|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
ORAL/WRITTEN CONTRAST OF MENTAL STATE REFERENCES IN OLDER CHILDREN by Jeanne E. Federico A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Department of Communicati on Sciences and Disorders College of Arts and Sciences University of South Florida Major Professor: Elaine R. Silliman, Ph.D. Committee Member: Ruth Huntley Bahr, Ph.D. Committee Member: Judith Becker Bryant, Ph. D. Committee Member: Sylvia F. Diehl, Ph.D. Date of Approval: March 11, 2005 Keywords: theory of mind, narratives, motivational, experiential, belief Copyright 2005 Jeanne E. Federico
i Table of Contents List of Tables iii List of Figures iv Abstract v Chapter One Literature Review, Purpose, and Resear ch Questions 1 Theory of Mind: A General Overview 3 The Theory-Theory View 3 The Modularity View 5 Social and Academic Relevance of a Theory of Mind 6 The Language Connection to Theory of Mind Development 8 Direction of the Relationship 8 Semantic Development of Mental States Meanings 9 Early development 9 Later development of a theory of mind 12 Later development of mental st ate references 13 Developmental Continuum: Oral-Lit erate Contrasts 15 Summary 19 Problem, Purpose, and Research Question 20 Chapter Two Method 22 Inclusion Criteria 22 Hearing screening 22 PPVT-III 22 Second-order false belief tasks 23 Participants 24 Materials 25 Second-order false belief tasks 25 Mental state verb priming task 25 Narrative tasks 26 Procedure 27 Second-order false belief tasks 28 Mental state verb priming task 28 Oral and written narrative tasks 30 Equipment 33 Data Reduction 33 Segmentation of oral and written narratives 33 Occurrence and classification of ment al state references 33
ii Qualitative analyses of belief verbs: Diversity, complexity, and depth 35 Examiner Agreement 36 Chapter Three Results 37 Data Analysis 37 T-unit Analysis 38 Analysis of Mental State References 40 Qualitative Analyses 44 Diversity 44 Complexity 46 Lexical depth 47 Summary of Findings 48 Chapter Four Discussion 50 Methodological Issues 50 Sample Representativeness 50 Influence of the Priming Task, Instructions, and Elicitation Method on Performance 51 Influence of the priming task 51 Influence of the instructions and elicitation method 52 The Age Interval 53 Conceptual Issues 54 Lexical Diversity and Complexity 54 Lexical Depth 55 Future Research Directions 56 Conclusion 58 References 59 Appendices 67 Appendix A Table 2. Age, Gender, Ethnicity, and Results of Inclusion Criteria 68 Appendix B Second-Order False Belief Tasks 70 Appendix C Score Sheet for False Belief Tasks 73 Appendix D List of Priming Verbs 79 Appendix E Episode Analysis of Frog Goes to Dinner 80 Appendix F Description of the Eight Still Frames 84 Appendix G Consent Form 85 Appendix H Eight Still Frames 88 Appendix I Script for Priming Task and Narrative Task 92
iii List of Tables Table 1. Descriptive Data for 9 and 11 year olds in the Oral and Written Domains 39 Table 2. Diversity of Belief Verbs Produced by the 9 and 11 year old Participants as Displayed in Order of Frequency for th e Oral and Written Narratives 45 Table 3. Most Frequent Use and More Comp lex Use of Belief Verbs as Produced In the Oral and Written Narratives 46
iv List of Figures Figure 1. Frequency of Use of Mental State References by Age 41 Figure 2. Differences Noted by Reference T ype and Modality 43
v Oral/Written Contrast of Mental St ate References in Older Children Jeanne E. Federico ABSTRACT The purpose of this study was to determin e differences in the use of mental state references by typically developing 9and 11year-old children. Follo wing a priming task that emphasized distinctions between physical and mental acts, ch ildren watched an 11minute textless video and then were asked to generate an oral and a written story that focused on the mental states of the multiple characters. Narratives were transcribed, and all mental state references were classified into motivational, experiential, and belief categories. Specific mental state references we re also analyzed to determine levels of semantic complexity. The study attempted to answer: 1) whether 9and 11-year-old typically developing children differed in their ability to use more complex mental state references and 2) whether this ability varied as a function of the oral versus the written modality. The sample consisted of 26 children, ages 9;0-9;11 (15 females, 11 males), and 24 children, ages 11;00-11;11 (14 females, 10 male s). The total sample (N=50) consisted of 32 Caucasian children, 15 African-American ch ildren, 2 Hispanic children, and 1 AsianAmerican. All children were selected from one urban elementary school located in West Central Florida, were from monolingual, E nglish-speaking homes, and were speakers of Standard American English.
vi A statistical analysis was conducted via a 3-way MANOVA, specifically, 2 (age 9 vs. age 11) x 4 (mental state categories) x 2 (oral versus written modality). In addition, a qualitative analysis was completed to comp are the frequency, complexity, and lexical depth of different types of mental state verbs produced in the oral a nd written narratives. Results revealed an unexpected lack of variability in age in the production of mental state references, possibl y due to the close age interval of participants. However, there was substantial variability in production across the three mental state categories for both age groups. Across the two modalities, results provided evidence for increasing lexical diversity, complexity, and depth in th e production of mental state references by the older children, especially fo r belief verbs. Future research directions are presented to investigate relationships between more adva nced theory of mind development and the lexical depth of mental state references.
1 Chapter 1 Literature Review, Purpose, and Research Questions In the last 15 years, childrens develo pment of a theory of mind has become an important research area, in cluding how theory of mind relates to autism spectrum disorder (e.g., Baron-Cohe n, 1995; Happe, 1994, 1995; Silliman et al., 2003). Simply put, theory of mind is an understanding of the mi nd of self and others as an active agent. Arriving at some sort of understanding of the mind is an important accomplishment for children, because it is centra l to understanding the social world. Children learn that internal mental states, such as beliefs and de sires, influence the actions of themselves and others. They learn to differentiate between intended and acciden tal behavior, between wishes and reality, and between truth and deception. A concept of mind leads to understanding human behavior, which is an im portant tool for navi gating the intricacies of human communication, includi ng the language of schooling. Theory of mind is often described as the understanding of ones own and others mental states, such as thinking and knowing (German & Leslie, 2000). It is also described as awareness of ones own and others emo tions, desires, and in tentions (Astington, 2001). This mental landscape, which Brune r (1986) calls the landscape of consciousness, gives meaning to actions and ev ents. To assign meaning to the actions of others, children must consider the mental processes involved. When children become aware of this mentalism, they have ac quired a basic theory of mind. Beyond a basic framework of theory of mind, children also de velop more specific conceptual structures and more complex levels of meaning for cognitive processes over the course of their
2 development (Booth & Hall, 1995; Wellman, 1 990). In effect, theory of mind is how children think about thinking. Most studies on childrens development of theory of mind have concentrated on the preschool and early elementary age year s. In addition, few studies have examined changes in the complexity of childrens use of mental state refere nces that denote the understanding of motivation, acti ons, and beliefs. Finally, mo st studies concerned with mental state references have focused on child rens performance in the oral domain, with minimal attention paid to how the modality of expression (ora l versus written) affects the complexity of mental state terms. This review of the literature is organi zed to address the relationship between theory of mind development and childrens de veloping use of mental state references. The first section presents a general overview of theory of mind fram eworks, specifically two theoretical perspectives th at attempt to account for development of a theory of mind. The social and academic relevance of a theory of mind is also addressed. In the next section, the linguistic aspect of theory of mind is discussed, focusing on the relationship of mental state referencing to theory of mind development. Earlier phases of development are contrasted with more advanced phases with an emphasis on cognitive verbs that express ways of knowing. The limited research on mental state development as assessed through writing is also overvi ewed, including the few studies with children having a language learning disability (LLD). The fina l section presents the purpose of this study, which investigates similarities and differences in the mental state referencing of 9-and11-year children in an oral -written contrast paradigm.
3 Theory of Mind: A General Overview Theory of mind is necessary in order to have a psychological concept of others (Wellman, Phillips, & Rodriguez, 2000). In othe r words, theory of mind allows one to see people in terms of their psychological states, in addition to their physical entities. It is the tool by which children attribute mental states to people, such as their thoughts, desires, beliefs, and intentions. Most critically, a well-developed co ncept of mind leads to the understanding that people are mo tivated by their desires, inte ntions, beliefs, and emotions (Wellman, 1990). Additionally, children with advancing conc epts of mind develop the insight that their own or anot her persons beliefs may not be true, and that a person can have a false belief (Astington, Harris, & Olson, 1988). As a re sult, childrens theory of mind underlies their understanding of human behavior. There are two major contrasting views of theory of mind development in children, the constructivist, or theory -theory, model and the modularity, or mechanism, view. The constructivist, or th eory-theory, model suggests that children acquire knowledge about the mind between the ages of 2 and 5 years old by means of actively constructing and refining a scientific theory about mental states. The modu larity, or mechanism, view purports that children are born with an innate mechanism, or specialized module, that leads to the development of theory of mi nd. Each of these mentalistic theories is reviewed briefly. The Theory-Theory View Gopnik and Wellman (1982) proposed that childrens understa nding of mental states developed through a process of theory construc tion and modification. Children develop theories about the mind and use th ese theories to explain human behavior.
4 Others, such as Astington (1998a), view this process similarly. From ages 2 to 5 years, children develop concepts of how things and people work. To understand another persons behavior, children must become aware of the fact that people act based on their thoughts, desires, and feelings Children must become aware of this mental processing before they can interpret the behavior of other people. In a similar theory-theory vein, Pern er (1995) considers theory of mind acquisition as an important intellectual change that begins around 4 years of age, when children usually master an explicit concept of false belief. This e xplicit understanding, whereby children predict what a character will do based on specific knowledge of the characters false belief, is typically assessed with a task that invol ves a displaced object scenario. For example, Wimmer and Perner (19 83) developed the Maxi task in which the character, Maxi, places a piece of chocolate in the kitchen cupboard and leaves the room to play. While Maxi is out of the room his mother moves the chocolate from the cupboard to a drawer. The child is then asked where Maxi will look for the chocolate when he returns. The correct response (in the drawer) depends on the childs ability to understand that Maxis behavior depends on his beliefs about object location rather than on the real situation (the displa ced object). In other words, re lative to the formation of an initial or first order concept of mind (Bartsch & Wellman, 1 995), the child must be able to predict that, in this kind of situation, a nother can accept a false belief as true. Perner (1995) refers to this insight as the unde rstanding of aboutness, or the realization that a proposition is about a world, whether real or pretend, a nd must be evaluated against that world in order to determine if it is true or false. For example, when presented with the statement Maxi believes that the chocolate is in the original cupboard, the proposition
5 the chocolate is in the original cupboard must be evaluated agai nst Maxis world in order for the child to determine if Maxis belief is true or false. One speculation is that children acquire this primary knowledge about the contents of others minds through an ongoing process of conceptual discovery. The Modularity View A competing theory is the modularity view. For example, German and Leslie (2000) reject the theory-theory view and, instead, propose a mechanism, or modular, view, which is borrowed from the big modul arity strand of generative linguistics. A module is designed for the specialized processi ng of an input, in th is case, the cognitive processing of the minds contents. Fodors (1981, cited in de Villiers, 2003, p. 428) concep t of modularity is that, processing is fast, mandatory, not accessible to introspection, and most importantly, encapsulated (i.e., impervious to the influe nce of general knowle dge or other cognitive processes). In terms of a theory of mind m odule, the speculation is that children are born with innate concepts that allow them to at tend to mental states (German & Leslie, 2000). These specialized attentional mechanisms then allow learning about me ntal states to take place. The ability to attend to beliefs beco mes more flexible over time and thereby increases success with false belief problems. Eventually, children develop competent reasoning skills when assessing mental states. The modularity concept has been seriously criticized, most recently by Thelen and Bate s (2003), who make the case that modularity is not a developmental theory about how mind and language emerge continuously over time. Instead, modularity is a biologically deterministic vi ew that fails to explain
6 adequately cognitive or language developmen t, much less individual differences in developmental trajectories. Regardless of the theoretical perspective, it appears that, at least for Western cultures, middle-class children typically deve lop a concept of mind at about the same time. Wellman, Cross, and Watson (2001) c onducted a meta-analysis of 178 separate child studies on theory-of-mind development. Their findings indicated that childrens understanding of belief, and, therefore, th eir understanding of th e mind, underwent great conceptual change in the preschool years. Wellm an et al. (2001) also argued against early competence explanations, claiming, as do German and Leslie (2001), that apparent basic theory of mind development between the ages of 3 and 5 years of age is masked by overly difficult tasks variables. Rather, Well man et al. found that the development of these insights was consistently evident acr oss different tasks and different Western cultures, suggesting that such a conception of the minds contents is a natural and universal way for children to understa nd peoples intentions and actions. From a social interactional perspective, little is know n about the contribution of variations in family socialization practices to individual differences in theory of mind development (Lillard, 1998). However, there is some information about the relevance of a well-integrated theory of mind fo r social and academic success. Social and Academic Relevance of a Theory of Mind Social interaction is dependent on theory of mind because childrens social lives depend on perspective taking or th e shared reciprocity of thou ghts, desires, feelings, and plans. For example, Gopnik and Astington (1988) see social relevance in the ability of children to attribute false beliefs to themse lves and others because this ability is
7 necessary to understand what dr ives the otherwise incomprehe nsible behavior of another person. These authors believe that children ac tually learn to unders tand changes in their own mental states through eval uation of differences between their own mental states and those of others. To illustrate, it may be difficult for children to examine their own contradictions, but it is not so easy to overlook contrary claims imposed by an older sibling. Moreover, how children think about thi nking and communicating even affects their critical thinking skills a nd, therefore, their success in sc hool in a variety of content areas (Astington, 1998b). Critical thinking depend s on the ability to consider ones own beliefs, recognize ones false beliefs, and take anothers perspectiv e, including authors perspectives. When children are exposed to literature during th eir school-age years, development of perspective taking is a neces sary element in order to understand the intentions of characters and historical actions of individuals. The abil ity to coordinate the multiple perspectives of several character s in a book is speculated as leading to successful reading comprehension (Dona hue & Foster, 2004). Development of perspective taking also leads to the evolving social competence that contributes to academic success (Astington, 2003). In sum, theory of mind and perspective taking are intertwined, as both are central to achieving the shared community of minds that characterizes the understanding of ones self and others (Nelson, 2004). Because comm unication is socia l interaction that activates mental processes of perspec tive taking (MacWhinney, in press, p. 27), perspective taking and shifting are central to communication. Th erefore, there is evidence
8 that an increasingly coordina ted theory of mind is essent ial for childrens successful cognitive, social, and language development. An important question concerns the ro le of language in theory of mind development. The majority of research on li nkages has concentrated on the contributions of syntactic development to perspective taking. Less work has examined the semantic domain, specifically, the development and elaboration of mental state references. The Language Connection to Theory of Mind Development Direction of the Relationship An initial study on theory of mind-language connections was conducted by Bretherton and Beeghly (1982). A total of 30 middle-class mothers and their children. were seen at home and in laboratory sessi ons at 10, 13, 20, and 28 months of age. The mothers were asked to listen to the childs speech and to note the us e of 73 internal-state words, divided into categories of pe rception, physiology, aff ect, volition/ability, cognition, and moral judgment/obligation. Findin gs led to the interp retation that young childrens mental state development and language development were interrelated, emerging with the onset of communicative intentions. A second perspective is that language competence leads to theory of mind development. For example, based on their study of children from 8to 11-years-old, Astington and Jenkins (1999) argued that ch ildrens theory of mind development was predicted by their general syntactic comp etence. Astington (2001) also stressed the importance of language in the development of false-belief understandi ng. She stated that false belief is central to theory-of-mind develo pment, due to the fact that it is a strong indicator of mentalistic understanding.
9 A third perspective is that theory of mind development is made public through language, and childrens everyda y conversations demonstrate their awareness of other peoples mental states (Wellman et al., 2000). However, this public display should not be confused with language as the generator of conceptual change in theory of mind development. Wellman et al. (2001) argued that changes in the conceptual knowledge base drive the language of beliefs and false beli efs rather than linguistic elements, such as the ability to produce tensed complements, se rving as a bootstrap for beliefs about the mind. de Villiers and de Villier s (2000) made the opposite case. Because sentences with mental state verbs often require complements, children must possess the ability to engage in false complement processing, such as the processing of Dad thought that he forgot the birthday present (when the child knows he did not forget) (Silliman et al., 2003, p. 239) before they can provide an adequate mental explana tion for two contradictory perspectives. Regardless of the perspectiv e taken, childrens use of mental state references provides a fairly reliable indicator of their le vel of theory of mind development. Children begin to demonstrate acquisiti on of a theory of mind around ag e 2 years, as evidenced by their use of mental state references indi cating their own desire s (e.g., want). They continue to demonstrate theory of mind deve lopment throughout the school-age years, as evidenced by an increasingly varied and comp lex vocabulary of mental state references. Semantic Development of Mental States Meanings Early development Bartsch and Wellman (1995) acknowledge that, when children learn to talk, their conversations provide a window into their understanding of peoples minds. Although individual children will differ in their onset of referring to
10 mental states, Bartsch and Wellman found that childrens talk about the mind developed in three phases. The earliest phase begins around the age of 2 years, when very young children talk about desires, mainly wants and likes, in many different contexts, such as their desires for objects and actions. Even 2-year-olds do not limit the expression of desires to items, such as toys and food. Their desire expressions also include action (e.g., the desire to run) and changes in st ate (e.g., the desire to be asleep). The second phase, beginning around the age of 3 years, occurs when children begin to talk about thoughts a nd beliefs, as well as desires. Even when children first begin to talk about these more cognitive concepts, Bartsch and Wellman (1995) found that their mental state expre ssions were not confined to ta lking of thoughts alone, whether real or imaginary. They also talked about beliefs (a thought that something is so) and false beliefs (a thought that something is so, when it is not). However, children in this phase continue to talk about desires more frequently th an thoughts and beliefs, and, typically, their explanations for a persons ac tions are based on their interpretation of a persons desires. In the third phase, beginning around the age of 4 years, thought s and beliefs begin to play a larger role in childrens understa nding of relationships between the intentions and actions of others (Bartsch & Wellman, 1995). They refer more often to thoughts and beliefs (e.g., think, know), a nd they take in account a pe rsons beliefs, both true and false, to explain their actions. In this phase, children appear to understand that people not only have thoughts and beliefs, but also that thoughts and beliefs are important elements in explaining why people do things. For ex ample, children understand that peoples
11 desires are shaped by their beliefs, and that false beliefs, rather than the reality of a situation, can determine a persons actions. Bretherton and Beeghly (1982) also found th at an emerging ability to attribute internal states to ones self and others b ecomes evident with the onset of communicative intention. Based on their natura listically obtained data, they discovered that children begin to speak about mental states late in their second year, and th at this skill becomes more fully developed in the third year. A lthough the children in their study tended to speak of their own mental states before t hose of other people, only a small minority of mental state words were produced that referr ed exclusively to themselves. In fact, the results indicated that 2-year-old children in terpreted their own and other peoples mental states, commented on their own or others e xpected or past expe riences, and discussed how their own or another persons mental state came about or might be changed. Threeyear-old children demonstrated the ability to analyze the goals and motives of others, as these related to the childs own goals and motives. Astington (1999) also found that children begi n to express mental states almost as soon as they begin to talk. Children used wo rds to anticipate and de scribe their own and others actions. They also produced references like will and gonna to refer to their own and others future actions. According to Astington, childrens acquisition of these references leads to their developing understand ing of intention, which then influences and increases their understanding of intentionality. Moreover, Jenkins and Astington (1996) found a strong positive correlation between false-belief understanding and general syntactic ability in children between 2 years 11 months and 5 years 5 months.
12 While middle-class preschool age children are capable of talking about mental states and generally use the mental state refere nces that are used most often in the home (Jenkins, Turrell, Kogushi, Lollis, & Ross, 2003; Meins et al., 2002; Ruffman, Slade, & Crowe, 2002), certain categories of mental stat e verbs appear to be more cognitively and semantically complex for young children (N ixon, 2005). For example, Nixon found that mental state verbs encoding degrees of certainty (e.g., know ) were easier for 4 year olds to produce than mental state verbs enc oding degrees of uncertainty, such as wish, think, and consider Nixon (2005) concluded that whether or not verbs referring to the certaintyuncertainty continuum were used depended on their inherent semantic, cognitive, and syntactic complexity (p. 35). Later development of a theory of mind Wellman (1990) found evidence of additional development of theory of mind in the years beyond age 6 years. Children get progressively better at considering other peoples thinking. Their proficiency at identifying other peoples beliefs and desires increases, as well as the ability to predict and explain others actions on th e basis of their beliefs and desires. They move from a more concrete level to a more inferential leve l of interpretation of the actions of others. For example, younger children see themselves a nd others in concrete categories such as their address and physical appearance, while a dolescents use more ab stract descriptions such as personal beliefs and motivation. In a ddition, there is a subs tantial developmental change after age 7 years in the use of tra it explanations to justify the actions of themselves and others (e.g., hes stubborn, she s shy). During this time, children also develop a stronger and more insightful self -concept, which include s the concept that ones self is reflected by ones mind. Wellm an (1990) believed that each phase of
13 theory of mind development built on the achievements of the previous phase, and that more complex development of a theory of mind is probable beyond age 6 years. Older children also become better at implicitly understanding the concept of someone thinking about thinking, specificall y, when others false beliefs are not explicitly stated. Perner and Wimmer (1985) identified this further development in conception as the acquisition of second-or der beliefs, and Flavell (1999) called it metacognition, which includes knowledge abou t the nature of peopl e as cognizers (p. 21). Later development of mental state references Despite Wellmans (1990) prediction that theory of mind development becomes more sophisticated afte r the age of 6 years, there have been few studies inve stigating advances. Si milarly, childrens increasingly sophisticated development of me ntal state references during the school-age years has not been pursued in other than a limited way, an important omission given Nixons (2005) findings that the production of particular kinds of mental state verbs are influenced by their semantic and cognitiv e complexity. Two studies involving older children are of interest here. One study, conducted by Schwanenflugel, Fabricius, and Noyes (1996), examined the continuing development of theory of mi nd in 8 to 11 year old children through their changes in the organization of cognitive or mental state verbs of knowing, such as thinking, guessing, and knowing Results indicated that ol der children constructed a semantic organization of mental verbs accord ing to the certainty aspects of mental processes, and that this semantic organiza tion became more complex and fine-tuned with further development. Older children and adults were remarkably similar in their semantic
14 organization of these verbs, and, consistent w ith Nixons (2005) results with 4 year olds, older children placed greater emphasis on the certainty aspects of mental activity, as evidenced by their increased unders tanding of the mental state verbs know, think, and guess For example, know, think, and remembe r were ordered as more certain than guess, estimate, and compare. In contra st, preschool-age children demonstrated a more absolute, or true/false, view of cer tainty. Schwanenflugel et al. (1996) speculated that the conceptual and semantic knowledge of older children evolve d to a degrees of certainty perspective due to subjectively expe rienced feelings of uncertainty. The more that children have experiences with uncertain ty, the more that they come to understand this concept as relative. The second study was conducted by Booth a nd Hall (1995), who also investigated the understanding of th e cognitive verb know in 3-, 6-, 9-, and 12-year-olds. Booth and Hall compared younger and older childrens und erstanding of six different levels of meaning of the cognitive verb know : 1. Perception (e.g. I heard your story); 2. Recognition (e.g., I know that face); 3. Recall (e.g., I know his phone number); 4. Understanding (e.g., I know why he did that ); 5. Metacognition (e.g., Pretending can be fun); and 6. Evaluation (e.g., He guessed the answer, but I know it). Children were presented video-taped skits involving inter actions between two hand-held puppets in one scenario and audio-taped stor ies in another scenario. Th e skits and stories required answers for the six levels of meaning for the cognitive know Results showed that the depth of childrens knowledge about the multiple meanings of know increased with age due to concep tual reorganization. For example, hierarchical knowledge of the cognitive verb know became more differentiated with
15 development, with 3-year-olds having the most restricted levels of meanings and 12-yearolds having the most flexible levels of meaning. Younger ch ildren encountered difficulty differentiating between mental acts, such as knowing, and physical acts, such as doing In contrast, older children realized that to know something, one had to mentally manipulate information. Consistent with these patterns, 3-year-olds exhibite d understanding of the least number of levels of meaning, 6and 9-year-olds demonstrated understanding of a moderate number of levels of meaning, and 12-year-olds had the most distinctions among the six levels of meaning. This pattern resulte d in a more differentiated and hierarchical mental state verb lexicon in older children. Developmental Continuum: Oral-Literate Contrasts Children develop language skills over an oral-literate continuum. They continually add new vocabulary words and a pply new language concepts (e.g., semantic, syntactic, and pragmatic concep ts) to their oral language knowledge, and they initially write like they talk. However, during the early school years, f eatures of oral and written language converge write like talk combines with talk like books (Rubin, 1987). Beginning around age 9 years, writing beco mes increasingly differentiated from speaking, with more complex forms of langua ge expressed in writing than in speech (Scott, 1999). Similarly, childrens mental state refere nces appear to emerge over a general developmental continuum. Based on the litera ture, motivational references (e.g., want) appear first, experiential re ferences (e.g., see) appear next and belief references that index degrees of uncertainty (e.g., think) a ppear last. In additi on, these mental state references demonstrate greater lexical depth over time in the evol ution of related and
16 unrelated meanings (Nixon, 2005). For example, a child may say, I see a dog at age 3 years and I see what you mean at age 10 years. McGregor (2004) refers to this proce ss of increasing lexical depth as slow mapping. Children develop initial maps of word meanings. This process is known as fast mapping, or incidental learning, based on childre ns implicit exposure to a multitude of word meanings. With multiple experiences with these meanings over time, including the critic al influences of reading and writing, children continue to refine those meanings. One slow mapping process that McGregor (2004) desc ribes is semantic elaboration, which allows words to assume multiple meanings (e.g., dogs bark, but bark also grows on trees). A second slow ma pping process is the development of a categorically organized semantic network, wh ich affords greater ease in lexical access because meanings are related in a superordinat e-subordinate structure. Figure 1 displays this continuum of development for mental st ate categories in their scope or breadth and depth.
17 Figure 1. Proposed developmental continuum with examples for three categories of mental state references in terms of their lexical breadth and depth Motivational References Experiential References Belief References want look, see, heard think tried mad, sad notice said found guess had to X stared decide meant to X disgusted realize aggravated Despite the growing interest in the sema ntic development of school-age children, only three studies have explicitly used the oral-written contrast design. Two investigations, Scott and Windsor (2000) a nd Fey, Catts, Proctor-Williams, Tomblin, and Zhang (2004), employed methodology in which child rens oral narratives were contrasted with their written narrative s. Scott and Windsor (2000) assessed the general language performance of school-age ch ildren with and without la nguage learning disabilities through spoken and written narratives. The primary focus was morphosyntactic integrity of inflectional morphemes. Fey et al. (2004) al so compared the oral and written narratives of typically developing child ren and children with language impairment for lexical and grammatical complexity, in addition to stor y content, organization, and overall quality. As previously noted, mental state references in either oral or writt en narratives were not examined in either study. Lexical De p th Lexical Breadth
18 Only a Swedish study of children, adol escents, and young adults, ages 9, 12, 15, and 20 years (Nordqvist, 2001) explicitly exam ined patterns of mental state verb production in oral and written narratives. A wordless picture book frequently used in cross-linguistic research on na rratives served as the elicit ation method for both the oral and written modes. Nordqvist established two categories of cognitive verbs. The first was cognitive verbs having a syntactic framing, or complementizer, function, e.g., The boy thought that the frog was lonely. The second categor y included four types of verbs that did not have the complement, or syntactic framing, function: a) those that expressed motivation (such as intends); b) those th at conveyed reflection, as in She is thinking about the frog or They wondered about the frog; c) t hose that communicated experiences like The boy thinks he hears something; and d) those that expressed having an opinion or belief, such as He thinks the frog is lost. All four age groups produced a greater number of non-framing verbs. In addition, a general trend for the three oldest age groups was that al l cognitive verbs were expre ssed to a greater extent in writing than in speaking (Nordqvist, 2001, p. 263). In general, a small number of cognitiv e verbs were generated across age groups ( N =132; n = 53 spoken, n = 79 written). Three variables may have influenced the frequency of cognitive verb production, whet her framed or unframed. The first is the relatively small sample size of each group (N = 15). Second, a wordless picture book designed for younger children may not be as app ealing to older childr en, adolescents, and young adults. Finally, the instructions to th e participants did not emphasize cognitive states. Participants were informed only to look first through the picture book and Then tell what is happening in the pictures (Nordqvist, 2001, p. 120).
19 Summary As noted earlier, the majority of previ ous studies on theory of mind development have focused on children from ages 2 to 5 y ears. The first studies of theory of mind investigated development in young children. Le slie (1987) found that children typically engage in pretend play by two years of age, adopting the perspectives of other people or characters. Around 2 years of age, most child ren use the mental state terms of want and know in everyday speech (Bartsch & Wellman, 1995). By age 3 years, children have developed some concept of desires and intentions (Wellman, 1990). Between the ages of 2 and 4 years, children begin to r eason about other peoples beliefs, can predict or explain actions based on false beliefs, a nd can actively deceive other people (Bartsch & Wellman, 1995). Most recent studies have consisted mainly of two types of task s: false belief tasks and tasks eliciting the use of mental stat e verbs. These studies have focused on young childrens awareness that someone can have a false belief (Astington & Jenkins, 1999) and their understanding of mental states as evidenced by their use of mental state verbs, such as think and know (Bartsch & We llman, 1995). Mental state verbs are evidence of childrens awareness of multiple perspectiv es, indicating other peoples states of mind and emotions (Dunn, 1999), other peoples intent ions and desires (Wellman et al., 2000), and other peoples knowledge, perceptions, a nd beliefs (Schwanenflugel et al., 1996). Minimal research has examined the know ledge and organization of specific mental state verbs in pr eadolescents, such as think, know, and guess (Booth & Hall, 1995; Nordqvist, 2001; Schwanenfluge l et al., 1996). Evidence of more complex use of these
20 belief references over time w ould suggest the continued deve lopment of a theory of mind throughout the school years. Problem, Purpose, and Research Question Many researchers have different theori es on theory of mind. It has been speculated that theory of mind plays a role in social interac tion skills and academic success. However, the majority of research to date has focused on the development of theory of mind in younger children. The empiri cal focus has been th e initial acquisition of theory of mind, neglecting the further de velopment of theory of mind in the later elementary years. The purpose of this study was to assess the continuing development of theory of mind in 9and 11-year old children as suggest ed by the use of more complex mental state terms. Oral and written narratives were contra sted to assess the use and complexity of mental state references in both modes. Oral narratives are stories expressed verbally, and written narratives are stories expressed in writing. Many researchers have been interested in what childrens use of mental states terms reveals about childrens developing theory of mind. However, these production studies have focused solely on the oral domain (Bretherton & Beeghly, 1982; Hyter & Westby, 1996; Wellman et al., 2000). Little da ta exist on the con tinuing acquisition and complexity of the mental state lexicon of older children or the influence of narrative modalities (i.e., oral vs. written) on the comple xity of mental state references. Previous research has examined the organization and knowledge of mental state verbs in older children (Booth & Hall, 1995; Schwanenflugel et al., 1996), but ha s not compared the complexity of the mental state verb le xicon across oral and written narratives. Only the
21 Swedish study by Nordqvist (2001) examined me ntal state references in older children using an oral-written contrast design. Since there is ev idence that children develop language skills over an oral-l iterate continuum, a difference would be expected between the complexities of mental state verb s in oral versus written narratives. The overall question was: Is acquiring th eory of mind a developmental process that continues during the sc hool age years as evidenced by the use of more complex mental state verbs? Two questions were ad dressed: 1) whether 9and 11-year-old typically developing children differed in their ability to use more complex mental state verbs, and 2) whether this ability differed as a function of the oral versus written modality.
22 Chapter 2 Method Inclusion Criteria All participants were selected from on e urban elementary school located in West Central Florida. To be included in the study, 9to 11-year-old ch ildren were typically developing in accord with four inclusion crite ria. These were: (1) not eligible or not previously eligible for special education or related services; (2) hearing sensitivity within normal limits; (3) age-appropriate vocabular y recognition according to performance on the Peabody Picture Vocabulary Test3rd edition (PPVT-III) (Dunn & Dunn, 1997); and (4) passing of a logical and/or a social second order false be lief task (Silliman, Diehl, Hnath-Chisolm, Bahr, Zenko, & Friedman, 2003). Hearing screening. Childrens hearing was screen ed before the testing session began. A Beltone Audio Scout audiometer calibrated to ANSI 1989 was used for an audiometric screening. Participants responde d to screening levels of 20dB HL at 1000, 2000, and 4000 Hz in order to pass the hearing screening. PPVT-III. The PPVT-III (Dunn & Dunn, 1997) was ad ministered as a measure of general vocabulary familiarity. The start item for the 9-year-old participants was item #73, set 7, while the start item for the 11-year -olds was item #85, set 8. According to the PPVT-III technical manual, the start item is the first item in the appropriate set of test questions designated for the test takers age. Th ese starting items were derived so that test takers in the middle 50% of any age group w ould meet the basal se t criterion. The basal set is the lowest set of items administered containing one or no errors. The ceiling set is the highest set of items administered cont aining eight or more errors. Testing was
23 discontinued as soon as the ceiling set was established. To be included in the study, a participants score needed to be within + 1 standard deviation of the mean for each age group ( M = 100, SD = + 15). Table 2 in Appendix A contai ns the PPVT-III scores for all participants. Second-order false belief tasks. It is assumed that a more advanced theory of mind is associated with more complex reference to mental states in both oral and written narratives. Therefore, two second-order, implic it, false belief tasks (Silliman et al., 2003), a logical inferencing task and a social in ferencing task, were used to assess more advanced concepts of mind characteristic of the 9-year to 11-year age span. To be included in the study, participants had to pa ss at least one of the two tasks. Scores obtained on the false belief tasks are also presented in Table 2 in Appendix A. The first implicit false belief task required logical inferencing and was adapted from Sullivan, Zaitchik, and Tager-Flusberg (1994) (see Appendix B). In this logical second-order false belief task, the false be liefs about characters actions were not explicitly stated. For example, a father buys his daughter a birthday present other than what she is expecting and hides the present; however, unknown to the father, the daughter saw what he got her for her birthday before the father gave her the gift. The father remains unknowing when Grandma asks him what he thinks the daughter believes he got her for her birthday. The implicit false beli ef question would then be: What would Dad say (to Grandma)? Silliman et al. (2003) de fined this task as one requiring logical inferencing because it is primarily premised on cause-effect inferencing. Cause-effect inferences necessitate that the child predict the missing in formation, in the case of the
24 above example, the result of th e explicitly stated cause or the effects of the conflicting beliefs of the father and da ughter that were the product of their different actions. The second implicit false belief task, a soci al inferencing task, focuses on more inner-directed perspectives that are motiv ated by affective states (see Appendix B). Silliman et al. (2003) describe this mode of reasoning as social reasoning, which represents a category of feeling/attitu de inferences (Johnson & von Hoff Johnson, 1986). For example, a child does not want a surprise birthday party because he becomes embarrassed when he is the center of attention; instead, he wants to go to a baseball game for his birthday. However, his father is plan ning a surprise birthday party because he does not know that his son gets easily discomfited. The father thinks that a surprise birthday party would make his son happy, even when the son tells the father he wants to go to a baseball game for his birthday. The fath er remains unknowing about his sons selfconsciousness when Grandma asks the father how her grandson will feel about having a surprise birthday party. The second-order imp licit false belief question would be: What would Dad say (to Grandma)? In this task, the beliefs reconstructed derive from the father and sons inner-directe d feelings and attitudes that motivate their actions. Participants A total of 55 children were assessed. Of the 55, 50 met th e inclusion criteria. The final sample consisted of 26 children ages 9;0-9;11 ( M = 9;5, Range 9;0-9;11) and 24 children ages 11;00-11;11 ( M = 11;3, Range 11;0-11;9). Of the 26 children in the 9-yearold group, 15 were female and 11 were ma le, 18 were Caucasian, 7 were AfricanAmerican, and 1 was Hispanic. Of the 24 ch ildren in the 11-year-old group, 14 were female and 10 were male, 14 were Caucasian, 8 were African-American, 1 was Hispanic,
25 and 1 was Asian-American. All children were from monolingual, English-speaking homes and speakers of Standard American English, according to classroom teachers. Approximately 70% of the school population wa s eligible for the federal lunch program, and the school received Title I funds. The 9year-old children were in third and fourth grade classes, and the 11-year-old children were in fifth grade classes. Table 2 in Appendix A contains a summary of the de mographic data for the 50 participants. Materials Second-order false belief tasks Both second order false belief tasks were displayed in a binder containi ng illustrations depicting the two different scenarios and scripted questions for both scenarios. The logical inferencing scenario consisted of 15 colored illustrations, and the social infe rencing scenario consisted of 12 colored illustrations. Protocol sheets were used to reco rd all responses (see Appendix C for sample). Mental state verb priming task A mental state and communication verb priming task was given first. This task was admini stered based on the assumption that children would show a preference for using mental state verbs and communication verbs when these verbs were primed prior to the narrativ e task, which comprise d the major portion of the study. The priming task consisted of two parts. In the first part, 16 3x5 cards with verbs printed on the front of each card were presen ted. A total of 8 cards had action verbs (e.g., run, play, jump ) printed on the front, and 8 cards had mental state verbs (e.g., think, know, guess ) printed on the front. The second part of the prim ing task consisted of an additional 16 3x5 cards with verbs printed on the front of each card. A total of 8 cards
26 had action verbs ( climb, crawl, push) printed on the front, and 8 cards had communication verbs (e.g., discuss, beg, complain) printed on the front (see Appendix D for a complete list of the priming verb s by category.) During piloting of the study, children had difficulty separating the three cat egories at the same time in one task. Therefore, the priming tasks were presented as two separate tasks, one task for mental state verbs and one task fo r communication verbs, to make the task simpler. Three headings printed on three different index cards were designed to assist children in separating the cards during the priming task. One card had a header with the word body printed on it and a picture of a child running next to the word. Another card had mind printed on it with a picture of a child sitting at a desk with an inquisitive expression on his face. The last card used as a header had the word communication printed on it with a picture of someone orati ng in front of a podium next to the printed word. Narrative tasks Both the oral and written narra tives were elicited from an 11minute, 41-second, textless film, Frog Goes to Dinner (Osbourn & Templeton, 1985) in CD-ROM format, based on the book by Mercer Mayer (1975). The multi-character story is about a boy who takes his pe t frog to dinner in an expens ive restaurant, unknown to his family. The frog wanders away from the boy and comes in contact with many different characters and has numerous adventures in the restaurant before the boy finally finds him. These adventures create conflic ts in which beliefs about conv entional social expectations for how to behave in a re staurant are violated. The film consisted of 13 episodes. An ep isode is the basic structure of a narrative that, at a minimum, must consist of an Init iating Event (IE) or Goal (G), Action (A), and
27 Outcome (O) (Westby, 1999). An episode analysis was conducted in order to estimate the amount of opportunities for ment al state occurrences within each epis ode. (A detailed analysis of the episodes and possible mental state occurrences is lo cated in Appendix E.) Eight still frames from the film were used during the elicitation of the narratives in order to reduce the working memory load on the participations while formulating their oral and written narratives (see Appendix F fo r descriptions of the eight frames included). The same three headers described under the priming task were also used during the narrative task to reduce working memory dema nds and help children stay focused on the task directions. Lined writing paper and penc ils with erasers were provided for the written mode of the narrative task. Procedure Prior to administration of the experime ntal tasks, completed consent forms describing the task procedures were obtained from the pare nts. Children signed assent forms as a prerequisite for their indi vidual participation (see Appendix G for consent/assent form). The participants were tested in the fall and early spring of the 2002-2003 school year. All testing was administered by one of two examiners, both of whom were Master Degree students in speech-language pathology. Testing, including the hearing screening, administration of the PPVT-III (Dunn & Dunn, 1997), second-order false belief tasks, and narrative elicitations, was accomplished in two sessions approximately one week apart and lasting approximately 45 minutes per session, for a total time that did not exceed 90 minutes. Testing occurred in a qui et room on the school campus. Breaks were provided as needed. Pencils a nd stickers were provided as rewards for participation.
28 Second-order false belief tasks. Before administration of the logical and social inferencing tasks, a brief set of instructi ons were read, including a purpose for the task. Children were told that their help was needed to determine what the characters in the picture scenarios were thinking and feeling. The illustrated binders were then presented to each child. Following the procedures of Sullivan et al (1994) and Silliman et al. (2003), six scripted questions accompanied each scenario (see Appendix B for the complete set of questions). These questions were: (1) a firs t order question, (2) a reality question, (3) a linguistic contrast question, (4) an ignoran ce question, (5) the sec ond order question, and (6) a justification que stion. The last question was incl uded to ascerta in childrens rationale for their response to the second or der question. If the second-order question was not answered, assistance was provided using a cloze procedure. For example, Dad says, Pam thinks I got her _______. The purpose of the cloze procedure, a form of guided assistance (Silliman et al., 2003), was to reduc e the complexity of language formulation that might be competing with access of the underlying concept. Responses to the six questions were record ed on data sheets. Points were assigned as follows to the second order questions for each task: 2 = correct answer without assistance, 1 = correct answer using the cl oze procedure, 0 = incorrect answer or no response. Each participant was required to obtain a score of at least 1 on the second order question (i.e., question #5) for eith er the logical or the social inferencing task in order to meet the passing criteria for the task. Mental state verb priming task. The purpose of the priming task was to maximize familiarity across participants about the difference among communication, cognitive, and
29 physical action verbs. Although mental states can be verbs, adjectives, or adverbs, only verbs were selected to provide a clear contrast among things you do with your body, things you do with your mind, and di fferent ways of communicating. The task required children to differen tiate between two t ypes of verbs and consisted of two parts. The first part of th e priming task asked children to differentiate between action verbs and cognitive verbs. The second part of the pr iming task asked the children to differentiate between action ve rbs and communication verbs. Examples of each type of verbs were individually printe d on index cards, and participants were asked to separate the cards into two piles: one pile of cognitive verbs and a second pile of physical action verbs for the first part of th e task, and one pile of communication verbs and a second pile of physica l action verbs for the second part of the task. Before beginning the task, children were given exam ples of physical acti on verbs, cognitive verbs, and communication verbs. The first part of the ta sk was introduced as follows: Were going to play a game. Im going to show you some verbs written on cards. Verbs are words that express what you can do with your body or with your mind. For example, you can use your mind to think and use your body to run. Were going to separate the two kinds of verbs in to two piles. One pile will be things you can do with your mind, and the other p ile will be things you can do with your body. Following this introduction, participants will be instructed as follows: Now its your turn. Put a ll the verbs about things you do with your mind in this pile, and all the verbs ab out things you do with your body in this pile.
30 Scripted prompts were used as necessary, such as, Tell me if that verb is something you do with your mind or your body. Participants were allowed a maximum of 10 minutes to complete the task. The second part of the task was introduced as follows: Were going to play another game. Now Im going to show you some more verbs written on cards. These verbs are wo rds that express what you can do with your body or words that express different ways of communicating or talking. For example, you can discuss when you communicate. Were going to separate the two kinds of verbs into two piles. One pile will be different ways of communicating or talking, and the other pile will be things you can do with your body. Following this introduction, participan ts were instructed as follows: Now its your turn. Put all the verb s that express different ways of communicating in this pile and all the verbs about things you do with your body in this pile. Scripted prompts were used as necessary, such as, Tell me if that verb describes a way of communicating or something you do with your body. Participants were allowed a maximum of 10 minutes to complete the tas k, although all of the children completed the task within 3 to 5 minutes. Oral and written narrative tasks The oral and written narratives were elicited using the 11 minute, 41 second wordless CD-ROM, Frog Goes to Dinner (Osbourn & Templeton, 1985) and were presented on a lapt op computer. It should be noted that, although the film is considered to be wordle ss, one vocal utteran ce is said by the boy, a
31 major character (upon entering the kitchen, the boy sees a chef preparing to cook the frog and yells, No!). The examiner was present during the presentation of the film to maximize the childs focus on the landscape of consciousne ss and to increase the encoding of mental state references. However, the examiner did not collaboratively watch the film with the children in order to reduce child assumptions about familiarity. Both the oral and written narratives were counterbalanced. A two-phase set of instructions, including a purpose, were provided as follows: Im writing a book and need some help coming up with a good story to go with the video Im going to show you. This vide o has many different characters. There is a boy, a frog, a mom, a dad, a cook, a waiter, and some other people. Your teacher told me that youre a good storyte ller. I need you to help me make up a good story to go with the video. Le ts watch the video together. After the introduction, the following instructi ons were given before both the oral and written tasks: I want you to make up the best stor y that you can to go with the video. Remember that to tell/write a good stor y, you should talk about whats going on in the characters minds. You need to say what they might be thinking, saying, and feeling. Tell me what you think they might be thinking, feeling, and saying. Here are some pictures from th e story to help you remember. A maximum of eight still frames printed fr om the film were available to the child following both the oral and written elicitations as a strategy to re duce possible verbal working memory demands that could compet e with language formulation demands (see
32 Appendix H). Scripted questions were used as necessary as another form of guided assistance to support childrens interpretation of the characte rs beliefs and their virtual dialogue, as well as to mainta in attentional focus on the fo rmulation of mental state references and quoted dialogue. Examples of scripted quest ions included: (1) What was going on in the characters minds? (2) What else could be going on in the characters minds? (3) What might this characte r have been saying? (4) Tell me how the characters are using their minds? (5) Why do you think that character did that? (6) How does the character feel? (7) If I could listen to the charac ters talking, what would I hear? (See Appendix I for complete scripts of the priming task and elicitation of narratives.) The rationale for the inclusion of film fr ames and scripted prompts derived from Wellman et al. (2001), who found that enhanc ing the salience of me ntal states was a significant variable for improving childrens pe rformance on false belief tasks regardless of age. Therefore, an assumption was that enhanced salience would also heighten production of mental states in oral and written narratives. The oral and written conditions were presented in random order to minimize a mode order effect. Participants were allowe d a maximum of 10 minutes to respond in the oral mode and 20 minutes to respond in the written mode. Depending on the school accommodations, the writing task was sometimes administered as a small group task.
33 Equipment Both audio and video recordings were used to maximize accurate data transcription. All of the sessions were vi deo recorded with a tripod using a Memorex SM1000 video recorder and audio recorded using an Optimus CTR-117 audio cassette recorder. An Optimus 33-3013 tie-clip microphon e was also used to ensure high quality sound. The wordless video film was presente d in CD-ROM format on a Dell Sx-700 laptop computer. Data Reduction All participants were assigned a numerical code to ensure anonymity. Individual oral and written narratives were iden tified only with this numerical code. Segmentation of oral and written narratives The oral narratives were transcribed and divided into T-units (termi nable unit) and then entered in to the Systematic Analysis of Language Transcripts (SALT; Miller & Ch apman, 1984). A T-unit is defined as a main clause plus all subordinate cl auses or non-clausal structures that are embedded within the main clause (Hunt, 1965). All clauses that began with a coordina ting conjunction (e.g., and, but) were considered separate clauses ex cept when there was co-referential subject deletion (Scott, 1988), for example, Mary got in her car and (s he) went to the store. The written narratives were also divided into T-units in the same manner. Occurrence and classification of mental state references. All transcriptions were analyzed for the presence of mental state terms. Possible mental states references, including verbs, adject ives, and adverbs, compiled from pr evious research with children from the late preschool to preadolescent ye ars served as the basis for classification (Astington, 1998a, 1998b; Bretherton & Beeghly, 1982; Ely & McCabe, 1993;
34 Schwanflugel et al., 1996). Th is classification schema consisted of 68 terms: 24 motivational references, 33 experiential re ferences, and 11 belief references (see Appendix E). It should be noted that this schema was not exhaustive, but represented potential examples that could possibly o ccur during the oral and written narrative elicitation. Consistent with the concept of a de velopmental continuum, mental state references were classified into motivational, experiential, and belief categories. Motivational mental states express de sire, need, and in tentionality (e.g., want, need, try ), including the intenti on to communicate (e.g., say, tell, ask, complain ). Other types of motivational mental references express the aspe ct of possibility as an intentional state (e.g., can, will, might ). Experiential mental states typically consist of perception references referring to sight, h earing, taste, and touch (e.g., see, hear, taste, smell, feel ) and situational emotion references (Bar on-Cohen, 1995) that may be motivated by physiological reactions to a mental state (e.g., thirsty, angry, hungry, happy ). Finally, belief mental states index broader conceptual and multidimensional understanding of others minds in terms of the mental pr ocesses that govern cognitive and verbal activities (e.g., know, guess, explain, realize, wonder, promise ) (Bartsch & Wellman, 1995; Bretherton & Beeghly, 1982; Schwanflugel et al., 1996). Specifically, beliefs refer to a continuum of degrees of un certainty to degrees of certainty. Belief verbs may also be constructed with modals to express uncertainty, for example, Tom thought that he might have been a good runner if he practiced more often. As such, belief references may reflect a more literate con ceptual understanding of mental states and occur more often in the written register of sc hooling than in the oral register typical of
35 everyday conversation. (See Appendix G for codi ng of the mental states and examples.) If any additional mental states were found in the data, they were included if the examiners agreed on the classification. Qualitative analyses of belief verbs: Diversity, complexity, and depth Mental state references were also analyzed qualitat ively. One analysis examined lexical diversity of mental state references be tween the two age groups in th e oral and written modes. A second analysis sought to dis cern patterns of most freque nt and more complex production of belief verbs only. One criterion for determining complexity was age specific. Belief verbs not used by the 9 year old group in either condition but produced only by the 11 year old group were interpreted as more complex. A second complexity criterion was ba sed on Beck, McKeown, and Kucan (2002). These authors categorized words into three leve ls of word tiers, with Tier 1 representing high frequency, or commonly used, words, such as baby, happy, and walk. On the other hand, Tier 2 meanings are more literate meanings consistent with the written language register (whether this register is used orally or in writing), such as comment, vicissitudes, and inquiry. These meanings are generally slow ma pped, or elaborated with experience, as opposed to the fast mapping of more comm only used words. Tier 3 meanings are specific to subject domains, such as mathema tics, geography, literary analysis, etc., for example, algebraic peninsula, and deconstruction. The analysis of more complex belief verbs drew on Tier 2 meanings. A final qualitative analysis was condu ced on a preliminary basis to examine the lexical depth of think As modified from Nordqvist (2001 ), variations in the meaning of
36 think were examined in their specific social contexts. The purpose was to determine the extent to which the meaning of think represented degrees of uncertainty and certainty. Examiner Agreement Two graduate students in speech-language pathology served as raters for the oral and written samples. Prior to analyzing the da ta, the students formulated a coding system for classification of mental state references into three categories (i.e., motivational, experiential, belief). The examiners reached 90% agreement when coding sample responses that were not included in the study. Any disagreement as to classification of a particular mental state refe rence was resolved by discussi on. Based on the piloting of the coding system, modifications we re made before the system was actually applied to the studys data. Intercoder agreement for coding consiste ncy using an independent coder was conducted for 16 narratives from the oral a nd written samples (16% of the total). The samples were randomly chosen, such that the two age groups and two conditions (oral versus written) were equally re presented (i.e., four 9-year-old oral narratives, four 9-yearold written narratives, four 11-year-old oral narratives, and four 11-year-old written narratives). The primary coder was a doctoral level student in Speech-Language Pathology. Training was accomplished by expl aining and identifying mental state references in practice narrative s not included in the sample. Prior to coding, consensus was reached on how to divide the transcripts into Tunits. Intercoder agreement was then calculated in a series of steps. First, percentage agreement for whether a T-unit contained one or more mental state references was determined. The coders obtained 92% agreemen t for identification of T-units containing
37 mental state references. Next, in those instan ces in which a T-unit contained one or more mental state references, the references we re coded as motivati onal verb, experiential adjective or verb, or belief verb. Cohens kappa was used to determine the consistency of response coding of the mental state referenc es. The resulting kappa for mental state references was k = .95. This represents a hi gh level of consistency for coding categories (Bakeman & Gottman, 1997).
38 Chapter 3 Results This study was designed to determine the ex istence of differences in the use of mental state references by 9and 11-year-o ld typically developing children. Quantitative and qualitative analyses were conduc ted to answer two questions: 1. Did 9and 11-year-old typically deve loping children differ in their use of mental state references? 2. Did this ability differ as a function of the oral versus written modality? Data Analysis Statistical analyses were conducted first to determine whether T-unit distributions differed according to age group and narrative co ndition. Then, differences were examined as related to among age group, the frequency of mental state refere nces by category, and narrative condition. A three-way MANOVA was c onducted for this purpose, specifically, 2 (Age 9 versus Age 11) x 4 (mental state cat egories) x 2 (oral versus written modalities). For this analysis, the independe nt variables were age of pa rticipants (9 and 11 years), type of narrative (oral versus written), and type of ment al state reference (motivational verbs, experiential verbs and adjectives, a nd belief verbs). The de pendent variable was frequency of mental state code responses. Raw data frequencies were normalized by dividing the number of specific mental state references by the total num ber of T-units. Post-hoc tests were run as appropriate. In addition, qualitative analyses were conducte d to examine the variety, complexity, and depth of belief verbs in the oral and written narratives for both 9 year olds and 11 year olds.
39 Results are organized as follows. Firs t, the MANOVA results of T-units are reported. Next, MANOVA results that address the two resear ch questions are presented. An additional qualitative analysis of belief verb complexity is then provided, followed by a summary of the important findings. T-unit Analysis All narratives were transcribed and di vided into T-units (Hunt, 1965). Next, the transcripts were entered in to the software program Systematic Analysis of Language Transcriptions (SALT; Miller & Chapman, 1984). SALT produced type and frequency data for the mental state references. A descriptive analysis was conducted to determine the average number of T-units, the number of words used, and the average Tunit length per narrativ e in the oral and written contrasts for both the 9and 11-year olds. As seen in the Table 1 summary, the 9 year olds produced an average of 57.1 T-units in the oral modality and 26.0 T-units in the written modality, while the 11 year olds produc ed an average of 70.6 T-units in the oral modality and 36.4 T-units in the written modality. The range in T-units for each age group indicates the wide vari ability between individuals. Average T-unit length for participants in the 9-year-old group was 9.9 in the oral modality and 9.1 in the written modality; average T-unit length for participan ts for the 11-year-old group was 9.4 in the oral modality and 10.2 in the written modality.
40 Table 1. Descriptive Data for 9 and 11 year olds in the Oral and Written Domains. Average # of Tunits: Oral Narratives Average # of Tunits: Written Narratives Average T-unit length: Oral Narratives Average T-unit length: Written Narratives 9 year olds 57.1 (Range = 20-154) 26.0 (Range = 8-68) 9.9 9.1 11 year olds 70.6 (Range = 23-148) 36.4 (Range = 9-56) 9.4 10.2 Three separate multivariate ANOVAs (MANOVAs) were conducted to determine differences attributable to age group and/or narrative condit ion (oral vs. written). The MANOVA for number of T-units revealed significant main effects for condition, F (1,48) = 52.399, p < .001, 2 = .522, and group, F (1,48) = 4.589, p = .037, 2 = .087. The interaction between narrative conditi on and age group was not significant, F (1,48) = .117, p = .733, 2 = .002. These findings suggest that th e 11-year-olds produced more T-units overall and that both age groups produced more T-units in the oral condition. The effect sizes suggest that the finding for the narrativ e condition (oral vs. wr itten) was a stronger finding than the group difference, which was of minimal significance. The second MANOVA considered the in fluence of age group and narrative condition on the number of words produced. Like the previous MANOVA, there were significant main effects for condition, F (1,48) = 55.138, p < .001, 2 = .535, and age group, F (1,48) = 7.610, p = .008, 2 = .137. The interaction between narrative condition and age group was not significant, F (1,48) = .023, p = .879, 2 = .000. These findings
41 support the previous findings, bot h in significance and effect si ze calculations, in that the 11-year-old group produced more words than the 9-year-old group and that more words were produced in the oral condition than in the written condition. The third MANOVA for differen ces in T-unit length revealed somewhat different findings. In this case, none of the findings we re significant. This s uggests that there is no difference between age groups or between na rratives conditions for T-unit length. In other words, T-unit lengths were co mparable across groups and conditions. Analysis of Mental State References The results of a three-way MANOVA rev ealed a significant interaction between age and reference type, F (3,144) = 4.999; p = .007, partial 2 = .094. Post hoc testing using the Bonferroni procedure revealed that one paired comparis on was significant. As illustrated in Figure 1, this significant differe nce was attributed to the experiential verb category. The use of experiential verbs was significantly greater for the 11 years olds than the 9 year olds. However, the partial 2 indicated that this interaction explained very little (9%) of the variance between these f actors, so this age di fference is of minor importance. This effect size interpretation is further substantiated by the lack of a significant main effect for age, F (1,48) = .942, p = .341. This finding suggested that the two age groups did not significan tly differ in the number of occurrences of mental state references. Therefore, both age groups app eared to use each mental state reference category with equal frequency. The significant two-way interaction between reference type and modality, F (3,109) = 11.407; p < .001; partial 2 = .192, is the primary finding in this MANOVA. The partial 2 accounted for 19% of the vari ance between these factors.
42 Figure 1. Frequency of Use of Mental State References by Age. 0 2 4 6 8 10 12 14 16Motivation Exp Adj Exp Verb Belief 9 years 11 years
43 When paired samples t-tests with a Bonfe rroni correction were conducted to evaluate comparisons across reference types within m odalities, 11/12 pairwi se comparisons of interest were significant ( p < .001). As illustrated in Figure 2, the non-significant comparison was between the belief and the mo tivational references in the oral modality. In this case, there was no significant differen ce in the oral use of these two verb types. The experiential verb category occurred with the greatest fr equency, followed by motivational references, then belief references and finally experiential adjectives. Paired samples t-tests with a Bonferroni correction re vealed that the freque ncy of mental state references was always great er in the or al condition. It should also be noted that the main effects for modality, F (1,144) = 60.913, p < .0001, partial 2 = .558, and reference type, F (3,144) = 72.92; p <.0001; partial 2 = .603 were significant. The significant main effect for modality indicated that mental state references were produced more frequently in the oral narratives for both age groups. The significant main effect for refe rence type suggested that th e reference categories occurred with different frequencies. Experiential verb s were produced most frequently in both modalities, with motivational verbs, belief ve rbs, and experiential adjectives occurring less frequently in that order for both modalities.
44 Figure 2. Differences Noted by Re ference Type and Modality. 0 2 4 6 8 10 12 14 16Motivation Exp Adj Exp Verb Belief Oral Written
45 Qualitative Analyses Qualitative analyses were completed to compare the diversity, complexity, and depth of mental state references in the oral and written narra tives between the two different age groups. Belief verbs were chosen for this analysis because they were best suited to show different levels of lexical complexity. Diversity. Table 2 displays the diversity of be lief verbs produced. The 9 year olds produced a total of 194 belief verbs, with 134 verbs in the oral modality and 60 verbs in the written modality; the 11 year olds produced a total of 196 belief verbs, with 122 verbs in the oral modality and 74 ve rbs in the written modality. In terms of variation, the 9 year olds produced 29 differe nt belief verbs in the oral condition and 22 different belief verbs in th e written condition, wh ile the 11 year olds produced 27 different belief verbs in the or al condition and 25 in the written condition. As the frequencies indicate, va riations of think, including think, thinks, thought, and thinking, were most common for both ag e groups in both modalities. For the 9 year olds, the next most frequently occurring be lief verbs were notice, know, and find out in the oral modality and know, decide d, and notice in the written modality. For the 11 year olds, the next most freque ntly occurring belief verbs were know, notice, and forget in th e oral modality and notice, check, and decide in the written modality.
46 Table 2. Diversity of Belief Verbs Produced by the 9 and 11 Year Old Participants as Displayed in Order of Frequency for the Oral and Written Narratives (Number in Parenthesis Reflects the Fr equency of the Verb). 9 year olds ( n =26) 11 year olds ( n = 24) Oral Narratives Think (37) Notice (9) Know, find out (7) Search (5) Check, decide, feel, forget, realize (4) Wonder (3) Figure out, guess, like, love (2) Allow, assume, discover, figure, hope, let, mind, remember, want, wish (1) Think (27) Know (12) Notice (11) Forget (8) Realize (7) Feel, find out (5) Decide, remember, wonder (3) Check, guess, like, make sure, search (2) Believe, check on, enjoy, exaggerate, figure out, plan, pretend, see (1) Written Narratives Think (14) Know (5) Decide, notice (4) Find (3) Check, figure out, forget, search (2) Discover, felt, find out, guess, had an idea, like, mind, propose, realize, saying in his mind (1) Think (18) Notice (9) Check, decide, realize (4) Forget, know (3) Like, worry (2) Believe, bet, explore, find out, guess, have no clue, hope, love, mind, remember, see, wonder (1)
47 Complexity. Table 3 shows the most frequently us ed belief verbs, as well as the more complex belief verbs, that each ag e group produced alone. The use of more complex belief verbs demonstrates the understanding of more complex cognitive processes. As Table 3 indicates, the 11 y ear olds demonstrated greater semantic complexity in their production of belief verbs, such as believe, exaggerate, plan, pretend, and worry These verbs refer to the certainty-uncertainty domain. Table 3. Most Frequent Use and More Comp lex Use of Belief Verbs as Produced in the Oral and Written Narratives. Most Frequent Use More Complex Use 9 year olds ( n = 26) Oral modality think, notice, know, find out Written modality think, know, decide, notice (References not used by 11 year olds) Oral modality assume, discover Written modality propose 11 year olds ( n = 24) Oral modality think, know, notice, forget Written modality think, notice, check, decide (References not used by 9 year olds) Oral modality believe, exaggerate, plan, pretend Written modality explore, worry, wonder In summary, the 9 year olds were capabl e of generating mental state verbs, such as assume, discovered, and wish in th e oral mode and proposed in the written mode. The 11 year olds did not produce thes e verbs. In contrast, the 11 year olds generated more tokens of complex be lief verbs than did the 9 year olds.
48 Lexical depth. When analyzing childrens use of think in context to determine lexical depth, Nordqvist (2001) differentiate d between think as experience versus think as having an opinion. However, th ere is no consensus on procedures for analyzing lexical depth; theref ore, to attempt a finder differe ntiation of meanings of the same word, random samples of the 11 year olds narratives were selected to examine variations in meaning of the belief verb think according to their specif ic social contexts. As a result of this preliminary analysis, fi ve belief variations emerged that indicated degrees of certainty or uncertainty. The varia tions included think as decide, think as realize, think as ask oneself, think as have an opinion, and t hink as believe. An example of each variation follows: 1) He thought (decided) he would bring the frog with him. In this situation, the boy was preparing to go to dinner with his parents; he saw the frog in the dresser drawer and decided (was certain he wanted) to take the frog with him. 2) and the boy looks worried that the guy might think (realize) that he has a frog with him.; Here, as the boy was getting out of the car at the restaurant, the parking valet heard the frog cr oak. The boys face registered uncertainty that the valet might realize he had a frog. 3) He was probably thinking (asking himself) why would a boy bring a frog to a restaurant. In the same valet situat ion, just described, the valets facial expression could be inte rpreted as registering uncertainty that he heard a sound akin to a frogs croaking; thus, he might have been asking himself about the reality of what he heard.
49 4) She thought (had an opinion) she was just exaggerati ng. In this event, a woman and a man were dini ng at a table and the frog, which had jumped into the salad, was served to the woman. The woman appeared uncertain because she thought that she saw a frog in her salad, but then the frog disappeared, leading her to have an opinion that she was just exa ggerating [imagining] the situation. 5) and he thought (believed) he had caught it with thes e three little bowls. In this scenario, the frog had made his way to the kitchen and was being chased by the kitchen personnel. One of the men tried to capture the frog with three upside-down mixing bowls, but then was uncertain about which bowl the frog might be under. However, he believed he had caught the frog underneath one of the bowls. The results from this preliminary analys is indicate that by age 11 years, some children are aware of different levels, or depths, of meaning for belief references. In addition, they are able to demonstrate their understanding of these levels through selecting variations that reveal the multidimensional aspects of the meaning of a word, such as think Summary of Findings Contrary to expectations, there were no significant differences between the 9 year old and 11 year old participants in their freque ncy of use of mental state references in the oral and written narratives. The results of th is study did indicate th at experiential verbs were used most frequently and experiential ad jectives were used l east frequently in both the oral and written modalities. Motivationa l and belief verbs were used relatively
50 equally in the oral domain and motivation refe rences were used more frequently than belief in the written domain. Both age groups al so produced more mental state references in the oral mode than in the written mode. A qualitative analysis of the belief verb category suggested the possibility of a transitional shift, in terms of both breadth and depth, from more or al to more literate lexical knowledge. This trend was evident more so for the 11 year olds than the 9 year olds. The 11 year old children in this study generated narr atives in both the oral and written modes that contained belief verbs reflecting more diversity, complexity, and lexical depth than did the 9 year old child ren. Although this developmental transition was slight, it was also noted by Nordqvist (2001).
51 Chapter 4 Discussion Prior studies have primarily investigated continuing theory of mind development in older children by focusing on the organiza tion of specific mental state verbs (i.e., know ) (Booth & Hall, 1995; Schwanenflugel, Fabricius, & Noyes, 1996). With one exception, the Swedish study by Nordqvist (2001), previous studies have not investigated different types of mental state references us ing an oral-written contrast paradigm. Thus, the results from this study do not have a ri ch comparative framework. To interpret the results, three methodological factors and two conceptual i ssues are discussed initially. Then, directions for future research, incl uding clinical implica tions, are presented. Methodological Issues At least three methodological issues impact ed on the quantity and quality of the narratives in both modalities. These issues were: a) sample representativeness; b) the influence of the priming tas k, instructions, and elicitation method on performance; and c) the age interval of the two groups in the study. Sample Representativeness A major question is whether the studys procedures solicited a representative sample of childrens knowledge of mental state references. One source of evidence for representativeness was the average number of T-units. The general rule of thumb for the oral domain is that a minimum of 50 T-units is needed to meet a representative criterion, especially when lexical diversity is a fo cus (e.g., Watkins, Kelly, Harbers, & Hollis, 1995). Both age groups in this study exceeded the 50 T-unit minimums. The 9 year old
52 group produced an average of 57.1 T-units, while the 11 year old group produced an average of 70.6 T-units. The criterion for representativeness is le ss clear for writing, for example, whether formal units or specific time intervals should serve as the method for obtaining a representative sample. Some (e.g., Hughes, McGillivray, and Schmidek, 1997) suggest that 50 to 60 T-units should be collecte d, while others (Apel, Masterson, & Niessen, 2004) recommend a minimum of 50-100 words. However, there are minimal data to support that 50-60 T-units or 50100 words are sufficient for anal ysis purposes in writing. The average written T-unit length collected in this study was 26.0 for the 9 year olds and 36.4 for the 11 year olds. Hence, as might be predicted, the older group was capable of generating more information th an was the younger group. A qualification is in order as the children in this study were given only 20 minut es to plan and write their written narratives. If given more time, it is possible that they would have produced a greater number of T-units in the written modality. In view of the absence of normative T-unit data for writing, a conservative conclusion is that the writing samples were representative of childrens abilities in general and their capab ility to produce mental state refe rences specifically. What seems most relevant for representativeness, howev er, is that children s knowledge production was assessed in both the oral and written domain. Influence of the Priming Ta sk, Instructions, and Elicit ation Method on Performance Influence of the priming task One difference between the present study and the study by Nordqvist (2001) was the implementa tion of a priming task to familiarize participants with the difference between cognitive verbs, communication verbs, and
53 action verbs. The advantage of this priming procedure was that the children had recent experience with categorizing ment al state references previous to generating their oral and written narratives, and, therefore, may have been less inclined to focus on only the actions in the video. An issue, however, is that childrens exposure to various cognitive and communication verbs prior to narrative ge neration may have led to a high incidence of the verb think a finding that will be addressed shor tly. Despite this possibility, both groups of children also generated numerous mental state references, including belief verbs, that were not used for the priming task. Influence of the instructions and elicitation method. One methodological factor that may have increased the participants en gagement in the narrative activities was the presentation of a purpose prio r to initiating either the oral or written conditions. Specifically, children were asked for their he lp in writing a book to go with the video, and knowledge of a meaningful purpose may have been contributed to their interest and engagement. In addition, the elicitation methods for both oral and written narratives included an emphasis on what the characters in the vi deo were thinking, feeling, and saying, which appeared to encourage a sharing and coordina tion of character perspectives. Instructions explicitly stated that To te ll/write a good story, you should ta lk about whats going on in the characters minds. You need to say what they might be thinking, saying, and feeling. Furthermore, the eight still frames presented to the children during production of the narratives were also chosen for the various fa cial expressions, reflecting different moods, thoughts, and feelings of the characters. Fr om an observational point of view, most children utilized the frames to some extent in their story generations.
54 The combined effects of these methodologica l factors, including the priming task, explicit directions with purpose, and still frames from the video, are unknown. However, one viable conclusion is that each contributed suppor t to the quantity and richness of the narratives obtained. The issue of presentation format, including modality differences, remains important for both educational and clinical purposes (Schneider & Dube, 2005). The Age Interval There was an unexpected lack of variability in age in the production of mental state references. In retrospect, a study limita tion was the participants age interval. The particular ages chosen for the study, 9 and 11 years, were selected because these are the ages when typical children become immersed in more literate language use in the school setting, including vocabular y, reading, and writing. Results showed substantial variability in production across the three mental state categories for both age groups, suggesting that differences within and between the ages were great. At some level, this variability is consistent with the transition from less oral to more literate language expression during the preadolescent years (Rubin, 1987; Scott, 1999), when children begin to shift from tal k like books to write like books. Because of this transition, and the marked variab ility in individual difference that seems associated with it, an expanded age interv al appears necessary for any follow-up study. As an example, Nordqvists (2001) study s howed greater differences in the lexical diversity and depth of mental state verb production between 9 year olds and 15 year olds than between the 9 year old and the 12 year ol d groups. Therefore, future research in this area should further explore lexical diversit y, complexity, and depth in children with greater age intervals than the tw o age groups included in this study.
55 Conceptual Issues Lexical Diversity and Complexity A critical conceptual question is whet her the patterns of results support a continuum of development for mental state references. The increas ing lexical diversity and complexity demonstrated by the 11 year olds, as opposed to the 9 year olds, would appear to indicate progressi on along a developmental con tinuum. This progression may provide evidence of a developm ental shift from more oral to more literate uses of vocabulary, depending on the situation. Eviden ce of a developmental shift was also acknowledged by Nordqvist (2001). The qualitative analyses provided evidence for increasing lexical diversity and complexity in the production of mental st ate references by the older children, as compared to the younger group. The 11 year olds generated more in stances, or greater diversity, of more semantically and cognitively complex mental state references, especially in the belief category. Six examples of the greater scope of complexity follow. Specifically, the 9 year old children produ ced less complex meanings denoting the characters certainty about key story events (examples 1 and 2) versus those produced by the 11 year old group (examples 3 6): 1. And the lady got the frog with the salad. But at first she doesnt know (9 year old oral narrative) 2. And then he finally noticed (9 year old written narrative) 3. And he realized that the frog wasnt there (11 year old or al narrative) 4. But they just pretended that it was nothing (11 year old oral narrative) 5. But the boy is unaware (11 year old written narrative)
56 6. his frog got out of his pocket unnoticed (11 year old written narrative) According to Beck et al. (2002), the belie f references produced by the 11 year old children would be examples of more literate use of Tier 2 words, or meanings more characteristic of the written language re gister. These written register meanings significantly contribute to a richly elaborated semantic network, which may not be evident until grade 6 for many children (McGregor, 2004). Examples 4, 5 and 6 are especially in teresting because al l three are derived meanings; that is, they all have a prefix ( pre -, un -) attached to the root words as the lexical mechanism for expressing more complex thinking about the absence of awareness in others. Examples of deriva tions were seldom found in the 9 year old transcripts. One conclusion is that individual 11 year old children appeared to be advancing in their second order coordination of the multidimensi onal relationships that index states of consciousness in others, especi ally those states that refl ect degrees of certainty or factivity. An individual profile analysis in future research would aid in comparing group and individual differences. Lexical Depth In regard to revealing lexical depth, the coding of the belief verbs may have been an issue, since only the verb itself was c onsidered based on the cl assification protocol. Unlike some previous studies (Booth & Hall, 1995; Schwanenflugel et al., 1996), belief verbs, such as think and know were not interpreted along the certainty-uncertainty continuum. Also, variations in meaning (s ee Nordqvist, 2001), such as the difference between think as experience versus think as having an opinion, were not fully analyzed in their syntactic or social contexts of occurr ence. However, a preliminary
57 analysis revealed five variations in meaning for the belief verb, think Four of the five variations referred to uncerta inty, while three of the four co-occurred with sentential complementation, either the marked or unmarked form, e.g., And the boy looks worried that the guy might think that he has a frog with him (marked form). She thought [ that ] she was just exaggerating (unmarked form). Thus, it appears that, in accord with Nixon s (2005) findings on preschoolers, even during preadolescence, syntactic constrai nts and cognitive co mplexity (p. 34), combined with inferences about the social s ituation, continue to influence the emergence of the varied meanings of belief verbs. The findings from the extended examination of these meanings suggest that childrens progressive understanding of others minds will require an analysis of both the linguistic and social contexts in which me ntal state references are expressed. This interpretation is consistent not only with the findings from Booth and Hall (1995) but also with a recommendation from Montgomery (2005) More valid assessment of theory of mind will require broadening the range of words and visual and situational cues employed in (theory of mind) tasks (Montgomery, 2005, p. 119). Future Research Directions The oral-written contrast paradigm dem onstrated the quantity and richness of narrative production possible from typically developing preadolescent children. Future research should focus on the inclusion of explic it priming tasks and e xplicit directions to assess their relevance in solicit ing greater scope and depth of mental state references. In addition, drawing on childrens situational knowledge of social contexts in which
58 thoughts and beliefs are attributed to other pe ople also seems beneficial in encouraging children to express the mental state references they are typically capable of producing. Finally, subsequent work should focus on sta ndards for a representative writing sample, comparing whether number of words, number of T-units, or time intervals are the most reliable units. Most importantly, analyzing belief verbs such as think and know in their linguistic and social contexts to determine significant variations in meaning would seem imperative in order to reveal the nuances of lexical depth not eviden t by categorization alone. Since belief is an abstract concept (you cant se e a belief, much less degrees of certainty or uncertainty), how children assi milate and express their evol ving understanding of such abstract concepts throughout their childhood is relevant to their understanding of others behaviors and their ski lls with social interactions. Be ing aware of the subtleties of meaning inherent in mental state references is an indication of mo re advanced cognitive, lexical, and, potentially, synt actic growth. Analyzing different levels of meaning for cognitive verbs of uncertainty may be a powerfu l way to paint a more complete picture of developmental differences in the semantic lexicon of older children. A second area of research concerns Au tism Spectrum Disorder (ASD). Past research on children with ASD indicates that this particular developmental disorder is characterized by significant limitations in theory of mind devel opment (Baron-Cohen, 1995). Communication and social interactions with others are hampered when children with autism are unable to attri bute mental states to other peop le or predict the behavior of others based on their interpreta tion of mental states Previous studies have not attempted to elicit mental state references from childre n with autism using a priming task or via an
59 oral-written contrast paradigm. Since there is wide vari ability in lexical, syntactic, and discourse skills across the autistic spectrum, in itial studies in this area should focus on higher functioning children, such as those with Aspergers syndrome. Conclusion In summary, the results of this study s uggest evidence of childrens progress on a developmental continuum in the acquisition of more complex mental state references, such as belief verbs, indicating that theory of mind development continues to advance in preadolescent children. Although only a slight developmental shift was evident in the production of mental state references between th e ages of 9 and 11 years, future research with broader age intervals may reveal more evidence of a developmental continuum that continues to progress in complexity a nd flexibility throughout adolescence and young adulthood. These results also offer evidence of the length of time required to acquire more intricate lexical knowledge. In this view theory of mind devel opment is not simply an acquisition that stops in the preschool y ears, but is an ongoing transformation of mind sharing and perspective taking that enrich so cial interactions with others throughout our lives.
60 References Apel, K., Masterson, J. J., & Niessen, N. L. (2004). Spelling assessment frameworks. In C. A. Stone, E. R. Silliman, B. J. Ehren, & K. Apel (Eds.), Language and literacy: Development and disorders (pp. 644-660). New York: Guilford Press. Astington, J. W. (1998a). Theory of mind, Humpty Dumpty, and the icebox. Human Development, 41, 30-39. Astington, J. W. (1998b). Theory of mind goes to school. Educational Leadership, 56, 46-48. Astington, J. W. (1999). The language of inte ntion: Three ways of doing it. In P.D. Zelazo, J. W. Astington, & D. R. Olson (Eds.), Developing theories of intention: Social understanding and self-control (pp. 295-315). Mahway, NJ: Lawrence Erlbaum Associates. Astington, J. W. (2001). The future of theory-of-mind research: Understanding motivational states, the role of langu age, and real-world consequences. Child Development, 72, 685-687. Astington, J. W. (2003). Sometimes nece ssary, never sufficient: False belief understanding and social competence. In B. Repacholi & V. Slaughter (Eds.), Individual differences in theory of mind: Implications for typical and atypical development (pp. 13-38). New York: Psychology Press. Astington, J. W., Harris, P. L., & Olson, D. (1988). Developing theories of mind. Cambridge: Cambridge University Press.
61 Astington, J. W., & Jenkins, J. M. (1999). A longitudinal study of the relation between language and theory-of-mind development. Developmental Psychology, 35, 13111320. Bakeman, R., & Gottman, J. M. (1997). Observing interaction: An introduction to sequential analysis (2nd ed.) New York: Cambridge University Press. Baron-Cohen, S. (1995). Mindblindness: An essay on autism and theory of mind. Cambridge, MA: MIT Press. Bartsch, K., & Wellman, H. M. (1995). Children talk about the mind. New York: Oxford University Press. Beck, I. L., McKeown, M. G., & Kucan, L. (2002). Bringing words to life. New York: Guilford Press. Booth, J. R., & Hall, W. S. (1995). Devel opment of the understanding of the polysemous meanings of the mental-state know. Cognitive Development, 10, 529-549. Bretheron, I., & Beeghly, M. (1982). Talking about internal states: The acquisition of an explicit theory of mind. Developmental Psychology, 18, 906-921. Bruner, J. (1986). The culture of education Cambridge, MA: Harvard University Press. de Villiers, J. G. (2003). Defi ning SLI: A linguistic perspective. In Y. Levy & J. Schaffer (Eds.), Language competence across populations: Toward a definition of specific language impairmen t (pp. 425-447). Mahwah, NJ: Lawrence Erlbaum. de Villiers, J. G., & de Villiers, P. A. (2000). Linguistic determinism and the understanding of false beliefs. In P. Mitchell & K. J. Riggs (Eds.), Childrens reasoning and the mind (pp. 191-228). East Sussex, UK: Psychology Press.
62 Donahue, M. L., & Foster, S. K. (2004). Integration of language and discourse components with reading comprehension: It s all about relati onships. In E .R. Silliman & L. C. Wilkinson (Eds.), Language and literacy learning in schools (pp. 175-198). New York: Guilford Press. Dunn, J. (1999). Making sense of the so cial world: Mindreading, emotion, and relationships. In P. D. Zelazo, J. W. Astington, & D. R. Olson (Eds.) Developing theories of intention: Soci al understanding and self-control. Mahwah, New Jersey: Lawrence Erlbaum Associates. Dunn, L. M., & Dunn, L. M. (1997). Peabody Picture Vocabulary Test (3rd edition). Circle Pines, MN: American Guidance Service. Ely, R., & McCabe, A. (1993). Remembered voices. Journal of Child Language, 20, 671696. Fey, M. E., Catts, H. W., Proctor-Williams, K., Tomblin, J. B., & Zhang, X. (2004). Oral and written story composition skills of children with language impairment. Journal of Speech, Language, and Hearing Research, 47, 1301-1318. Flavell, J. H. (1999). Cognitive development: Childrens knowledge about the mind. Annual Review of Psychology, 50 21-45. German, T. P., & Leslie, A. M. (2000). Attend ing to and learning about mental states. In P. Mitchell & K. J. Riggs (Eds.) Childrens reasoning and the mind. East Sussex, UK: Psychology Press. German, T. P., & Leslie, A. M. (2001). Ch ildrens inferences from knowing to pretending and believing. Journal of Developmental Psychology, 19 59-83.
63 Gopnik, A., & Astington, J. W. (1988). Child rens understanding of representational change and its relation to the understandi ng of false belief and the appearancereality distinction. Child Development, 59, 26-37. Gopnik, A., & Wellman, H. (1992). Why the childs theory of mind really is a theory. Mind and Language, 7, 145-171. Happe, F. G. E. (1994). An advanced test of theory of mind: Understanding of story characters thoughts and feelings by able autistic, mentally handicapped, and normal children and adults. Journal of Autism and Developmental Disorders, 24, 129-154. Happe, F. G. E. (1995). The role of age and verbal ability in the theory of mind task performance of subjects with autism. Child Development, 66, 843-855. Hughes, D., McGillivray, L., & Schmidek, M. (1997). Guide to narrative language: Procedures for assessment. Eau Clair, WI: Thinking Publications. Hunt, K. W. (1965). Grammatical structures written at three grade levels (Research Report No. 3) Champaign, IL: National Council of Teachers of English. Hyter, Y. D., & Westby, C. E. (1996). Usi ng oral narratives to assess communicative competence. In A. G. Kamhi, K. E. Pollack, & J. L. Harris (Eds.), Communication development and disorders in African Amer ican children: Research, assessment, and intervention (pp. 247-284). Baltimore, MD: Paul H. Brookes. Jenkins, J. M., & Astington, J. W. (1996). Cognitive factors and family structure associated with theory of mi nd development in young children. Developmental Psychology, 32, 70-78.
64 Jenkins, J. M., Turrell, S. L ., Kogushi, Y., Lollis, S., & Ross, H. S. (2003). A longitudinal investigation of the dynamics of mental state talk in families. Child Development, 74 905-920. Johnson, D. D., & von Hoff Johnson, B. (1986). Highlighting vocabulary in inferential comprehension instruction. Journal of Reading, 29, 622-625. Leslie, A. M. (1987). Pretense and represen tation: The origins of theory of mind. Psychological Review, 94, 412-426. Lillard, A. (1998). Theories behind theories of mind. Human Development, 41, 40-46. MacWhinney, B. (in press). The emergence of gr ammar from perspective. In D. Pecher & A. Zwaan (Eds.), The grounding of cognition: The ro le of perception and action in memory. McGregor, K. K. (2004). Developmental de pendencies between lexical semantics and reading. In C. A. Stone, E. R. Sillim an, B. J. Ehren, & K. Apel (Eds.), Handbook of language and literacy: Developmental and disorders (pp. 302-317). New York: Guilford Press. Mayer, M. (1975). Frog goes to dinner New York: A Puffin Pied Pipe. Meins, E., Fernyhough, C., Wainwright, R., G upta, M., D., Fradley, E., & Tuckey, M. (2002). Maternal mind-mindedness and attach ment security as predictors of theory of mind understanding. Child Development, 73, 1715-1726. Miller, J., & Chapman, R. (1984). Systematic Analysis of Language Transcriptions [Computer software]. Madison, WI: La nguage Analysis Laboratory, Waismon Center, University of Wisconsin.
65 Montgomery, D. E. (2005). The developmental orig ins of meaning for mental terms. In J. W. Astington & J. A. Baird (Eds.), Why language matters for theory of mind (106-122). New York: Oxford. Nelson, K. (2004). Commentary: The future of ToM lies in CoM. Newsletter of the International Society for the Study of Behavioural Development, 1 (Serial No. 45), (pp. 16-17). Nixon, S. M. (2005). Mental state verb producti on and sentential complements in fouryear-old children. First Language, 25, 19-37. Nordqvist, A. (2001). Speech about speech: A developmental study on form and function of direct and indirect speech. Unpublished doctoral dissertation, Gteborg University, Sweden. Osbourn, S. (Producer) & Templeton, G. (D irector). (1985). Frog goes to dinner [Film]. Rochester, NY: Phoenix Films. Perner, J. (1995). The many faces of belief: Reflections on Fodors and the childs theory of mind. Cognition, 57, 241-269. Perner, J., & Wimmer, H. (1985). John thinks that Mary thinks th at attribution of second-order beliefs by 5to 10-year-old children. Journal of Experimental Psychology, 39, 437-471. Rubin, D. L. (1987). Dive rgence and convergence between oral and written communication. Topics in Language Disorders, 7, 1-18. Ruffman, T., Slade, L., & Crowe, E. ( 2002). The relation between childrens and mothers mental state language and theory-of-mind understanding. Child Development, 73, 734-751.
66 Schneider, P., & Dube, R. V. (2005). Story presentation effects on childrens retell content. American Journal of Speech-Language Pathology, 14, 52-60. Schwanenflugel, P. J., Fabricius, W. V., & Noyes, C. R. (1996). Developing organization of mental verbs: Evidence for the devel opment of a constuctiv ist theory of mind in middle childhood. Cognitive Development, 11, 265-294. Scott, C. (1999). Learning to write. In H. Catts & A. Kamhi (Eds.), Language and reading disabilities (pp.224-258). Boston, MA: Allyn and Bacon. Scott, C. M. (1988). Producing complex sentences. Topics in Language Disorders, 8, 4462. Scott, C. M., & Windsor, J. (2000). General language performance measures in spoken and written narrative and expository discourse of school-age children with language learning disabilities. Journal of Speech, Language, and Hearing Research, 43, 324-339. Silliman, E. R., Diehl, S. F., Hnath-Chisol m, T., Bahr, R., Zenko, C. B., & Friedman, S. (2003). The understanding of mental and em otional states in adolescents with autism spectrum disorder: Does discourse support matter? Language, Speech, and Hearing Services in the Schools, 34, 236-252. Sullivan, K., Zaitchik, D., & Tager-Flusger g, H. (1994). Preschoolers can attribute second-order beliefs. Developmental Psychology, 30, 395-402. Thelen, E., & Bates, E. (2003). Connectioni sm and dynamic systems: Are they really different? Developmental Science, 6 pp.378-391.
67 Watkins, R. V., Kelly, D. J., Harbers, H. M., & Hollis, W. (1995). Measuring childrens lexical diversity: Differentiating typi cal and impaired language learners. Journal of Speech and Hearing Research, 38 1349-1355. Wellman, H. M. (1990). The childs theory of mind. Cambridge, MA: The MIT Press. Wellman, H. M., Cross, D., & Watson, J. (2001). Meta-analysis of theory-of mind development: The truth about false belief. Child Development, 72, 655-684. Wellman, H. M., Phillips, A. T., & Rodriguez, T. (2000). Young childrens understanding of percepti on, desire, and emotion. Child Development, 71, 895-91. Westby, C. E. (1999). Assessing and facilitating text comprehension problems. In H. W. Catts & A. G. Kamhi (Eds.), Language and reading disabilities (pp. 154-223). Needham Heights, MA: Allyn & Bacon. Wimmer, H., & Perner, J. (1983) Beliefs about beliefs: Repr esentation and constraining function of wrong beliefs in young childrens understanding of deception. Cognition, 13, 103-128.
69 Appendix A: Table 2. Age, Gender, Ethnicity, and Results of Inclus ion Criteria Measures for the 50 Participants. Age Gender Grade Ethnicity PPVT-III Logical Task Social Task 9;0 Female 3rd Caucasian 85 2 0 9;0 Male 3rd AA 89 1 0 9;1 Female 3rd Caucasian 85 2 2 9;1 Female 4th Caucasian 115 2 2 9;1 Male 3rd AA 121 2 2 9;3 Male 4th Caucasian 99 2 2 9;4 Male 4th AA 100 2 2 9;4 Male 4th AA 118 2 2 9;4 Female 3rd Caucasian 115 2 2 9;5 Female 4th AA 114 2 2 9;5 Male 4th Caucasian 106 2 2 9;5 Female 3rd Caucasian 112 2 2 9;6 Female 4th Hispanic 108 2 2 9;6 Male 3rd Caucasian 102 2 2 9;6 Female 3rd Caucasian 94 0 2 9;7 Male 4th Caucasian 103 2 2 9;7 Female 4th Caucasian 115 0 2 9;8 Female 4th Caucasian 107 2 2 9;9 Male 3rd Caucasian 109 2 2 9;9 Female 4th Caucasian 104 2 2 9;10 Female 4th Caucasian 99 2 2 9;11 Female 4th Caucasian 103 2 2 9;11 Female 4th Caucasian 91 2 2 9;11 Female 4th Caucasian 110 2 0 9;11 Male 3rd AA 106 2 2 9;11 Male 3rd AA 120 2 2 11;0 Female 5th Hispanic 99 2 2 11;1 Female 5th Caucasian 93 2 2 11;1 Male 5th Caucasian 108 2 2 11;2 Female 5th Caucasian 107 2 2
70 Appendix A: (Continued) Age Gender Grade Ethnicity PPVT-III Logical Task Social Task 11;2 Female 5th AA 89 2 0 11;2 Female 5th AA 105 2 2 11;2 Male 5th Caucasian 124 2 2 11;2 Female 5th AA 103 2 2 11;3 Female 5th Asian 94 2 2 11;3 Female 5th Caucasian 114 2 2 11;3 Male 5th Caucasian 111 2 2 11;3 Male 5th Caucasian 114 2 0 11;3 Female 5th Caucasian 111 2 2 11;3 Male 5th AA 107 2 2 11;3 Male 5th AA 103 2 2 11;4 Female 5th AA 103 2 2 11;4 Female 5th Caucasian 89 2 2 11;5 Male 5th AA 87 2 2 11;5 Male 5th Caucasian 93 2 2 11;5 Male 5th AA 105 2 2 11;5 Male 5th Caucasian 110 2 2 11;8 Female 5th Caucasian 99 2 2 11;8 Female 5th Caucasian 98 2 2 11;9 Female 5th Caucasian 108 2 2
71 Appendix B: Second-Order False Belief Tasks Logical Inferencing Scenario: 1. Implicit False Belief: (modified from Sullivan et al., 1994) This is a story about Pam and her Dad. Toda y is Pams birthday and shes having a big party tonight. Dad is surprising her with a ne w bike that he has hi dden in the living room. See? Here is the surprise bike. Pam and Dad are in the kitchen talking about her birthday. Pam says, Dad, I really want a new bike for my birthday. Now remember, Dad wants the bike to be a surprise, so he says, Sorr y, I didnt get you that. I got you roller blades instead. First Order Question : What does Pam think Dad got her for her birthday ? (If necessary, Fill in: Pam thinks Dad got her ____.) Reality Question : What did Dad really get her for her birthday? (If necessary, Fill in: Dad really got Pam ___.) Then Pam says to Dad, O.K., well Im goi ng over to my friends house. Ill be home later. On her way out, Pam goes into the li ving room to get her umbrella because its raining. In the living room, she finds her new bike! She thinks to herself, Yes! Dad did not get me roller blades. He really got me a bike. Dad does not see Pam go into the living room and find the bike. Linguistic Contrast Question : Does Dad know that Pam saw her bike in the living room? Later, Pams grandmother comes over for the party. Grandma asks Dad, Does Pam know what you got her for her birthday?
72 Appendix B: (Continued) Ignorance Question : What does Dad say? ( If necessary, Fill in: Dad says ____.) Now remember, Dad does not know that Pam saw wh at he got her for her birthday. Then Grandma asks Dad, What does Pam th ink you got her for her birthday? Second Order Question : What does Dad say? (If necessary, Fill in: Dad says, Pam thinks I got her ___.). Justification Question : Why does Dad say that? Social Inferencing Task: 2. Implicit False-Belief: This is a story about Frank and his Dad. Toda y is Franks birthday. Frank wants to go to a baseball game for his birthday. He does not wa nt a surprise party for his birthday. Frank hates surprise parties. He gets embarra ssed when everyone looks at him and yells Surprise! Franks Dad is giving him a surprise birt hday party. Dad bought Frank balloons and a birthday cake and hid them in the living room. Dad does not know that Frank gets embarrassed at surprise parties. Dad thinks Frank would be glad to have a surprise party. Frank and Dad are in the kitc hen talking about his birthday. Frank says, Dad I really want to go to a baseball ga me for my birthday. Now remember, Dad wants the party to be a surprise, so he says, Fra nk, thats a good idea. I will think about it. First Order Question : What does Frank hope that Dad will do for his birthday? (If necessary, Fill-in: Frank hopes that Dad will _____).
73 Appendix B: (Continued) Reality Question : What did Dad really do for Fra nks birthday? (If incorrect, say: But remember, Dad wants to surprise Frank with th e party. If necessary, Fill-in: Dad really ____). Then Frank says to Dad, Great, Im going to tell my friend next door, Ill be home later. On his way out, Frank sees the ball oons and cake hidden in the living room. He thinks, Oh no! Dad is giving me a surpri se party; we are no t going to the baseball game. Remember, Dad does not see Frank go in to the living room and find the balloons and cake. Linguistic Contrast Question : Does Dad know that Frank is disappointed? (If necessary, Fill in: n/a). Later, Franks grandmother comes over for the party. Grandma asks Dad, Does Frank like surprise parties? Ignorance Question : What does Dad say? (If necessary, Fill-in: Dad says__). Now remember, Dad does not know that Frank gets embarrassed at surprise parties. Then Grandma asks Dad, How will Frank feel about having a surprise party? Second Order Question : What does Dad say? (If necessary, Fill-in: Frank will be ____). Justification Question : Why does Dad say that?
74 Appendix C: Score Sheet for False Belief Tasks Name: _______________________ Date: ___-___-___ School:_______________________ Gender: M F Ethnicity: _____________________ Grade: _______ Birthday: ___-___-___ Age: _______months Introduction: I am writing some stories for little children. I want you to tell me if you think little children would like these stories. I am going to read two stories to you. Then I will ask you questions about the stories. I wa nt you to look at the pictures and listen carefully. If you have any questions, you can ask me. Logical Inferencing Scenario: Implicit False Belief: (modified from Sullivan et al., 1994) This is a story about Pam and her Dad. T oday is Pams birthday and shes having a big party tonight. Dad is surprising her w ith a new bike that he has hidden in the living room. See? Here is the surprise bike. Pam and Dad are in the kitchen talking about her birthday. Pam says, D ad, I really want a new bike for my birthday. Now remember, Dad wants the bike to be a surprise, so he says, Sorry, I didnt get you that. I got you roller blades instead. First Order Question : What does Pam think Dad got her for her birthday ? (If necessary, Fill in: Pam thinks Dad got her ____.) 2 = Correct answer w ithout assistance. 1 = Correct answer with cloze procedure. 0 = Incorrect answer or no response. Childs answer: __________________________________________ Correct answer: Rollerblades
75 Appendix C: (Continued) Reality Question : What did Dad really get her for her birthday? (If necessary, Fill in: Dad really got Pam ___.) 2 = Correct answer w ithout assistance. 1 = Correct answer with cloze procedure. 0 = Incorrect answer or no response. Childs answer: __________________________________________ Correct answer: A bike. Then Pam says to Dad, O.K., well Im going over to my friends house. Ill be home later. On her way out, Pam goes into the living room to get her umbrella because its raining. In the living room, she finds her new bike! She thinks to herself, Yes! Dad did not get me roller blades. He really got me a bike. Dad does not see Pam go into the living room and find the bike. Linguistic Contrast Question : Does Dad know that Pam saw her bike in the living room? 2 = Correct answer w ithout assistance. 1 = Correct answer with cloze procedure. 0 = Incorrect answer or no response. Childs answer: __________________________________________ Correct answer: No. Ignorance Question : Later, Pams grandmother comes over for the party. Grandma asks Dad, Does Pam know what you got her for her birthday? What does Dad say? ( If necessary, Fill in: Dad says ____.)
76 Appendix C: (Continued) 2 = Correct answer w ithout assistance. 1 = Correct answer with cloze procedure. 0 = Incorrect answer or no response. Childs answer: __________________________________________ Correct answer: No. Second Order Question : Now remember, Dad does not kno w that Pam saw what he got her for her birthday. Then Grandma asks Dad, What does Pam think you got her for her birthday? What does Dad say? (If necessary, Fill in: Dad says, Pam thinks I got her ___.). 2 = Correct answer w ithout assistance. 1 = Correct answer with cloze procedure. 0 = Incorrect answer or no response. Childs answer: __________________________________________ Correct answer: Rollerblades. Justification Question : Why does Dad say that? 2 = Correct answer w ithout assistance. 1 = Correct answer with cloze procedure. 0 = Incorrect answer or no response. Childs answer: __________________________________________ Correct answer: Because he doesn t know that Pam saw the bike.
77 Appendix C: (Continued) Social Inferencing Task: Implicit False-Belief: This is a story about Frank and his Da d. Today is Franks birthday. Frank wants to go to a baseball game for his birthday. He does not want a surprise party for his birthday. Frank hates surpri se parties. He gets embarrassed when everyone looks at him and yells, Surprise! Franks Dad is giving him a surprise birthday party. Dad bought Frank balloons and a birthday cake and hid them in the living room. Dad does not know that Frank gets embarra ssed at surprise parties. Dad thinks Frank would be glad to have a surprise party. Frank and Dad are in the kitchen talking about his birthd ay. Frank says, Dad I really want to go to a baseball game for my birthday. Now remember, Dad want s the party to be a surprise, so he says, Frank, thats a good idea. I will think about it. First Order Question : What does Frank hope that Dad will do for his birthday? (If necessary, Fill-in: Frank hopes that Dad will _____). 2 = Correct answer w ithout assistance. 1 = Correct answer with cloze procedure. 0 = Incorrect answer or no response. Childs answer: __________________________________________ Correct answer: He hopes he will take him to a baseball game. Reality Question : What did Dad really do for Franks birthday? (If incorrect, say: But remember, Dad wants to surprise Frank with th e party. If necessary, Fill-in: Dad really ____).
78 Appendix C: (Continued) 2 = Correct answer w ithout assistance. 1 = Correct answer with cloze procedure. 0 = Incorrect answer or no response. Childs answer: __________________________________________ Correct answer: He is going to throw him a surprise party. Then Frank says to Dad, Great, Im goi ng to tell my friend next door, Ill be home later. On his way out, Frank sees the balloons and cake hidden in the living room. He thinks, Oh no! Dad is giving me a surprise party; we are not going to the baseball game. Remember, Dad does not see Frank go into the living room and find the balloons and cake. Linguistic Contrast Question : Does Dad know that Frank is disappointed? (If necessary, Fill in: n/a). 2 = Correct answer w ithout assistance. 1 = Correct answer with cloze procedure. 0 = Incorrect answer or no response. Childs answer: __________________________________________ Correct answer: No. Ignorance Question : Later, Franks grandmother comes over for the party. Grandma asks Dad, Does Frank like surprise pa rties? What does Dad say? (If necessary, Fill-in: Dad says__). 2 = Correct answer w ithout assistance. 1 = Correct answer with cloze procedure.
79 Appendix C: (Continued) 0 = Incorrect answer or no response. Childs answer: __________________________________________ Correct answer: Yes. Second Order Question : Now remember, Dad does not know that Frank gets embarrassed at surprise parties. Then Grandma asks Dad, How will Frank feel about having a surprise party? What does Dad say? (If necessary, Fill-in: Frank will be ____). 2 = Correct answer w ithout assistance. 1 = Correct answer with cloze procedure. 0 = Incorrect answer or no response. Childs answer: __________________________________________ Correct answer: He will like it. Justification Question : Why does Dad say that? 2 = Correct answer w ithout assistance. 1 = Correct answer with cloze procedure. 0 = Incorrect answer or no response. Childs answer: __________________________________________ Correct answer: Dad does not know that Frank gets embarrassed at surprise parties.
80 Appendix D: List of Priming Verbs Mental State Verbs Action Verbs Think Run Know Play Guess Swim Understand Fall Decide Eat Notice Jump Concentrate Walk Study Hit Communication Verbs Action Verbs Explain Climb Question Crawl Complain Cut Demand Draw Beg Push Remind Brush Promise Ride Discuss Drive
81 Appendix E: Episode Analysis of Frog Goes to Dinner Episode 1 The boy is in his bedroom getting dressed for dinner. He hears the frog croak. He gets the frog out of the drawer. He hears the car horn honk and runs out of the room. He comes back for the frog and puts him in his pocket. He leaves in the car with his mom and dad. Episode 2 They arrive at the restaurant. The valet is waiting to park the car. The frog croaks, and the valet hears him. The boy and the valet look at each other. The boy looks surprised and scared He is afraid the valet will say something. He runs into the restaura nt after his mom and dad. Episode 3 The boy and his parents are si tting in the restaurant looking at menus. The frog jumps out of the boys pocket, but the boy doesnt know it. A band is playing music. The frog climbs up on the stage and jumps into the saxophone. The musician blows the sa xophone, and the frog flies out.
82 Appendix E: (Continued) He lands in the lobster tank. Episode 4 The boy checks his pocket and realizes the frog is gone. He looks worried His parents are still looking at the menus. He sneaks under the table and goes to look for the frog. Episode 5 The frog swims to the side of the tank and is trying to get out. A lobster is swimming toward him with his pincers out. Will the frog get out in time? He climbs out of the tank be fore the lobster gets him. Episode 6 The frog jumps into a bowl of salad. The waiter serves the salad to a couple at the table. They look hungry They start eating the salad, and they like it. Then the lady finds the frog in her salad. She looks up and appears shocked When she looks down again, the frog is gone. Then she sees the frog on the mans head. She looks horrified and points to the mans head. The man gets excited and tries to get the frog off his head.
83 Appendix E: (Continued) The frog jumps in a water pitche r that a waiter is carrying. The mans toupee flies into a lady s soup bowl at another table. The lady at the other table looks surprised The couple with the salad r un out of the restaurant. Episode 7 The waiter does not see the frog in the water pitcher. He pours the frog into a water glass at a table where a man and a woman are holding hands. The couple at the table get mad and leave the restaurant. Episode 8 The boys parents look up from their menus and realize he is missing. They look concerned The boy is still searching for the frog. Episode 9 The frog jumps on a souffle being served at a table. The souffle collapses. The people at the table see the frog and look disgusted The frog jumps onto a tray of dirty dishes. Episode 10 The dishwasher sees the frog in the kitchen on a plate. He catches the frog under a bowl. He cant figure out which bowl the frog is under.
84 Appendix E: (Continued) The frog is still under a bowl and is getting away. The dishwasher tries to catch him. The pastry chef trips over him and falls into a cake. A waiter comes into the kitchen an d also trips, dropping his tray. Another waiter sees the chaos and decides not to go into the kitchen. Abbreviated Episode 11 The boys parents are still walk ing around the restaurant and looking for the boy. They appear worried Episode 12 The cook sees the big mess in the kitchen and then sees the frog. The cook looks really mad He catches the frog and looks at him. He holds up a big knife. The boy runs in and yells, Stop! Episode 13 The boy and his parents arrive at home. The boy is worried that he is in trouble. The boy goes upstairs. The parents look at each other and start to laugh The boy turns around and sees his parents laughing He knows he is not in trouble. The boy is in his room and starts to laugh too.
85 Appendix F: Description of the Eight Still Frames Frame 1 The boy comes back to his room to get his frog. Frame 2 The valet hears the frog croak a nd looks questioningly at the boy. Frame 3 The frog is about to jump out of the boys pocket unknown to the boy. Frame 4 The boy sneaks under the table to look for the frog. Frame 5 A woman is staring at the frog on a mans head and looks horrified. Frame 6 The dishwasher catches the frog under a bowl. Frame 7 The boys parents are looking fo r him and appear worried. Frame 8 The parents look at each other and start to laugh.
86 Appendix G: Consent Form 1. General Information of the Research Study: Your child is invited to participate in a research study called Talking and Writing about Thinking: Childrens Use of Special Meanings . Children use special words to talk and write about thinking, for example, think, know, guess, remember, want to, mean to, and can. Children learn that other peop le also think in these ways. This is when children learn to see other peoples points of view. Our pur pose is to see if 9and 11-year children differ in their ability to use these types of words when talk ing or writing. This study is being conducted by faculty and students in the Department of Co mmunication Sciences and Disorders at the University of South Flor ida (USF). Your countys school district and the administrators and teachers at your chil ds school have allowed us to do this study. We are asking for permission to include your child. We hope to have 100 children in grades 4 and 6 participate in the study. 2. Description of the Research Study: This study will include 9-year-old and 11-year-o ld children. The initial part of the study will consist of a hearing screening and admi nistration of a general vocabulary test. Two tasks that draw on childrens ability to r eason about physical and social consequences will also be administered. These tasks require understanding of other peoples beliefs, including false beliefs. The children will then be shown a 12-minute film, Frog Goes to Dinner. The children will be asked to talk and wr ite about the events in the film, focusing on the thoughts and feelings of the characters Children find this type of activity to be fun. Testing will be accomp lished in two sessions no longer than 45 minutes each. 3. Benefits of the Study: The results of this study will help educat ors by leading to a better understanding of how oral language skills relate to reading a nd writing abilities. St ory telling and reading comprehension often require the ability to take other perspectives or points of view. We assume that, if a child is able to understand the perspectives of different characters in a story, there will be a corresponding increase in reading comprehension. This study is the first step of further research that will benefit children who are struggling with reading and writing. 4. Volunteering for the Study: Your decision to allow your child to particip ate in this research study is completely voluntary. You are free to allow your child to participate in this research study or to withdraw him/her at any time. If you choose not to allow your child to participate, or if you withdraw him/her, there will be no pena lty to you or your child. Neither refusal to participate nor the results obtained from th e study will in any way affect your child grades or eligibility for a special progra m in Hillsborough County. You or your child will not be paid or receive special considerations for participa tion in this study. There are no known risks to participants. 5 Confidentiality of Your Childs Records: During this study, all students will be videotaped and audiotaped; however, all videotapes, audiotapes, and test results will be kept confidential to the extent of the law.
87 Appendix G: (Continued) Authorized research investigators, agents of the Department of Health and Human Services, and the USF Institutional Review Bo ard may inspect records from this project. While the results of this study may be repor ted at professional m eetings, published in professional journals, or used for training graduate student s in Communication Sciences and Disorders, your childs anonymity will be maintained. Each chil ds individual data will be coded by number without the childs na me and school. All data will be kept in a locked language laboratory in the Departme nt of Communication Sciences and Disorders at the University of South Florida. The room also has an electronic security system for additional security. 6. Instructions: Please read and sign the consent form (printed on the back) a nd promptly return it to your child teacher by ___________________. You may also return the consent form or get additional information from Dr. Elaine Silliman at the USF Department of Communication Sciences and Disorders. We would greatly appr eciate your help in allowing your child to be a part of this important study. If you have additional questions, please call Dr. Silliman at (813) 974-9812. Once the study is completed, we will be happy to provide a summary of results to a ny parent or guardian who requests a copy. If you or your child have question about your childs rights as a person who is taking part in a research study, you or your child may contac t a member of the Division of Research Compliance of the University of South Florida at (813) 974-5638. Childs Name__________________ D.O.B._______________________ Teachers Name________________ Your Consent By signing this form I agree that: I have fully read or have had read and e xplained to me in my native language this informed consent form describing a research project. I have had the opportunity to question one of the persons in charge of this research and have received satisfactory answers. I understand that I am being asked to allow my child to participate in research. I understand the risks and benefits, and I freely give my consent to allow my child to participate in the research pr oject outlined in this form, under the conditions indicated on it. I have been given a signed copy of this informed consent form, which is mine to keep. ____________________________ ____________________ _______________ Signature of Parent of Participant Printed Name of Parent Date
88 Appendix G: (Continued) Childs Assent Statement The research study called Talking and Writing about Thinking has been explained to me. I agree to be in this study. _________________________ _________________________ ____________ Signature of Child Printed Name of Child Date _________________________ _________________________ ____________ Signature of Parent Printed Name of Parent Date _________________________ _________________________ ____________ Signature of Investigator Printed Name of Investigator Date _________________________ _________________________ ____________ Signature of Witness Printed Name of Witness Date By signing this form I agree that: Particip ants have been provided with adequate information relative to the study. A phone number has been provided in case of questions. _________________________ _________________________ ____________ Signature of Investigator Printed Name of Investigator Date Institutional Approval of Study and Informed Consent This research project/study and informed c onsent form were reviewed and approved by the University of South Florida Instituti on Review Board for the protection of human subjects. This approval is valid until the date provided below. The board may be contacted at (813) 974-5638. Approval Consent Form Expiration Date: _____________ Revision Date: _____________
89 Appendix H: Eight Still Frames
90 Appendix H: (Continued)
91 Appendix H: (Continued)
92 Appendix H: (Continued)
93 Appendix I: Script for Prim ing Task and Narrative Task Priming Task Introduction 1 (mind vs. body): Were going to play a game. Im going to show you some verbs written on cards. These verbs can express so mething you can do mostly with your body or mostly with your mind. For example, you mostly use your mind to think and you mostly use your body to run. Were going to separate the two kinds of verbs into two piles. On e pile will be things you do mostly with your mind, and the other pile wi ll be things you mostly do with your body. Instructions 1: Now its your turn. Put all the verbs you do mostly with your mind in this pile, and all the verbs you do mostly with your body in this pile. Scripted prompt 1: Tell me if that verb is something you do mostly with your mind or mostly with your body. Introduction 2 (communication vs. body): Were going to play another game. Im going to show you more verbs written on cards. These verbs can express what you do mostly when communicating or something you do mostly with your body. For example, you can communicate or talk with a yell or a whisper and you can use your body to crawl. Were going to sepa rate the two kinds of verbs into two piles. One pile will be th ings you do mostly when you are
94 Appendix I: (Continued) communicating, and the other pile will be things you mostly do with your body. Instructions 2: Now its your turn. Put all the verbs you do mostly when you communicate in this pile, and all th e verbs about things you mostly do with your body in this pile. Scripted prompt 2: Tell me if that verb is something you do mostly when you communicate or mostly with your body. Narrative Task Introduction: Im writing a book and need some help coming up with a good story to go with the video Im going to show you. This video has many different characters. There is a boy, a frog, a mom, a dad, a cook, a waiter, and some other people. Your teacher told me that youre a good storyteller. I need you to help me make up a good story to go with the video. You can watch the video while I do some work over here. Instructions: I want you to make up the best story that you can to go with the video. Remember that to tell/wri te a good story, you should talk about whats going on in the characters minds. You need to say what they might be thinking, saying, and feeling. Tell me what you think they might be thinking, feeling, and saying. Here are some pictures from the story, to help you remember.
95 Appendix I: (Continued) Scripted questions (as needed): (1) What was going on in the characters minds? (2) What else could be going on in the characters minds? (3) What might have this character been saying? (4) Tell me how the characters are using their minds. (5) Why do you think that character did that? (6) How does the character feel? (7) If I could listen to the charac ters talking, what would I hear? Scripted questions for frames: 1a. What is going on in the little boys mind? 1b. If I could listen to the boy talk, what would I hear? 2. This is the guy that parked the car. Whats going on in his mind? 3. Whats going on in the frogs mind? 4. Whats going on in the little boys mind? 5. Look at her face. Whats going on in the ladys mind? 6. I wonder what hes doing. Whats going on in the chefs mind? 7a. There are the parents. Whats going on in the parents minds? 7b. If I could listen to the pare nts talking, what would I hear? 8a. What is going on in the parents minds? 8b. If I could listen to the pare nts talking, what would I hear?
96 Appendix I: (Continued) If the written narrative is not readable, the following script will be employed: Okay, I want to make sure that I have th is right. Could you r ead it to me out loud so that I can tape record it?
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 001681062
007 cr mnu|||uuuuu
008 051222s2005 flu sbm s000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001056
Federico, Jeanne E.
Oral/written contrast of mental state references in older children
h [electronic resource] /
by Jeanne E. Federico.
[Tampa, Fla.] :
b University of South Florida,
Thesis (M.S.)--University of South Florida, 2005.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
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 103 pages.
ABSTRACT: The purpose of this study was to determine differences in the use of mental state references by typically developing 9- and 11-year-old children. Following a priming task that emphasized distinctions between physical and mental acts, children watched an 11-minute textless video and then were asked to generate an oral and a written story that focused on the mental states of the multiple characters. Narratives were transcribed, and all mental state references were classified into motivational, experiential, and belief categories. Specific mental state references were also analyzed to determine levels of semantic complexity.The study attempted to answer: 1) whether 9- and 11-year-old typically developing children differed in their ability to use more complex mental state references and 2) whether this ability varied as a function of the oral versus the written modality.The sample consisted of 26 children, ages 9;0-9;11 (15 females, 11 males), and 24 children, ages 11;00-11;11 (14 females, 10 males). The total sample (N=50) consisted of 32 Caucasian children, 15 African-American children, 2 Hispanic children, and 1 Asian-American. All children were selected from one urban elementary school located in West Central Florida, were from monolingual, English-speaking homes, and were speakers of Standard American English. A statistical analysis was conducted via a 3-way MANOVA, specifically, 2 (age 9 vs. age 11) x 4 (mental state categories) x 2 (oral versus written modality).
Adviser: Elaine R. Silliman.
Theory of mind.
x Speech-Language Pathology
t USF Electronic Theses and Dissertations.