the Annual Meeting of the Teachers of Englis h to Speakers of Other Languages. Baltimore, MD. (ERIC Document Reproduction Service No. ED 373 539). Chanlin, L. (1998a). Students' cognitive styl es and the need of visual control in animation. Journal of Educational Computing Research, 19 (4), 353-65. Chanlin, L. (1998b). Animation to teach students of different knowledge levels. Journal of Instructional Psychology, 25 (3), 166-175.. ChanLin, L., & Chan, K. (1996). Computer graphics and met aphorical elaboration for learning science concepts. (ERIC Document Reproduction Service No. ED 329 390). Chapelle, C. A. (1998). Multimedia CALL: Lessons to be learned from research on instructed SLA. Language Learning and Technology, 2 (1), 22-34. Chapelle, C. A. (2001). Computer applications in second language acquisition: Foundations for teaching, testing and research. Cambridge: CUP. Chen, L. (1995). Effects of static graphics animated graphics, and digital video on students' performance and attitude in computer-a ssisted instruction. Univ ersity of Texas-Austin Dissertation Abstracts International, 56 10A, (University Microfilms No. AAI96-03816). Childress, M. D. (1995). Effects of three mu ltimedia instructional presentation formats containing animation and narration on recall and problem-solving performance. Virginia Polytechnic Institute and State University Dissertation Abstra cts International, 56 09A, (University Microfilms No. AAI95-44240). Chomsky, N. (1957). Syntactic Structures. The Hague: Mouton. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Earlbaum Associates. Curtin, C., & Shinall, S. (1984). Computer-assisted reading lessons. CALICO, 1 (5). Davis, J. N., & Lyman-Hager, M. A. (1997). Computers and L2 reading: Student performance, student attitudes. Foreign Language Annals, 30 (1), 58-72. DeKeyser, R. (1994). Implicit and explic it learning of L2 grammar: A pilot study. TESOL Quarterly, 28 (1), 188-94.
94 DeKeyser, R. (1995). Learning second language grammar rules: An experiment with a miniature linguistic system. Studies in Second Language Acquisition, 17 (3), 379-410. DiDonato, R., Clyde, M., & Vasant, J. (1999). Deutsch, Na klar! (3rd ed.). Columbus, OH.: McGraw Hill. Dollenmayer, D., & Hansen, T. (1999). Neue Horizonte. Boston: Houghton Mifflin. Duchastel, P. C. (1978). Illu strating instructional texts. Educational Technology 36-39. Ehlers-Zavala, F. P. (1999). Reading an il lustrated and non-illustrated story: Dual coding in the foreign language classr oom. Illinois State University; Dissertation Abstracts International, 60 08A. (No. AAG99-42643). Ellis, R. (1990). Instructed second language acquisi tion: Learning in the classroom. Oxford: Blackwell. Ellis, R. (1994). The study of second language acquisition. New York: Oxford University Press. Enyedy, N. (1997). Constructing understanding: The role of anima tion in interpreting representations. AERA 1-15. Fodor, J.A., Bever, T.G., & Garrett, M. F. (1974). The psychology of language: An introduction to psycholoinguistics and generative grammar. New York: McGraw Hill. Gagne, R. M. (1968). Media and the learning process. (ERIC Document Reproduction Service No. ED 022 368). Gagne, R. M. (1970). Learning theory, educational me dia, and individualized instruction. (ERIC Document Reproduction Service No. ED 039 752). Garrett, N. (1986). The problem with gramma r: What kind can the language learner use. Modern Language Journal, 70 133-148. Gass, S., & Varonis, E. M. (1985). Variati on in native speaker speech modification to non-native speakers. Studies in Second Language Acquisition, 7 (1). Gregg, L. W., & Farnham-Diggory, S. (1975). Content and structure in learning. (ERIC Document No. ED 110 515). Hall, Christopher. ( 1998). Overcoming the grammar deficit: the role of information technology in teaching German grammar to undergraduates Canadian Modern Language Review, 55 (1), 41-60.
95 Hammoud, R. (1982). Utilisation de l'image mental et du champ d'associations dans l'enseignement du vocabulaire arabe des dbutants adultes francophones. [Use of mental imagery and associative fields in the teac hing of Arabic vocabulary to adult francophone beginners]. Unpublished doctoral dissertation, Laval University, Quebec, P.Q., Canada. Hansen, S., Schrimpsher, D., & Narayanan, N. (1998). From algorithm animations to animation-embedded hyperm edia visualizations. [Online] Available: http://citeseer.nj.nec.com/410448.html Hativa, N., Sarig, O., & Lesgold, A. (1991). Timing students' answers in CAI. Journal of Computer-Based Instruction, 18, (1). Hays, T. (1996). Spatial abilities and the eff ects of computer animation on short-term and long-term comprehension. Journal of Educational Computing Research, 14 (2), 139-55. Hill, J.R., & Hannafin, M.J. (1997). Cognitive strategies and learning from the World Wide Web. Educational Technology Research and Development, 45 (4), 37-64. Housen, A. (2001). The Role of form-focused instru ction in second language acquisition (PowerPoint) [On-line]. Available: http://www.uniss.it/facolta/lingue /varie/files/Instruction &SLA-2.ppt Hsieh, F. (1992). Effects of animation and manipulation of adult learning of mathematical concepts. Purdue University. Dissertation Abstracts International, 53 09A, (University Microfilms No. AAG93-01313). Hulstijn, J. H. (1993). When do foreignlanguage readers look up the meaning of unfamiliar words? The influence of task and learner variables. Modern Language Journal Sum, 77 (2), 39-47. Hulstijn, J. H. (2000). The use of computer technology in experimental studies of second language acquisition: a survey of some techniques and some ongoing studies. Language Learning and Technology, 3 (2), 32-43. Hwang, J. (1999). Attention and second language acquisition: The role of detection in the L2 acquisition of English passives. University of Oregon. Dissertation Abstracts International, 60, 08A. (University Microfilms No. AAG99-40407). Jonassen, D. H. (Ed.). (2001). Handbook of research for educational communications and technology. Retrieved March 14, 2002, from http:// www.aect.org/Intranet/P ublications/edtech/
96 Justice, L. M. (1999). The effect and affect of animated visual cues within a computerbased learning program. Ohio State University. Dissertation Abstract s International, 60, 05A. (University Microfilms No. AAG99-31625). Kehoe, M., & Stasko, J. (1996). Using animations to learn about algorithms: An ethnographic case study. [Online] Available: http://citeseer.nj.nec.com/60789.html Knowlton, J. (1996). A socioand psycho-lingui stic theory of pi ctorial communication. In D.H. Jonassen (Ed.), Handbook of research for educati onal communications and technology New York. Simon & Schuster Macmillan, 755-794. Kramsch, C. (1983). Interaction in the cl assroom: learning to negotiate roles and meanings. Unterrichtspraxis, 16 (2). Krashen, S. (1981). The "fundamental pe dogagical principle" in second language teaching. Studia Linguistica, 35 (1-2), 50-70. Krashen, S. (1981b). Second language acquisition and second language learning. Oxford; New York: Pergamon Press. Krashen, S. (1985). The input hypothesis: Issues and implications London: Longman. Krashen, S., & Terrell, T. (1983) The natural approach: Language acquisition in the classroom New Jersey: Alemany Press/Regents/Prentice Hall. Kulhavy, R.W., Lee, B., & Caterino, L.C. (1985) Conjoint retention of maps and related discourse. Contemporary Educati onal Psychology, 10, 28-37. Kulhavy, R.W., Stock, W.A., & Kealy, W.A. (1993). How geographic maps increase recall of instructional text. Educational Technology Research and Development, 41 47-62. Kulhavy, R.W., Stock, W.A., Peterson, S.E., Pridemore, D.R., & Klein, J.D. (1992). Using maps to retrieve text: A test of conjoint retention Contemporary Educational Psychology, 17, 56-70. Lai, S. (2000). Influence of audio-visual presentations on learning abstract concepts. International Journal of Instructional Media, 27 (2), 199-206. Lally, C. G. (1998a). Grammar in the sec ond language classroom: An ever-changing role. (ERIC Document Reproduction Service No. ED 424 752). [Online]. Available: http://www.edrs.com/members/fcla.cfm?an=ed424752
97 Lally, C. G. (1998b). The implicit versus ex plicit division in langua ge learning theory, method, and practice. Unterrichtspraxis, 31, 154-159. Large, A. (1996). Computer animation in an instructional environment. Library & Information Science Research, 18 (1), 3-23. LeBlanc, L. B., & Lally, C. G. (1998). A Comparison of instructor-mediated versus student-mediated explicit language instru ction in the communicative classroom. French Review, 71 734-746. Lee, S. (1996). The effects of computer animation and cognitive style on the understanding and retention of sc ientific explanation. Virginia Polytechnic Institute and State University. Dissertation Abstracts Internation, 57, 10A. (University Microfilms No. AAG9710954). Lee, H. (1997). The use of animation as a tool for concept learning. Ohio State University. Dissertation Abstracts Internation, 58, 05A. (University Microfilms No. AAG9731665). Leutner, D., & Plass, J.L. (1998). Measuri ng learning styles with questionnaires versus direct observation of preferenti al choice behavior in authentic learning situations: the visualizer/verbalizer behavior observation scale (VV-BOS). Computers in Human Behavior, 14 (4). Long, M. H., & Porter, P. A. (1985). Gr oup work, interlanguage talk, and second language acquisition. TESOL Quarterly, 19 (2), 207-28. Lovik, T.A., Guy, J. D., & Chavez, M. (1997). Vorsprung Boston: Houghton Mifflin. Marsh, Valeri (1998). Total physical response storytelling: A comm unicative approach to language learning. Learning Languages, 4 (1), 24-28. Mattoon, J. (1991). Effects of control mechan isms in instructional simulation. (ERIC Document Reproduction Service No. ED336077). Mayer, R. E. (1997). Multimedia learni ng: Are we asking the right questions? Educational Psychologist, 32 (1), 1-19. Mayer, R.E., & Anderson, R.B. (1992). The in structive animation: Helping students build connections between words and pictures in multimedia learning Journal of Educational Psychology, 84 444-452.
98Mayer, R.E., & Anderson, R.B. (1991). Animati ons need narrations: An experimental test of a dual-coding hypothesis. Journal of Educational Psychology, 83 (4), 484-90.Mayer, R.E., & Gallini, J.K. (1990). When is an illustration worth ten thousand words? Journal of Educational Psychology, 82 (4), 715-26. Mayer, R.E., & Moreno, R. (2002). Anima tion as an aid to multimedia learning. Educational Psychology Review, 14 (1), 87-99.Mayer, R.E., & Moreno, R. (1998). A split-at tention effect in multimedia learning: Evidence for dual processing systems in working memory. Journal of Educational Psychology, 90 (2), 312-20. Mayer, R.E., & Sims, V.K. (1994). For whom is a picture worth a thousand words? Extensions of a dual-coding theory of multimedia learning. Journal of Educational Psychology, 86, 389-401. Mayton, G. (1991). Learning dynamic processes from animated visuals in microcomputer-based instruction In: Proceedings of selected research presentations at the annual convention of the Association for Educational Communications and Technology. (ERIC Document Reproduction Service No. ED 334 999). Mayton, G. (1990). The effects of the anim ation of visuals on the learning of dynamic processes through microcomputer-based instruction. Dissertation Abstracts International, 51 12A. (University Microfilms No. AAG91-11760).McCarthy, M., & Carter, R. (1995). Spoken gram mar: What is it and how can we teach it? ELT Journal, 49 3. McGallagher, J. (1975). Vocabulary retention of lo wer-class students in language experience and in basal text appr oaches to the teaching of reading Unpublished dissertation. University of South Carolina. (ERIC Document Reproduction No. ED 126 444). Mitchell, J. T. & Redmond, M. L. (1993) Rethinking grammar and communication. Foreign Language Annals, 26 (1), 13-19. Moeller, J., Adolf, W., HoecherlAlden, G., & Lalande, J. (2000). Deutsch heute (7th ed.). Boston: Houghton Mifflin.
99 Moore, D.M., Burton, J.K., & Myers, R.J. (2001). Multiple-channel communication: The theoretical and research foundations of multimedia. In D.H. Jonassen (Ed.), Handbook of research for educational co mmunications and technology, 26, 851-875. Retrieved March 14, 2002, from http://www.aect.org/Intranet/Publications/edtech/ Moray, N. (1969). Attention: selective pro cesses in vision and hearing. London: Hutchinson Educational. Moray, N. (1959). Attention in dichotic list ening: Affective cues and the influence of instructions. Quarterly Journal of Ex perimental Psychology, 11 56-60. Morrison, J. B., Tversky, B., & Betrancour t, M. (2000). Does animation facilitate learning? International Journal of Human-Computer Studies [Online] Referenced June 2, 2002. Available: http://w5.cs.uni-sb.de/~butz/ AAAI-SSS2000/cameready/JMorrison00.pdf Morse, T. (1999). An overview of tec hnology applications in special education. School Business Affairs, 65 (2). Morton, J. (1979). Facilitation in word rec ognition: Experiments causing change in the logogen model. In P. A. Kolers, M. Wrolstead, & H. Bouma (Eds.). Processing of visible language (Vol. 1). New York: Plenum Press. Nagata, N. (1997). An experimental comp arison of deductive a nd inductive feedback generated by a simple parser. System, 25 (4). Nagata, N. (1995). An Effective Applicati on of Natural Language Processing in Second Language Instruction. CALICO Journal, 13 (1), 47-67. Nagata, N. & Swisher, M.V. (1995). A study of consciousness-raising by computer: The effect of metalinguistic fee dback on second language learning. Foreign Language Annals, 28 (3), 337-47. Nicolls, C. (1996). The effect of comput er animation on students' understanding of microbiology. Journal of Research on computing in Education, 28 (3), 359-371. Okolo, C., & Hayes, R.(1996). The impact of animation in CD-ROM books on students' reading behaviors and comprehension. (ERIC Document Reproduction Service No. ED 395 434).
100 Packard, A. (1994). The effect of animation on the in struction of the principles of ANOVA. Paper presented at the Annual Meeting of the Mid-South Educational Research Association. Nashville, TN. (ERIC Document Reproduction Service No. ED 387 529). Packard, A. (1996). Exploration of Brunswik learning environment for instruction of basic sampling concepts. (ERIC Document Reproduction Service No. ED 393 442). Packard, A. et al. (1997). Incorporating individual learning styles into computer assisted environments for instru ction basic research and statistical concepts. Paper presented at the Eastern Educational Research Association. Hilton Head, SC. (ERIC Document Reproduction Service No. ED 408 328). Paivio, A. (1971). Imagery and Verbal Processes. New York: Holt, Rinehart and Winston. Paivio, A. (1990). Mental representations: A dual coding approach New York: Oxford University Press. Paivio, A., Clark, J. M., Pressley, M., & Desrochers, A. (unpublished paper). Imagery mnemonic and control strategies in second-language vocabulary learning. Paivio, A., & Desrochers, A. (1979). E ffects of an imagery mnemonic on second language and comprehension. Canadian Journal of Psychology, 33 (17-28). Paivio, A., & Desrochers, A. (1981). Mnem onic techniques in s econd language learning. Journal of Educational Psychology, 73 (6). Paivio, A., & Lambert, W. (1981). Dual coding and bilingual memory. Journal of Verbal Learning and Verbal Behavior, 20 (5). Palmer, S.E. (1978). Fundamental aspects of c ognitive representations. In E. Rosch and B.B. Lloyd (Eds.). Cognition and categorization New York: Lawrence Erlbaum Associates. Park, 0., & Hopkins, R. (1993). Instructi onal conditions for using dynamic visual displays: A review. Instructional Science, 22 1-24. Pica, T. (1996). Second language learning thr ough interaction: Multiple perspectives. Working Papers in Educ ational Linguistics, 12 (1), 1-22. Pica, T. (1994). Research on negotiation: What does it reveal about second-language learning conditions, processes, and outcomes? Language Learning, 44, (3).
101 Pienemann, M. (1995). Second language acquisition: A first introduction Paper presented at the National Language and Literacy In stitute of Australia, University of Western Sydney. (ERIC Document Reproducti on Service No. ED 425 637). Plass, J. L., Chun, D. M., Mayer, R.E., & Leutner, D. (1998). Supporting visual and verbal learning preferences in a second-la nguage multimedia learning environment. Journal of Educational Psychology, 90 (1), 25-36. Poohkay, B., & Szabo, M. (1995). Effects of animation and visuals on learning high school mathematics. In: Proceedings of the Association for Educational Communications and Technology, Anaheim, CA. (ERIC Reproducti on Service Document No. ED 380 122). Purpura, J. E. (1997). An analysis of the relationships between test takers' cognitive and metacognitive strategy use and second language test performance. Language Learning, 47 (2), 289-325. Ramirez, A., & Rivard, S. (1998). Hypermedia aids for advanced learning in complex and ill-structured knowledge domains. In: Proceedings of the International Academy for Information Management (IAIM) Annual Conf erence. Helsinki, Finland. (ERIC Document Reproduction Service No. ED 431 435). Rehaag, D. M., & Szabo, M. (1995). An experiment on eff ects of redundant audio in computer based instruction on achievement, at titude, and learning time in 10th grade math. Paper presented at the Annual Me eting of the Association for Educational Communications and Technology. Anaheim, CA. (ERIC Reproducti on Service Document No. ED 380 123). Rieber, L. P. (1989a). A review of animation research in computer-based instruction. (ERIC Document Reproduction Service No. ED 308 832). Rieber, L. P. et al. (1989b). The effects of computer ani mated lesson presentations and cognitive practice on adult learning in physical science. (ERIC Document Reproduction Service No. ED 308 834). Rieber, L. P. (1989c). The effects of computer animated lesson presentations and cognitive practice activities on young child ren's learning in physical science. (ERIC Document Reproduction Service No. ED 308 833).
102 Rieber, L. P. (1989d). The effects of comput er animated elaboration strategies and practice on factual and application lear ning in an elementary science lesson. Journal of Educational Computing Research, 5 (4), 431-44. Rieber, L. P. (1990a). The effects of animated vis uals on incidental learning and motivation. (ERIC Document Reproduction Service No. ED 323 943). Rieber, L. R. (1990b). Animation in computer-based instruction. Educational Technology, Research and Development, 38 (1), 77-86. Rieber, L.P. (1991). Animation, incident al learning, and continuing motivation. Journal of Educational Psychology, 83 (3), 318-328. Rieber, Lloyd P. (1995). A historical review of visualization in human cognition. Educational Technology Research and Development, 43 (1), 45-56. Rieber, L.P. (1996a). Animation as fee dback in a computer-based simulation: Representation matters. Educational technology re search and development, 44 (1), 5-22. Rieber, L. P. (1996b). Feedback and elaboration within a computer-based simulation: A dual coding perspective. (ERIC Document Reproduction Service No. ED 397 829). Rieber, L. P., & Hannafin, M. J. (1988). E ffects of textual and animated orienting activities and practice on learning fr om computer-based instruction. Computers in the Schools, 5 (1-2), 77-89. Rieber, L., & Parmley, M. W. (1992). Effects of animated computer simulations on inductive learning with adults: a preliminary report In: Proceedings of Selected Research and Development Presentations at the Conven tion of the Association for Educational Communications and Technology. (ERIC Document Reproduction Service No. ED 348 019). Rittschof, K. A., & Kulhavy, R. W., (1998) Learning and remembering from thematic maps of familiar regions. Educational Technology, Research and Development, 46 (1), 19-38. Robinson, P. (1995). Attention, memory, and the "noticing" hypothesis. Language Learning, 45 (2), 283-331. Robinson, P. (1997). Generalizability and au tomaticity of second language learning under implicit, incidental, enhanced, and instructed conditions Studies in Second Language Acquisition, 19 (2), 223-47.
103 Sadoski, M., & Paivio, A. (2001). Imagery and text: A dual c oding theory of reading and writing. Mahwah, NJ: Lawrence Erlbaum Asssociates. Savignon, S.J. (1991). Communicative language teaching: State of the art. TESOL Quarterly, 25 (2), 261-77. Schleppegrell, M., & Oxford, R. (1988) Language learning strategies for Peace Corp volunteers. (ERIC Document Reproduction Service No. ED 328 877). Schmidt, R. (1992). Awareness and second language learning. Annual Review of Applied Linguistics, 13, 206-26. Schnotz, W. (2002). Towards an integrated view of learning from text and visual displays. Educational Psychology Review, 14 (1), 101-119. Schnotz, W., & Grzondziel, H. (1996). Knowledge acquisition with static and animated pictures in computer-based learning. Paper presented at the A nnual Meeting of the American Educational Research Association, New York, NY. (ERIC Document Reproduction Service No. ED 401 878). Seliger, H. (1975). Inductive method a nd deductive method in language teaching. International Review of Applied Linguistics in Language Teaching, 13 (1). Selinker, L. (1972). Interlanguage. International Review of Applied Linguistics in Language Teaching, 10 (3), 209-231. Sevin, D., & Sevin, I. (2002). Wie Geht's. (6th ed.). New York: Heinle. Sharwood-Smith, M. (1991). Speaking to ma ny minds: On the relevance of different types of language information for the L2 learner. Second Language Research, 7 (2), 118-32. Sheen, R. (1996). The advantage of exploiti ng contrastive analysis in teaching and learning a foreign language. IRAL, 34, (3). Shepard, R.N. (1978). The mental image. American Psychologist, 33 125-137. Shih, Y., & Alessi, S.M. (1994). Mental mode ls and transfer of learning in computer programming. Journal of Research on Computing in Education, 26 (2), 154-175. Spada, N., & Lightbown, P. (1990). Focus-on-form and corrective feedback in communicative language teaching: E ffects on second language learning. Studies in Second Language Acquisition, 12 (4), 429-48.
104 Speelman, P. K. (1997).The effects of computer generated text slides with animation on short-term retention of knowledge. Wayne State University. Dissertation Abstracts International, 58, 11A. (University Microfilms No. AAG98-15383). Stasko, J. (1996). Do algorithm animations aid learning? Technical Report GIT-GVU96-18, GVU Center, Georgia Institute of Technology. Stasko, J., & Badre, A. (1993). Do algorithm animations assi st learning? An empirical study and analysis Proceedings of ACM INTERCHI'93 61-66. Szabo, M. (2001). Survey of in structional technol ogy research. [Online] Referenced July 13, 2002. Available: http://www.quasar.ualberta.ca/ed media/readingsnc/Nrefsza.html Szabo, M., & Poohkay, B. (1996). An experimental study of animation, mathematics achievement and attitude toward computer-assisted instruction. Journal of Research on computing in Education, 28 (3), 390-402. Tabachnick, B.G., & Fidell, L.S. (1989). Using multivariate statistics New York: HarperCollins. Tenberg, R., Tenberg, R., & Ainslie, S. (1998). Deutsch Plus: A German experience: A self-guided course for beginners learning German. Columbus, OH.: McGraw Hill. Tennyson, R. D. (1985). Interactive effect of information structure, sequence of information, and process learning time on rule learning using computer-based instruction. Educational Communication and Technology Fall, 33 (3). Terrell, T.D., Tschirner, E., & Nikolai, B. (2000). Kontakte. (4th ed.). New York: McGraw-Hill. Terrell, T.D., Tschirner, E., & Nikolai, B. (2000b). The Natural Approach. [Online] Referenced March 3, 2002. Available: http://www.mhhe.com/socscienc e/german/kontakte/natural.htm Tomlin, R. S., & Villa, V. (1994). Atten tion in cognitive science and second language acquisition. Studies in Second Language Acquisition, 16 (2), 183-203. Towers, R. L. (1994). The effects of animat ed graphics and static graphics on student learning in a computer-based instruct ional format. University of Kentucky. Dissertation Abstracts International, 55 0A, (University Microfilms No AAG94-26185). Treisman, A. M. (1966). Our limited attention, Advances in Science 22 600-611.
105 Treisman, A. M., & Davies, A. (1973). Divided attention to ear and eye. In S. Kornblum (Ed.), Attention and Performance IV London: Academic Press, pp. 101-117. Tversky, B. (1995). Cognitive origins of graphic productions. In A. Wylde (Ed.), Understanding images: Finding meaning in digital imagery. New York: Marchese. Tversky, B. (1994). Spatial mental models from descriptions. Journal of the American Society for Information Science, 45 (9), 656-668. VanPatten, B. (1987). On babies and bathwa ter: Input in fore ign language learning. Modern Language Journal, 71 (2). VanPatten, B. (1993). Grammar teachi ng for the acquisition-rich classroom Foreign Language Annals, 26 435-50. Van Patten, B., & Cadierno, T. (1993a). Input processing and second language acquisition: A role for instruction. The Modern Language Journal, 77 45-57. Van Patten, B., & Cadierno, T. (1993b). Exp licit instruction and input processing. Studies in Second Language Acquisition, 15 225-43. Van Patten, B., & Oikkenon, S. (1996). Explanati on versus structured input in processing instruction. Studies in Second Language Acquisition, 18 495-510. Verdi, M.P., Johnson, J.T., Stock, W. A., Kulhavy, R.W., & Whitman-Ahern, P. (1997). Organized spatial displays and texts: Effects of presentation order and display type on learning outcomes. The Journal of Experimental Education, 65 (4), 303-317. Verdi, M. P., & Kulhavy, R. W. (2002). Learning with maps and texts: An overview. Educational Psychology Review, 14 (1), 27-40. Viteli, J. (1989). Learning styles and individual differenc es in learning english idioms via computer assisted language learni ng in english as a second language. Paper presented at the Annual Meeting of the European Association of Research on Learning and Instruction. Madrid, Spain. (ERIC Document Reproduction Service No. ED 320 559). Williams, J. (1999). Memory, attention, and inductive learning. Studies in Second Language Acquisition, 21 (1), 1-48. Williamson, V. M. (1992). The effects of computer animation emphasizing the particulate nature of matter on the understandings and miscon ceptions of college chemistry
106 students. University of Oklahoma. Dissertation Abstracts International, 53 06A, (University Microfilms No AAG92-30141). Williamson, V.M., & Abraham, M.R. (1995). The effects of computer animation on the particulate mental models of college chemistry students Journal of Research in Science Teaching, 32 (5), 521-34. Winn, W., & Solomon, C. (1993). The effect of the spatial arrangement of simple diagrams on the interpretation of English and nonsense sentences Educational Technology, Research and Development, 41 (1), 29-41. Wong, A. L. (1994). The use of animation in computer-assisted instruction. Carleton University. Dissertation Abstract s International, 55 12A, (University Microfilms No AAINN92927). Yang, N. (1995). Effective awareness-raising in language learning strategy training Paper presented at the Annual Meeting of the Teachers of E nglish to Speakers of Other Languages (29th, Long Beach, CA, March 26-April 1, 1995). (ERIC Document Reproduction Service No. ED392278). Yoshii, M. (2000). Second language incidental vocabulary retention: The effect of text and picture annotation types. University of South Florida ; Dissertation Abstracts International, 61, 08A. (No. AAI99-84602). Yoshii, M., & Flaitz, J. (2002). Second langua ge incidental vocabul ary retention: The effect of text and pi cture annotation types. CALICO Journal 20, (1). Young, R. (1983). The negotiation of meaning in children's foreign language acquisition. ELT Journal, 37 (3).
108 Appendix A. Note to the ParticipantsReturn to text GERMAN MODAL AUXILIARY VERBS An Experiment In Computer Assisted Language Learning By Lizz Caplan-Carbin You are invited to participate in a study to test your pre-instruction and postinstruction knowledge of German modal auxiliary ve rbs. There are also a few questions concerning your experience with German language learning and with second language acquisition in general. Following the pretest you will be invited to view a presentation about German modal verbs, after which, you will be as ked to take another test. This posttest is designed to test your immediate understanding of the new material. This is an experimental study in second language acquisition, for which your participation is respectfully requested. The content of the study, modal verbs, is an important part of your German language curriculum, therefor e, viewing the instruct ional presentation is a requirement of the course. However, your particip ation with regards to all testing instruments is strictly voluntary. That means that you do not ha ve to take any of th e tests involved in the study. Your grade in your German I course will not be affected in any way by your decision to accept or decline this invi tation to participate.* Should you choose to take part in this expe riment, you will be assigned a code number, and your test scores will not be identifiable with your name. Your test scores, therefore can not be calculated into your course grade in any way. Your sincere participation is requested so that your contribution to the knowledge gathered here may be accurately accessed. There will be 10 minutes allotted to take the pretest and to answer the short questionnaire. The presentation will last 20 minutes, and then you will have 20 minutes to answer the posttest questions. Since this is a very tight schedule, the pretest must begin promptly at the start of the next class period. Again, I ask for your cooperation and sincere e ffort, in order to provide the most accurate picture of the effect of instruction. Thank you for your assistance with this study. *When you turn in your posttest exam, you will be given a lottery ticket for a chance at one of three gift certificates to Mr. Dunderbaks Germ an Restaurant, worth $5, $10, and $15. The winning tickets will be drawn during the Phi Sigma Iota / German Club Weihnachtskaffee.
109 Appendix B. Pretest Survey Responses Return to text Participant#group Studied 1-2 semesters Lived in Germany German at home Fluent/lacking grammar Just starting Studied 1-2 semesters lived in Near-native Fluent Survival level 1101hs 12010L. & I. 13010F. L. & G. 14010S. 15010S. 16010S. I. & F.S. 17013 yrs.1 18010S. 21012 yrs.11F.G. 2201hs.8 yrs.11S. 230120 yrs.3 yrs.1S. F. & R. 23010S. 25010F. 26010F. S. & IP. 27010S.SpainS.Cz 31012 yrs2 yrsPolandE. (L2)Pol. 32014 yrs.R. 33010I. & F.IAlb.F. 34010L. 35016 yrs.3 yrs.S. 36019 yrs.5 yrs. 110220 yrsS. & I. 12020F. & I. 13020S. & F. 14027 yrs 15020S. 160225 yrsF. & I.I. 17020S. & R.S. 18020S. & I. 19020Sp. & L. 2102110S. 22021 yr. 23028 yrs.monthsS.RomaniaE. (L2)Rom.G. 24020S. 25020E. (L2)S. 26026 mos.0 270214 yrs.1L.MaltaG. 31026 yrs.F. 3202months0 33024 yrs.5 yrs.G. 34021 yr.TaiwanC. 35023 yrs.1 36020S. 37020
110 Appendix C. Relevant Websites To view the animated version of th e word order instruction, point a WWW browser to the following URL: http://www.webgerman.com/study/modals.html Bookmarks to websites on German grammar: http://webgerman.com/germlinks/main.htm#grammar Animated German grammar presentations: http://webgerman.com/Presentations
111 Appendix D. Pretest Return to text A. Check the box next to the most appropr iate meaning for each modal verb. drfen to be aware of to be allowed to to be subscribed to knnen to be obligated to to be able to to be informed of mssen to be obligated to to be hoping to to be interested in wollen to be willing to to be intent on to be counting on sollen to be supposed to to be afraid of to be ready for mchten to wish for to wait for to plead for B. Fill in the four blanks, next to each of the following six pronouns, with the proper forms of the four modal verbs at the top of each column. mchten knnen drfen mssen 1. ich ____________ ______________ _____________ ____________ 2. du ____________ ______________ _____________ ____________ 3. er ____________ ______________ _____________ ____________ 4. wir ____________ _____________ _____________ ____________ 5. ihr ____________ _______________ _______________ ____________ 6. Sie ____________ ______________ _____________ ____________
112 Appendix D. (Continued)C Compose sentences by placing a word onto each blank line next to the jumbles below. 1. du duwarten wartendrei dreiStunden Stundenhier hiermusst musst _____ _____ ______ ______ _______ ______ 2. kann kannspielen spielensehr gut sehr guter erGitarre Gitarre ______ ______ ______ ____ __ ______ ______ 3. jetzt jetztRuhe Ruhemuss mussich ichetwas etwashaben haben ______ ______ ______ ____ __ _____ ______ 4. l nger l ngerk nnen k nnennicht nichtwarten wartenSie Sie ______ ______ _______ _______ ______ 5. m chtet m chtetKreditkarte Kreditkarteihr ihrbezahlenbezahleneiner einerm it m it ______ _______ _______ _______ _______ _____
113 Appendix E. Posttest Return to text Rewrite each sentence to make it match the translation beneath. The modal is given in parentheses. 1. Ich bleibe hier nur kurz. (drfen) _____________________________________________________________. I may stay here only a little while. 2. Du kaufst hier billig ein. (knnen) _____________________________________________________________. You can shop cheaply here. 3. Es regnet bald. (mssen) _____________________________________________________________. It must rain soon. 4. Ihr bereitet eure Hausaufgaben vor. (sollen) _____________________________________________________________. You should prepare your homework. 5. Du fhrst mit dem Auto aufs Land. (sollen) _____________________________________________________________. You should drive into the country by car. 6. Ich warte nicht lang. (wollen) _____________________________________________________________. I dont want to wait long. 7. Sie kommt morgen zum Mittagessen. (mchten) _____________________________________________________________. She would like to come to lunch tomorrow. 8. Wir gehen heute ins Museum. (sollen) _____________________________________________________________. We should go to the museum today. 9. Er kommt um acht Uhr an. (wollen) _____________________________________________________________. He wants to arrive at 8:00.
114 Appendix E. (Continued) 10. Sie steht schon so frh auf. (wollen) _____________________________________________________________. She wants to get up so early. 11. Man raucht in diesem Gebude nicht. (drfen) _____________________________________________________________. One may not smoke in this building. 12. Ihr seht den Horizont noch. (knnen) _____________________________________________________________. You can still see the horizon. 13. Ihr reist viel mehr. (mssen) _____________________________________________________________. You must travel a lot more. 14. Sie verpassen die Strasse nicht. (knnen) _____________________________________________________________. You cant miss the street. 15. Er fhrt um acht Uhr ab. (mchten) _____________________________________________________________. He would like to leave at eight o-clock. 16. Du bringst das mit. (drfen) _____________________________________________________________. You may bring that along with you. 17. Ich bezahle mit einer Kreditkarte. (mchten) _____________________________________________________________. I would like to pay with a credit card. 18. Du machst die Tr bitte zu. (mssen) _____________________________________________________________. You have to close the door, please.
115Appendix F. Qualitative Instruments Return to text Participant _________ 1. How would you describe your prior experience with German? Studied 1-2 semesters, a long time ago how long ago? ______________ Spent some time in Germany how long?________________ German was used in the home setting how often? _________________ Conversational ability, but need grammar skills. Just learning for the first time now.2. What is your experience with language learning? Studied 1-2 semesters of ___________________________ (what language/s?) Spent some time in ____________ (what country/ies) how long? _________ Near-native speaking abilit y in _____________________ (what language/s?) Fluent speaking ability in __________________________ (what language/s?) Survival level speaking ability in ____________________ (what language/s?)Posttest QuestionnaireComments about the test or th e instructional presentation? ____________________________________________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ Pretest Questionnaire
116 Appendix G. Spatial Conventions Study In order to test the conventi onal expectations that influen ced the design of the animations for this experiment, a small case study was performed. The animation sequence of the instructional program was segmented into five se parate screens, one for each discreet motion or transformation. These also corresponded to the fi ve steps of the grammar transformation. The text of the sample sentence was changed to sc rambled letters forming nonsense words that were meant to convey no meaning whatsoever. Five members (one staff, one professor, one student, and two TAs) of a university foreign language department helped with the case study. The participants were asked to view each of the five animations, depicting a single motion of one word or part of a word, plus a sequence that combined the five into a single animated transfor mation. For each of the five discreet animations, they were asked to (1) supply the verb which would describe the animation, and (2) choose the most appropriate descriptive verb from a prin ted list. For the combined sequence, they were asked to (3) provide a verbal de scription of the even t, and (4) rate the accuracy of a verbal description supplied by the researcher. The sample sentence, consisting of invented, nonsense "words", appeared in the center of the screen as "Tuo Inkus tsiv jni blkii Uadkna." Th e word "yallic" appeared at the bottom of the screen. For each of the five screens, participants were asked to first describe the motion in their own words and then choose a descriptive verb from a printed list as follows: Replace Position Emerge Converge Revert Push Exchange Contact Repress Pull Enmesh Center Release Partition Envelope Crawl Remove Prepare Engage Conjugate Participants were then shown an animation sequence combining the five separate motions into a single event. Shown the sequence a second time, they were asked to listen to a verbal description, synchronized to matc h the animation, and rate the accu racy of the description using a percentage value. The following transformations were made throughout the presentation: Tuo Inkus tsiv jni blkii Uadkna. Tuo Inkus yull jni blkii Uadkna tsivic. (yallic yull) (tsiv tsivic) Participant responses were recorded on a mini -CD. The researcher explained the procedures to each participant and then remained silent dur ing the presentation and recording of comments. Below is a transcript of the participants' responses recorded during the study. The numbers 1-6 represent the five screens and the combined anim ated sequence. For #7 participants were asked to rate how well they agree that the following se ntence described the animated event: "This is a sentence, notice this position, this emerges from this, and takes the place of this, pushing it to the end, where it takes another form." The word "t his" was accompanied by pointing to various parts of the screen. Return to text
117 Appendix G (Continued.) Participants' Descriptions Of Animations Participant 1 1. surrounding, emerge 2. conjugate 3. replacement, replace 4. moving position to the end, remove to end 5. changing the ending, conjugation thing, movi ng to the end and changing the ending 6. conjugation of the verb, moving the existing ve rb towards the end, making it agree with the second subject, or now modifying something else 7. 100% Participant 2 1. a box coming up, a rectangle / emerge 2. is that maybe the base word? / exhange 3. push, exchange and replace 4. replaced one word and moved it / replace (move) 5. is that like a conjugation of a word? / conjugation 6. ( silence) 7. 100% Participant 3 1. emerge, zooming in on 2. conjugate, exchange 3. replace 4. moving 5. position, moving to the end, pushed to the end 6. replace, move 7. 100% Participant 4 1. emerge, highlight 2. conjugate, emerge 3. replace, remove 4. remove, move 5. position, push 6. conjugation, replacement 7. 100%
118 Appendix G (Continued.) Participant 5 1. highlighting 2. conjugate, conjugate 3. replace, replacing 4. exchanging and moving it, Replacing the one and moving it to the end of the sentence. 5. declension, conjugation 6. conjugation 7. 100% Participants were also asked to give a verbal description of the entire animation. Here are four responses given when participants were as ked to describe the entire animation sequence. One participant remained silent. Participant descriptions of total animated sequence. 1. You're replacing "tziv" with "yull" moving that to the end, conjugating it in some way. 2. Zooming in on a word to replace it with anot her and then adding the word at the end in another form. 3. Conjugation of the verb, moving the existing ve rb towards the end, ma king it agree with the second subject, or now modifying something else. 4. Might be infinitive. The conj ugation of the verb is replaci ng another, which was moved to the end of the sentence and was ch anged to another part of speech. Summary of responses for each of the five screens Emerging (x2) A box coming up Surrounding A rectangle Zooming in on Highlighting Conjugate (x4) Exchange (x2) Is that maybe the base word? Replacement (x5) Push Exchange Remove Moving (x5) To the end (x2) Remove (x2) Exchange (x2) Conjugation (x4) Another form (x2) Changing the ending Moving to the end (x2)
119Appendix H. Instructional Screens with Narrative Script Return to text Slide 1 In this program you are going to learn about modal verbs. / There are three areas of consideration concerning modal verbs: / their meanings, / their special conjugations, / and their effect on word order. Slide 2 mssen mssen sollen sollen wollen wollen mchten mchten knnen knnen drfen drfenModal verbs have meanings that express mood. Every modal verb can be combined with any action infinitive to form a modal verb phrase. A modal verb phrase expresses action and the mood or attitude toward that action.Meanings of Modal Verbs Meanings of Modal VerbsYou understand .You should understand.Example:Modal verb Action verb The six modal verbs: sollen, knnen, drfen, mchten, wollen, and mssen, / are verbs that express mood. Every modal verb can be combined with any action infinitive to form a modal verb phrase. / A modal verb phrase expresses action and the mood or attitude toward that action. / For example, adding the modal verb should to the sentence You understand, expresses an attitude toward the action You should understand. Should is the modal verb, while understand is the action verb. Slide 3 mssen mssen sollen sollen wollen wollen mchten mchten knnen knnen drfen drfen You should understand. You can understand. You may understand. You would like to understand. You want to understand. You must understand. Any modal verb can be combined with any infinitive to express a new mood or attitude about the action verb. You should understand. differs only in mood from the phrase, you can understand. The following phrases each express a different attitude about the idea you understand : Just as in English, the infinitive or action verb is interchangeable with any other infinitive, and the modal verb is interchangeable with any other modal. Slide 4 He does it.machen Er macht es .to do: to be supposed to ought to shouldEr soll es machen.He should do it.sollensollento be supposed to ought to should Modal verb sollen / means should / It also means ought to / as well as to be supposed to These three meanings are essentially equivalent. / Take the action verb to do / rendered as machen / in the simple sentence Er macht es. / He does it Add a modal verb / such as sollen / meaning should as in / Er soll es machen, / and the meaning is now he should do it. Or He ought to do it. Or even He is supposed to do it. Meaning Meaning Conjuga Conjuga Order Order tion tion Word WordModal Verbs Modal Verbs
120Appendix H. (Continued)Slide 5 He carries it.tragen Er trgt es .to carry: to be able to canknnenEr kann es tragen.to be able toHe can carry it.canknnen Modal verb knnen / means can / It also means to be able to These two meanings are basically equivalent. Take the action verb to carry / rendered as tragen in the simple sentence / Er trgt es. / He carries it ./ Add the modal verb knnen / meaning can as in / Er kann es tragen, and the meaning is now / he can carry it. Or / He is able to carry it. / Slide 6 We stay here.bleiben Wir bleiben hier.to stay: to be permitted to to be allowed maydrfenWir drfen hier bleiben.to be allowed to to be permitted toWe may stay here.maydrfen Modal verb drfen means may It also means to be allowed to or to be permitted to These three meanings are roughly equivalent. With the action verb to stay rendered as bleiben in the simple sentence Wir bleiben hier. We stay here. Adding the modal verb drfen meaning may changes the sentence to Wir drfen hier bleiben -We may stay here. Also rendered as We are allowed to stay here, and We are permitted to stay here. Slide 7 You help us.helfen Ihr helft uns.to help:mchtenIhr mchtet uns helfen.to have a desire toYou would like to help us.would like tomchtenwould like to to have a desire to Modal verb mchten means would like to It also means to have a desire to relatively equivalent meanings. Along with the action verb to help German helfen, the sentenceIhr helft uns. You help us, and the addition of the modal verb would like to meaning mchten gives us Ihr mchtet uns helfen, or You would like to help us /. Slide 8 You are becoming skilled.to intend toYou want to become skilled.werdenwollenDu wirst begabt. Du willst begabt werden.to become: to want to to intend to to want to to plan towollen Modal verb wollen means / to want to / in the sense of / to intend to Take the action verb / to become / which is werden in German. The simple sentence / Du wirst begabt / or You are becoming skilled. / With the addition of the modal verb / wollen meaning to want to, / to intend to becomes / Du willst begabt werden, and the meaning is now / Y ou want to become skilled, in the sense of / You intend to become skilled. / Expressing intention, wollen is stronger and more definite than desire mchten. For this reason, / wollen is often rendered as to plan to in English. /
121Appendix H. (Continued)Slide 9 to give:They give more. They must give more.mssen mssenSie geben mehr. Sie mssen mehr geben.to be obligated to to have to must to be obligated to to have to mustgebenmssen Modal verb mssen means / must It also means to have to / as well as to be obligated to These three meanings are basically equivalent. Here is the action verb / to give / -geben in the simple sentence / Sie geben mehr. / They give more. Adding the modal must / with mssen / as in Sie mssen mehr geben expresses the meaning, / They must give more. Or / They have to give more / Or even They are obligated to give more. / Slide 10 to be obligated to to have to must to intend to to want to would like to to have a desire to may to be permitted to to be allowed to to be able to can to be supposed to ought to should mssen mssen sollen sollen wollen wollen mchten mchten knnen knnen drfen drfen English also expresses mood with modal verbs, but they usually not always single words. They are often a conglomeration of words, such as / to be supposed to, / to be allowed to / to have a desire to / to be obligated to. German uses single words to convey these longer verb phrases. / German also offers single words to convey multiple meanings for English speakers. / To be allowed to, / to be permitted to / and may are all rendered into Germ an with the one word, / drfen. / To have to, / to have an obligation to, / and must are all equally expressed with the single German word / mssen. Slide 11 The second thing to consider is the conjugation of modal verbs. / Notice that the er form does not have its customary ~t ending. For modal verbs, the er form has the same endings as the ich form. Since the upper-case, formal Sie form is identical to the plural sie form, only the lower-case word will be used in this presentation. Ich mchte has the same verb form as Er mchte. The rest of the subject pronouns take their customary verb endings: du mchtest, wir mchten, ihr mchtet, and sie mchten. Mchten is the only word that has the -e endings. The other modal verbs all have no endings for these forms, as the following slides will show. ich du erNotice that the er form does not have its customary ~t ending. -e -st -e Mchten is the only word that has the -e endings. The other modal verbs all have no endings for these forms. Wir ihr sie -en -t -en mchte mchtest mchte mchten mchtet mchten e e Sie -st -en -t -en -Conjugation of modal verbsFor modal verbs, the er form has the same endings as the ich form. sie
122Appendix H. (Continued)Slide 12 Modal verbs sollen and mchten have regular conjugations. st en t en ich du er wir ihr sie soll soll soll soll soll soll ich du er wir ihr sie mchte mchtest mchte mchten mchtet mchten Modal verbs sollen and mchten have regular conjugations. ich soll, du sollst, er soll, wir sollen, ihr sollt, sie sollen ich mchte, du mchtest, er mchte wir mchten, ihr mchtet, sie mchten. The stem vowel of both sollen and mchten remains unchanged for all conjugations Slide 13 st en t en will will will woll woll woll ich du er wir ihr sieExcept for sollen and mchten modal verbs have a stem-vowel change for the ich , du and er forms. stem-vowel change stem-vowel change stem-vowel change stem-vowel change kann kann kann knn knn knn darf darf darf drf drf drf muss musst muss mssen msst mssen Except for sollen and mchten all of the other modal verbs have a stem-vowel change for the ich, du and er forms. / Wollen for example, changes to will for ich and er forms and willst for the du form. ich will, du willst, er will It takes the unique / stem-vowel change, o>i But remains wir wollen, ihr wollt and sie wollen. / Modal verb knnen becomes ich kann, du kannst, er kann, but wir knnen, ihr knnt, and sie knnen. The ich, du, and er forms all undergo the equally unique / stemvowel change >a. / Modal verb drfen becomes ich darf, du darfst, er darf, but wir drfen, ihr drft and sie drfen. The ich, du, and er forms all undergo the unique / stem-vowel change >a. / Modal verb mssen becomes ich muss, du musst, er muss, but wir mssen, ihr msst and sie mssen. / The stem-vowel change undergone by the ich, du, and er forms is simply the loss of the original umlaut. compare: Wir mssen and ich muss ihr msst and du musst Slide 14 ich du er mchte mchtest mchte mchten mchtet mchten wir ihr sie will willst will wollen wollt wollen kann kannst kann knnen knnt knnen darf darfst darf drfen drft drfen muss musst muss mssen msst mssen soll sollst soll sollen sollt sollenHere are all of the modal verb conjugations, with the irregular stem-vowel changes highlighted. Here are all of the modal verb conjugations, with the irregular stem-vowel changes highlighted.
123Appendix H. (Continued)Slide 15 Word Order Word OrderSince the modal verb must be conjugated to agree with the subject, it takes the pivotal second position in statement sentences. Since the modal verb must be conjugated to agree with the subject, it takes the pivotal second position in statement sentences. The modal remains dependent upon its action verb, which is placed in the last position in all sentences, whether statements, questions or imperative commands. The modal remains dependent upon its action verb, which is placed in the last position in all sentences, whether statements, questions or imperative commands.Dependent Infinitive A German sentence only conjugates one verb per clause, therefore, if there is a modal verb in second position, the dependent action verb is found at the end of the clause in its infinitive form. Dependent Infinitive A German sentence only conjugates one verb per clause, therefore, if there is a modal verb in second position, the dependent action verb is found at the end of the clause in its infinitive form. A third important area to consider about modal verbs is their effect upon word order. Since the modal verb must be conjugated to agree with the subject, it takes the pivotal second position in statement sentences. The modal remains dependent upon its action verb, which is placed at the end of the clause, in all senten ces, whether statements, questions or imperative commands. A German sentence only conjugates one verb per clause, therefore, the dependent action verb is given in its infinitive form. It is then called a dependent infinitive. Slide 16 ( modals1 .html) Der Junge ein Stck Kuchen. isst wollen to want toessen to eat The boy eats a piece of cake.will Listen to the sample sentence, Der Junge isst ein Stck Kuchen. The action verb essen (to eat) is conjugated match the subject, Der Junge. To add a modal verb, such as wollen conjugate the modal verb to agree with the subject of the sentence. der Junge will Slide 17 ( modals2.html ) essen. Der Junge ein Stck Kuchen wollen to want towill isstThe boy wants to eat a piece of cake. Next, imagine the conjugated m odal verb placed into the verb second position. The conjugated modal displaces the action verb, which is pushed to the end of the sentence, where it reverts to infinitive form. Der Junge will ein Stck Kuchen essen.
124Appendix H. (Continued)Slide 18 ( modals3.html ) essen. Der Junge ein Stck Kuchen willThe boy wants to eat a piece of cake. The action verb is now called a dependent infinitive. wants to eat The action verb is now called a dependent infinitive. Der Junge will ein Stck Kuchen essen . Slide 19 ( modals4.html ) Das Brot man tglich frisch. kauft sollen (ought to)kaufen (to buy) One buys bread fresh daily.sollImagine the modal verb conjugated to agree with the subject of the sentence. In the sample sentence, Man kauft Brot tglich. The action verb kaufen (to buy) is conjugated to match the subject, man. To add a modal verb, such as sollen conjugate the modal verb to agree with the subject of the sentence. man soll Slide 20 ( modals5.html ) Das Brot man tglich frischkauft kaufen.One must buy bread fresh daily.sollen (ought to)muss soll kauft Now, imagine the conjugated modal verb muss placed into the verb second position. The conjugated modal displaces the action verb, kauft which is pushed to the end of the sentence, soll is now the main verb... Das Brot soll man tglich frisch kaufen. Slide 21 ( modals6.html ) The action verb is now called a dependent infinitive kaufen. Das Brot man tglich frisch sollOne bread fresh daily. should buy The action verb, kaufen is now the dependent infinitive. Das Brot soll man tglich frisch kaufen .
125Appendix H. (Continued)Slide 22 ( modals7.html ) Das Mdchen das Zimmer rumt mssen to have toaufrumen (to clean up) The girl cleans up the room. auf muss In the next sample sentence, Das Mdchen rumt das Zimmer auf. the action verb aufrumen (to clean up) is conjugated to match the subject, Das Mdchen. It has a separable prefix auf. Das Mdchen rumt das Zimmer auf. To add a modal verb, such as mssen conjugate the modal verb to agree with the subject of the sentence. Mdchen... muss Slide 23 ( modals8.html ) rumtDas Mdchen das Zimmer muss mssen to have to The girl must clean up the room. aufrumen rumt Now, imagine the conjugated modal verb muss placed into the verb second position, and the action verb moving to the end of the sentence. The conjugated modal displaces the root of the action verb, rumt which is pushed to the end of the sentence, and reunited with its own separated prefix, auf Das Mdchen muss das Zimmer aufrumen. Slide 24 ( modals9.html ) Das Mdchen das Zimmer muss The girl must clean up the room. aufrumen must clean up A separable-prefix verb will always become whole again in the infinitive form. Das Mdchen muss das Zimmer aufrumen . Slide 25 ( modals10.html ) Modal VerbsMeaning Conjugation Word Orderknnen > k a nn knnen > canDie Frau mchte eine Cola trinken. trinkt mchten click To summarize the topic of modal verbs : The three aspects to consider are: Meanings, Conjugations, and Word Order. This concludes the presentation on German modal verbs. Thank you for your attention.
126Appendix I.Score Sheet PRETEST Participant #____________ C. W O 1. __ __ 2. __ __ 3. __ __ A. 1. ___2. ___3. ___4. ___5. ___6. ___ B. 1. __ __ __ __ 2. __ __ __ __ 3. __ __ __ __ 4. __ __ __ __ 5. __ __ __ __ _____ M _____ C ___W ____O ____ M ____C ____W Total C. W O 1. __ __ 2. __ __ 3. __ __ A. 1. ___2. ___3. ___4. ___5. ___6. ___ B. 1. __ __ __ __ 2. __ __ __ __ 3. __ __ __ __ 4. __ __ __ __ 5. __ __ __ __ _____ M _____ C ___W ____O ____ M ____C ____W Total
127Appendix I. (Continued) Posttest Score Sheet C = conjugation CM modal verb inflection CI dependent infinitive CS separable prefix infinitive W = word order finite position O = word order infinitive position Participant #____________ CM CI CS W O 1. __ __ __ __ 2. __ __ __ __ __ 3. __ __ __ __ 4. __ __ __ __ __ 5. __ __ __ __ 6. __ __ __ __ 7. __ __ __ __ 8. __ __ __ __ 9. __ __ __ __ __ 10. __ __ __ __ __ 11. __ __ __ __ 12. __ __ __ __ 13. __ ____ __ 14. __ __ __ __ 15. __ __ __ __ __ 16. __ __ __ __ __ ___CM ___CI ___CS ____C ____W Total Rater #________
128Appendix J. Posttest Survey Responses Return to text Scale: 5 = very positive 4 = somewhat positive 3 = neutral 2 = somewhat negative 1 = very negative Animated Group Comments 1. Great presentation helped me a lot, truly!!! 5.0 2. I liked it. We should have this in cl ass! Very helpful in seeing and hearing! (Best way to learn!) 5.0 3. Complete and thorough plus reinfo rced by the written test afterwards. 5.0 4. I like this way of teaching better th an the one we are using in class now. It was very helpful and very organized. 4.8 5. The word order animation was very effective. 4.8 6. Well presented. Wish I could have taken notes! 4.8 7. It was very clear and I could und erstand and remember well from it. 4.8 8. I liked the presentation because it used visual techniques to help teach language. I also like the fact that pa tterns in language usage were shown. 4.8 9. The presentation was very helpful. I found it easier to understand with both verbal and visual instructions. 4.8 10. I liked the different use of colors and how the words moved to the end of the sentence when replaced with the new modal verb. 4.4 11. I liked the slides to the instruction. 4.0 12. Good job on presentation despite computer problems. 4.0 M = 4.7
129Appendix J. (Continued)Static Group Comments1. I like it very much. I l earn a lot from the visual presentation. The word order was presented very clearly and effectively. 5.02. I understood the information much better after watching the presentation.4.83. This should be done more often. 4.24. I enjoyed the presentation. Examples were good.4.25. I found it helpful.4.06. Well I thought it was very informative. The German voice that read was a bit monotone but otherwise I think it was grea t. I wish there was a recap at the end because I forgot how to spell mochte. Danke.4.07. Helpful. 3.68. It was understandable. 3.49. Pretty simple.3.210. Instruction was a little fast a nd confusing, wasn't too bad was understandable. 2.411. The presentation was boring -should have more interaction. Possible practice of conjugation before taki ng the posttest. The second half was unnessecary [sic] as to the fact we sa w the sentences and the "Word Order", construction in the firs t half. This should have been a second separate presentation for better clarifing. Very good power point. 1.8M = 3.7
130Appendix K. Comment Rating Scale Return to text a) I like this way of teaching better than the one we are using in class now. It was very helpful and very organized. 5 4 3 2 1 b) I like it very much. I learn a lot from the visual presentation. The word order was presented very clearly and effectively. 5 4 3 2 1 c) Well presented. Wish I could have taken notes! 5 4 3 2 1 d) I liked the different use of colors and how the words moved to the end of the sentence when replaced with the new modal verb. 5 4 3 2 1 e) Complete and thorough plus reinforced by the written test afterwards. 5 4 3 2 1 f) It was understandable. 5 4 3 2 1 g) This should be done more often. 5 4 3 2 1 h) I enjoyed the presentation. Examples were good. 5 4 3 2 1 i) I liked it. We should have this in class! Very helpful in seeing and hearing! (Best way to learn!) 5 4 3 2 1 j) The presentation was boring should ha ve more interaction. Possible practice of conjugation before taking the posttest. The s econd half was unnessecary [sic] as to the fact we saw the sentences and the "Word Order" construction in the first half. This should have been a second separate presen tation for better clarifing [sic]. Very good power point. 5 4 3 2 1 k) Great presentation helped me a lot, truly!!! 5 4 3 2 1 l) I liked the slides to the instruction. 5 4 3 2 1 m) The word order animation was very effective. 5 4 3 2 1 n) Good job on presentation despite computer problems. 5 4 3 2 1 o) Instruction was a little fast and confusing, wasn't too bad was understandable. 5 4 3 2 1 p) Helpful. 5 4 3 2 1 q) I liked the presentation because it used vi sual techniques to help teach language. I also like the fact that patterns in language usage were shown. 5 4 3 2 1 r) It was very clear and I could understand and remember well from it. 5 4 3 2 1 s) Well I thought it was very informative. The German voice that read was a bit monotone but otherwise I think it was great. I wish there was a recap at the end because I forgot how to spell mochte. Danke. 5 4 3 2 1 t) I understood the information much better after watching the presentation. 5 4 3 2 1 u) Pretty simple. 5 4 3 2 1 v) I found it helpful. 5 4 3 2 1 w) The presentation was very helpful. I fo und it easier to understand with both the verbal and visual instructions. 5 4 3 2 1 Please rate the following comments according to the scale at right.Very Somewhat Neutral Somewhat Very Positive Positive Negative Negative
About the Author Elizabeth Caplan received a Bachelo rs Degree in Theater Arts from the University of Northern Colorado in 1979 and an M.A. in Germanic Language and Literature from Washington University in 1989. She began teaching German at University of South Florida in 1995, when she entered USF's pilot Ph.D. program for Second Language Acquisition and Instructional Technology. While in the program at the University of South Florida, Ms. Caplan has been an instructor of German in the Department of World Language Education (WLE) in the College of Arts and Sciences She is currently editor of The Forum, the newsletter of Phi Sigma Iota (PSI), an international forei gn language honors society, and she has served as president of the local PSI chapter, Beta Zeta. She has also been the faculty liaison for the USF German club and the coordinator of WLE' s German TAs. She made several paper presentations at regional meetings of the Florida Foreign Language Association (FFLA) and at national meetings of the American Council on the Teaching of Foreign Languages (ACTFL) and the Second Language Acquisition Research Forum (SLRF). Ms. Caplan is currently an assistant pr ofessor of German at the University of Tennessee-Martin.
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001413345
007 cr mnu|||uuuuu
008 030703s2002 flu sbm s000|0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0000042
Caplan, Elizabeth A.
The effects of animated textual instruction on learners' written production of German modal verb sentences
h [electronic resource] /
by Elizabeth A. Caplan.
[Tampa, Fla.] :
University of South Florida,
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 130 pages.
ABSTRACT: This study investigated the effects of animation for a technology-assisted German grammar presentation on modal verbs. The premise was that many intangible concepts of dynamic grammar involve syntactic components that possess visuo-spatial characteristics. It was further speculated that these characteristics could be more effectively represented by animated versus static instructional presentations.The supposition that animation would lend pedagogical advantage was supported by dual coding theory (Paivio, 1971, 1990), which posits two functionally separate representational systems, the verbal and the nonverbal, with dynamic mental imagery residing solely in the nonverbal system. The strength of dually coded information is that it is represented in both subsystems and, due to referential associations that cross between the two, is more easily retained and recalled.Under two treatment conditions, 44 university students of beginning German (GER 101) received large-screen multimedia instruction concerning the meanings and conjugated forms of German modal auxiliary verbs, and the grammatical rules which govern sentence structure. The independent variable was the type of visualization: static or animated text. The dependent variables were participants' total test scores as well as their individual scores on each of two task types: conjugation and word order. In addition, a posttest survey asked participants for their opinions of the instructional treatments.Participants in both treatment groups achieved high scores on the posttest with no significant difference between them; however, the posttest survey showed that the groups did differ significantly in their opinions of the treatments, with those in the animated group reporting more positive reactions to the presentation. Detailed planning and lengthy preparation of both treatments may explain the high scores for both groups, and the elementary nature of the content may also account for the resulting ceiling effect. Animation should be studied further, especially with respect to more preliminary tasks, more complex tasks, as well as in concert with other aspects of multimedia, such as interactivity, user-control, practice, and feedback.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
Co-adviser: Feyten, Carine
Co-adviser: Kealy, William
Second language acquisition
Study and teaching
Computer-assisted instruction for foreign speakers.
second language acquisition.
computer assisted language learning.
t USF Electronic Theses and Dissertations.