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 001920445
007 cr mnu|||uuuuu
008 080109s2007 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002124
Reconciliation in mandrills (Mandrillus sphinx)
h [electronic resource] /
by Pete Otovic.
[Tampa, Fla.] :
b University of South Florida,
ABSTRACT: This study aimed to examine whether mandrills (Mandrillus sphinx) reconcile their conflicts. The data were collected from a captive group of nine mandrills (5 males and 4 females) at the Lowry Park Zoo that ranged in age from 3 to 16 years at the time of study. After a conflict was observed, the behavior of one of the two former opponents was documented for a period of ten minutes using continuous recording methods. On the next possible observation day, at the same time of the previous conflict, the behavior of the same individual was recorded for an additional ten minutes. Former opponents exchanged peaceful or affiliative signals sooner after a conflict than during control periods. These post-conflict signals were selectively directed towards former opponents, and were most likely to be exchanged in the first two minutes after a conflict's termination. The silent bared-teeth face comprised 62.5% of the first peaceful interactions between former opponents. The best predictor of the likelihood of reconciliation was the dyad's baseline rate of silent bared-teeth face exchange. Mandrill dyads with higher rates of silent bared-teeth face exchange at baseline had higher conflict rates and spent less time in non-aggressive proximity than those with lower rates of silent bared-teeth face exchange. These results are consistent with the Insecure Relationship Hypothesis, which posits that individuals with insecure relationships are more likely to reconcile because their relationships are more likely to be damaged by a conflict than those with secure relationships. The exchange of peaceful post-conflict signals did not appear to have an effect on the behavior of the former opponents.
Thesis (M.A.)--University of South Florida, 2007.
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 66 pages.
Advisor: Sarah Partan, Ph.D.
Communication conflict resolution.
t USF Electronic Theses and Dissertations.
Reconciliation in Mandrills (Mandrillus sphinx ) by Pete Otovic A thesis prepared in partial fulfillment of the requirements for the degree of Master of Arts Department of Psychology College of Arts and Sciences University of South Florida Co-Major Professor: Sarah Partan, Ph.D. Co-Major Professor: Judith Bryant, Ph.D. Jon Rottenberg, Ph.D. Ann Weaver, Ph.D. Date of Approval: May 21, 2007 Keywords: aggression, agonism, communi cation conflict resolution, primate Copyright 2007, Pete Otovic
Acknowledgements I would like to thank Judy Bryant, Sarah Partan, Jon Rottenberg, and Ann Weaver for their invaluable guidance and suggestions. I woul d also like to thank Sara Saiger for all of her love and support. Finally, I would like to thank the Lowry Park Zoo staff, including Angela, Tracy, Jane, Chris, and Lee Ann fo r all of their assistance, support, and transcendental cooperation.
i Table of Contents List of Tables .....................................................................................................................iii List of Figures ....................................................................................................................iv Abstract ...............................................................................................................................v Chapter One ........................................................................................................................1 Group Life ...............................................................................................................1 Conflict Management ..............................................................................................2 Reconciliation .........................................................................................................4 Effects of Reconciliation.........................................................................................6 Third-Party Affiliation ............................................................................................6 Why Animals Reconcile.........................................................................................7 Predictors of Reconciliation ..................................................................................10 Variation in Conciliatory Tendency ......................................................................11 Mandrills ...............................................................................................................13 Do Mandrills Reconcile? ......................................................................................14 Hypotheses to be Tested .......................................................................................16 Chapter Two .....................................................................................................................18 Subjects .................................................................................................................18 Data Collection .....................................................................................................19 Baseline Data Collection ...........................................................................19 Conflict Data Collection ...........................................................................19 Post-Conflict Observations .......................................................................19 Matched-Control Observations .................................................................20 Chapter Three ...................................................................................................................22 Reconciliation .......................................................................................................23 Reconciliation with the PC-MC Method ..................................................23 Reconciliation with the Time Rule ...........................................................24 Reconciliation with the Rate Method .......................................................24 Selective Attraction ...................................................................................25 Silent Bared-Teeth Face as Conciliatory ..................................................25 Additional Analyses ..............................................................................................26 Conciliatory Tendency ..............................................................................26 Sex Effects ................................................................................................27 Kinship Effects ..........................................................................................27 Effects of Percentage of Time in Peaceful Proximity ...............................28 Effects of Baseline Silent Bared-Teeth Face Exchange Rate ...................29 Effects of Peaceful Signal Exchange After a Conflict ..........................................30 Displacement Behavior .............................................................................30 Relationship Disturbance ..........................................................................31 Chapter Four .....................................................................................................................33
ii Limitations and Future Directions ........................................................................42 List of References .............................................................................................................46
iii List of Tables Table 1 Typical Approaches to Analyzing Reconciliation Data, 53 Operationalization of Conflicts and Operationalization of Affiliation Table 2 Demographics of Mandrills in Sample 54 Table 3 Mandrill Ethogram 55 Table 4 Summary of Conflicts 58 Table 5 Baseline Silent Bared-Teeth Face Rate, CCT, and Number of 60 Conflicts by Dyad Table 6 CCTs as a Function of Baseline SBTF Rate with CCTs Calculated 61 For Entire Group Table 7 Rates of Displacement Behavior After Conflicts Followed by 62 Peaceful Signals (a) and After C onflicts not Followed by Peaceful Signals (b) Between Former Opponents
iv List of Figures Figure 1 Patterns of post-conflict behavior 63 Figure 2 Peaceful signal exchange over time in post-conflict and matched63 control observations Figure 3 Rate of silent bared-teeth face exchange in post-conflict and 64 matched-control observations Figure 4 Percentage of s ilent bared-teeth face exchange with former 64 opponents in PC and MC Figure 5 Conflict rate as a function of % time in peaceful proximity 65 Figure 6 Conflict rate as a functio n of baseline SBTF exchange 65 Figure 7 % of 1 st peaceful contacts in post-conflict and matched-control 66
v Reconciliation in Mandrills (Mandrillus sphinx ) Pete Otovic ABSTRACT This study aimed to examine whether mandrills ( Mandrillus sphinx ) reconcile their conflicts. The data were collected from a captive group of nine mandrills (5 males and 4 females) at the Lowry Park Zoo that ranged in age from 3 to 16 years at the time of study. After a conflict was observed, the behavior of one of the two former opponents was documented for a period of ten minutes using continuous recording methods. On the next possible observation day, at the same time of th e previous conflict, the behavior of the same individual was recorded for an additiona l ten minutes. Former opponents exchanged peaceful or affiliative signals s ooner after a conflict than duri ng control periods. These postconflict signals were selectively directed towa rds former opponents, and were most likely to be exchanged in the first two minutes after a conflicts termin ation. The silent bared-teeth face comprised 62.5% of the first peaceful interactions between former opponents. The best predictor of the likelihood of r econciliation was the dyads baselin e rate of silent bared-teeth face exchange. Mandrill dyads with higher rate s of silent bared-teeth face exchange at baseline had higher conflict ra tes and spent less time in non-aggressive proximity than those with lower rates of silent bare d-teeth face exchange. These re sults are consistent with the Insecure Relationship Hypothesis, which posits that individuals with insecure relationships are more likely to reconcile b ecause their relationships are mo re likely to be damaged by a
vi conflict than those with secure relationships. The exchange of peaceful post-conflict signals did not appear to have an effect on th e behavior of the former opponents.
Mandrill Reconciliation 1 Chapter One Introduction Group Life Living in stable social groups can have many benefits as well as liabilities for wild animals. Gregarious animals may enjoy decreased predation rates due to increased alertness for approaching predators (Caro, 1986; va n Schaik, 1983), dilution (Wrona & Dixon, 1991), confusion of the predator (Fel s, Rhisiart, & Vollrath, 1995) a nd cooperative defense against predators (Bertram, 1975). Group-living animal s may also acquire foraging benefits, which include obtaining prey that require cooperative efforts (C reel & Creel, 1995) and the facilitation of finding food (Templeton & Gira ldeau, 1995). Finally, finding a mate is made easier for social animals (Wiley, 1991). On the other hand, in addition to increased risk of infection from parasites (van Vuren, 1996), greg arious animals are susceptible to having conflicts of interest over lim ited resources such as food a nd mates (Krause & Ruxton, 2002), direction and speed of travel (Menzel, 1993), and time spent performing cooperative tasks (van Schaik & van Noordwijk, 1986). A conflict of interest among group-living anim als can have various immediate and longterm outcomes. Immediately following conf lict, opponents may show a range of responses, from tolerance and avoidance of open conflic t to aggression (de W aal, 2000). Aggressive encounters may yield negative consequences, whic h include the risk of physical injury from the initial encounter (Set chell, 2005), renewed attack following an initial conf lict (Aureli & van Schaik, 1991; York & Rowell, 1988) and damage to the quality of social relationships (Aureli,
Mandrill Reconciliation 2 Cords, & van Schaik, 2002). Because these c onsequences can disturb typical interaction patterns, conflict may reduce the benefits of living in a group. For example, Aureli (1992) found that wild long-tailed macaques foraged for shorter periods of time following an aggressive conflict than during ba seline. He hypothesized that this is due to reduced tolerance of the recipient of aggression ar ound a preferred resource (e.g., f ood) or to the fact that the recipient must pay more attention to the other group members to avoid renewed attack, thereby diverting its attention away from foraging. Aggressive conflict also has non-social consequences. Both aggressors (Aureli, 1997; Castles & Whiten, 1998b) and recipients of aggressive behavior (Aureli, van Schaik, & van Hooff, 1989) increase displacement behavior (e.g., self-scratching) after a c onflict, which indicates uncertain ty and anxiety in non-human primates (Maestripieri, Schino, Aureli, & Troi si, 1992; Schino, Schucchi, Maestripieri, & Turillazzi, 1988). Uncertainty and anxiety are stress responses; persistent activation of stress responses reduces an individuals fitness by im pairing immune system functioning, growth and development, reproductive ability and by causing brain damage (Henry, 1982; von Holst, 1985; Kaplan, 1986; Sapolsky, 2005; Uno, Tarara, Else Suleman & Sapolsky, 1989). Both social and non-social consequences of conflict provide the impetus for individuals to develop means to reduce or mitigate it. Conflict Management Due to the adverse consequences of aggr ession, conflict management strategies are thought to have evolved both to prevent and re pair damage following aggression (Preuschoft & van Schaik, 2000). One way to manage conflic t is through the use of honest signals that communicate an individuals competitive ability and thus help to prevent conflict from escalating into full-blown aggression These signals, such as the bl ue throat color in male tree
Mandrill Reconciliation 3 lizards, Urosaurus ornaus (Thompson & Moore, 1991), white forehead patches in male collared flycatchers, Ficedula albicollisare (Part & Qvarnastrom, 1997) and the red and violet coloration on the face, rump, and genitalia of male mandrills, Mandrillus sphinx (Setchell & Wickings, 2005), are referred to as badges of st atus (Rohwer, 1982) and are usually triggered by the release of hormones. In animals that live in stable groups, a second way to manage aggressive conflict is through the formation of dominance hierarchie s (de Waal, 1986), which serve to regulate access to pref erred resources when they are defendable (Silk, 1987). A third way is through ritualized beha vioral displays, which are so metimes nested in dominance hierarchies and function to communicate individual emotions or intentions (Hinde, 1985). Such communications can prevent conflict from intensifying into phys ical aggression. Examples include threat displays (e.g., staring in baboons and ches t beating in gorillas; Estes, 1991), formalized indicators of dominance (e.g., mock biting in stumptail macaques; Demaria & Thierry, 1990), and formalized indicators of subordinance (e.g., the silent bared-teeth display in rhesus macaques; de Waal & Luttrell 1985; the pant-grunt and bob display in male chimpanzees, Nishida & Hiraiwa-Hasegawa, 1987). These displays tend to be unidirectional in species with dominance hierar chies, meaning that a formal dominance (or subordinance) indicator is only emitted by the dominant (or subordinate) individual in a dyad. In order to maintain the groups integrity, conflict management also must include a means to resolve conflicts that do escalate in to aggression (de Waal, 1987). Affiliative postconflict reunions between former opponents may serve to mitigate the effects of aggressive conflicts, which would decrea se some of the risks of aggressive encounters.
Mandrill Reconciliation 4 Reconciliation Aggression has historically been viewed as an anti-social instinct, virtually impossible to control, that serves to disperse conspecifics in order to facilitate more equitable access to resources (Lorenz, 1966). Research that cont ributed to this viewpoi nt involved studying aggressive interactions between individuals that did not need one another, did not know each other, and would not be likely to see one another again (J ohnson, 1972). Hence, aggression was thought to result only from the expression of internal factors such as hormones and genes and external factors such as past experience ; the consequences of aggression for social relationships were ignored. This perspective is sometimes refe rred to as the Individual Model of Aggression (de Waal, 2000). It predicts th at 1) aggression will be rare among closely bonded individuals, 2) contact following aggressi on will be aggressive (due to motivational continuity), and 3) aggression will result in dispersal of the opponents. However, in a pioneering study, de Waal and van Roosmalen (1979) discovered that following an aggressive interaction in chimpanzees, former opponents so ught friendly contact with one another shortly after a conflict. More specifica lly, this crucial stu dy revealed a higher frequency of affiliative behavior between former opponents following a c onflict than during baseline. In addition, there were specific behaviors, such as kissi ng and embracing, which occurred more frequently in the first post-conflict interaction than in late r post-conflict interactions. This affiliative postconflict contact was labeled reconciliation. This re search was a major contribution to the gradual paradigm shift from a focus on th e strictly negative connotations surrounding aggression to a standpoint that considers a ggression as a social means of negotiating relationships arising from a c onflict of interest, which is kno wn as the Relational Model of Aggression (de Waal, 1996). In contrast to th e Individual Model of A ggression, the Relational
Mandrill Reconciliation 5 Model predicts that 1) aggression and peaceful post-conflict interactions will be common among closely bonded individuals, 2) post-conf lict contact between opponents will be affiliative and will occur at a higher rate than at baseline, and 3) peaceful post-conflict interactions will reduce anxiety of the opponents and restore baseline proximity between them (de Waal, 2000). Reconciliation may take a number of forms. It includes shorter latencies to friendly contact (de Waal & van Roosmalen, 1979) or frie ndly vocalizations (Silk et al., 1996) between former opponents after a conflict re lative to baseline or to a matched-control time period. Some researchers also label th e post-conflict interact ion reconciliation if the former opponents are in proximity to one another sooner afte r the conflict than duri ng a corresponding matchedcontrol or baseline period (C ords, 1993). Former opponents in many gregarious animals besides chimpanzees have demonstrated an increased tendency to affiliate in one way or another shortly after a conflic t. These include mountain gorillas (Watts, 1995), bonobos (de Waal, 1987), capuchins (Verbeek & de Waal, 1997), vervet monkeys (Cheney & Seyfarth, 1989), patas monkeys (York & Rowell, 1988) sooty mangabeys (Gust & Gordon, 1993), golden monkeys (Ren et al., 1991), colobus mo nkeys (Bjornsdotter et al., 2000), spectacled langurs (Arnold & Barton, 1997) baboons (Castles & Whiten, 1998a; Petit & Thierry, 1994a; Silk et al., 1996; Swedell, 1997; Zaragoza & Colmenares, 1997), macaques (Abegg et al., 1996; Aureli et al., 1994; Aureli et al., 1997; Aureli et al., 1989; Demaria & Thierry, 2001; Judge, 1991; Matsumura, 1996; Petit & Thierry, 1994b; de Waal & Ren, 1988; de Waal & Yoshihara, 1983), lemurs (brown lemurs, Kappe ler, 1993; ringtailed le murs, Palagi et al., 2005), dolphins (Weaver, 2003), hyenas (Hofer & East, 2000; Wahaj et al., 2001), and domestic goats (Schino, 1998). Anecdotal evidence has also b een reported for mouflons
Mandrill Reconciliation 6 (Pfeffer, 1967), lions (Schalle r, 1972), dwarf mongooses (Rasa, 1977), and feral sheep (Rowell & Rowell, 1993). Effects of Reconciliation As predicted by the Relational Model of Aggr ession, peaceful post-conflict interactions between former opponents have many positive effect s on the individuals involved. First, after such peaceful reunions, the risk of renewed attack is dramatically reduced relative to unreconciled conflicts (Aureli & van Schaik, 1991b; Castles & Whiten, 1998a; Cords, 1992). Second, baseline tolerance between former oppone nts is restored following a peaceful reunion (Cords, 1992), which suggests reduced fear in the subordinate and reduced aggression in the dominant individual. Third, following a friendl y reunion, rates of selfdirected behavior are reduced in both the aggressor and the recipi ent of aggression relative to both unreconciled conflicts and those conflicts wh ere affiliative post-conflict co ntact was sought from a third party member (i.e., an individual not involved in the conflict) (Aur eli, 1997). Finally, rates of affiliative behavior are restored to baseline le vels following a peaceful reunion between former opponents (Koyama, 2001). Third-Party Affiliation Peaceful post-conflict interactions may also be observed between one of the individuals involved in a conflict and a diffe rent group member not involved in the conflict. This is referred to as post-conflict third-party affiliati on (Call, Aureli, & de Waal, 2002). There are two basic forms of third-party aff iliation after a conflict. Solicitation refers to when one of the animals involved in a conflict in itiates affiliative interaction from a bystander, and consolation refers to when the bystander initiates affiliative interaction with one of the animals involved in the conflict (Palagi et al., 2004). Consolation in particular has only been observed thus far in
Mandrill Reconciliation 7 chimpanzees (de Waal & van Roosmalen, 1979), bonobos (Palagi et al., 2004), and stumptail macaques (Call, Aureli, & de Waal, 2002). A lthough it is relatively ra re, consolation may replace reconciliation un der certain circumstances, such as when the former opponent is not available for interaction (de Waal & Aureli, 1996). Why Animals Reconcile Thus far, two hypotheses explain the majority of the occurrences of and variation in peaceful post-conflict interac tions, the Valuable Relationshi p hypothesis and the Uncertainty Reduction hypothesis. The Valu able Relationship hypothesis sugge sts that individuals within dyads reconcile in order to repa ir the damage to their relations hips because these relationships confer fitness benefits to the in dividuals. According to this hypothesis, reconciliation may be viewed as communication between conspecifics about the valu e of each relationship (van Schaik & Aureli, 2000). Individuals within more valuable dyadic relationships are predicted to reconcile at a higher rate than individuals within less valuable dyadic relationships (Cords & Aureli, 2000). Researchers disa gree about how to measure valu e in a relationship. Some posit that dyads with strong bonds, which include high rates of affiliative interaction and frequent proximity, are likely to derive value from their relationship (Kummer, 1978). Others have suggested that high rates of affiliation are only indicative of the compatibility of the dyad and should not be confused with a valuable rela tionship, which should be characterized by an exchange of benefits that increase the fitness of each indivi dual in the dyad (Cords & Aureli, 2000). Behavioral qualities or dispositions that could increase anothers fitness (and hence be of value) include toleran ce around preferred resources, food sharing, cooperation while hunting, protection against predation and othe r conspecifics, reproductive receptivity, and support in aggressive encounters (van Schaik & Aureli, 2000). Beneficial behaviors confer
Mandrill Reconciliation 8 value to a partner, but they are predicated on both the availability of the partner for advantageous relations and individual traits (such as size, dominance status, reproductive condition) that make the rela tionship more valuable (Cords & Aureli, 2000). Empirical evidence supporting the Valuable Relationshi p hypothesis has been documented by Aureli (1997), who found that long-tailed macaques displa yed a higher self-scratching rate after a conflict with individuals with whom they exchan ged a high rate of affiliative behavior than after conflicts with individuals without such bonds. Aureli also found that those individuals with strong dyadic bonds (and who displayed in creased self-scratching after a conflict) reconciled more often than individuals with weaker bonds. Perhaps the strongest evidence supporting this hypothesis is the experimental research of Cords and Thurnheer (1993), who discovered that pairs of long tailed macaques in creased their reconcilia tion rate substantially after they were trained to coope rate in order to obtain food. The Uncertainty Reduction hypothesis suggests th at individuals reconcile in order to signal benign intentions and terminate the fight ing, thereby attenuating th e uncertainty of the former opponents (Aureli & van Schaik, 1991b; S ilk, 1996). This hypothesis is based on the assumption that, after a conflict, opponents display increased rate s of displacement behavior as a result of the anxiety arising from risk of re newed aggression and possibly the status of the damaged relationship (Aureli, 1997; Aureli, Cords, & van Schaik, 2002). Evidence that supports the reduction of uncertainty includes a decrease in self-dir ected behavior and increased tolerance between former opponents after reconciling (Aur eli & van Schaik, 1991b; Cheney et al., 1995; Cords, 1992). It is plausible that both hypotheses are corr ect, the former being an ultimate explanation of why gregarious animals reconcile and the la tter being a proximate explanation (Cords &
Mandrill Reconciliation 9 Aureli, 1996). Although the two hyp otheses are not mutually exclusive, Silk (1996) has argued that the empirical ev idence seems to support the Uncertainty Reducti on hypothesis, whereas the evidence supporting the relationship repa ir function of the Valuable Relationship hypothesis is ambiguous. As an alternativ e explanation for a higher occurrence of reconciliation in dyads with va luable relationships, Silk hypothe sized that since peaceful postconflict interactions facilitate future friendly in teractions, and individu als within dyads with strong bonds are more highly motivated to interact affiliatively than individuals within dyads without strong bonds, individuals with strong bonds would be more likely to exchange peaceful post-conflict behavior than those without such bonds. Furthermore, Silk conjectured that in order to conclude that reconciliation has a relationship repair function, long-term effects on the relationship should be demonstrated. A lthough results from her research with baboons did demonstrate an increase in affiliative vocal izations (e.g., grunts) in the ten minutes after a conflict relative to a matched-control period, they did not reveal an increase in affiliative behavior between reconciled opponents in the te n days following the co nflict relative to the affiliative behavior between unreconciled oppon ents (Silk et al., 1996). Based on this evidence, Silk concluded that the function of peacef ul post-conflict interactions is not to repair relationships, but merely to signal the end of a conflict with no long-term guarantees. A long-term increase in affiliative behavi or after a reconciled conflict may not be required to demonstrate a relationship repair function. Instead, restor ation of affiliative behavior to baseline le vels, along with a reduction in aggres sive behavior, ma y be sufficient (see Aureli, Cords & van Schaik, 2002). As ev idence to support this idea, Koyama (2001) found that affiliative behavior between Japa nese macaques in the ten days following a reconciled conflict returned to ba seline levels. She also found a decrease in affiliative and
Mandrill Reconciliation 10 increase in aggressive behavior among dyads in the ten days following unreconciled conflicts relative to baseline. Predictors of Reconciliation The majority of published research on reconciliation in gregarious non-human primates includes evidence supporting its existence, although the tend ency to reconcile varies considerably both within and be tween groups and species. The lo west percentage of peaceful post-conflict behavior has been reported by Chen ey and Seyfarth (1989), who found that vervet monkeys reconcile only seven percent of their conflicts. de Waal and Ren (1988) found that stumptail macaques reconcile 56 pe rcent of their conflicts, which is the highest percentage of conflicts followed by affiliative behavior thus far. According to de Waal (2000), the conditions for a particular spec ies to reconcile incl ude an ability to discriminate among individuals in a group, the existen ce of conflicts of interest betw een group members, an ability to remember previous conflicts, and advant ages to the preservation of cooperative relationships. Aureli, Cords, and van Schaik (2002) suggested that within-group aggression, not simply conflicts of interest, must also be a requisite, since conflict resolution may not be as pervasive in groups where conflict management st rategies prevent the aggressive escalation of conflicts. Kappeler (1993) failed to find reconciliation in a semi-captive group of ringtailed lemurs, Lemur catta but also found low rates of aggressi on between individuals that tolerate and/or support each other. Aureli et al. (2002) made a few further amendments to de Waals stipulations. One modification is that for reconciliation to occur there must be increased risks for renewed aggression in the period of time imme diately following a conflict, since one of the primary effects of peaceful pos t-conflict reunions is a re duced probability of renewed aggression between former opponents. Reconcil iation is also predicted to be a conflict
Mandrill Reconciliation 11 management strategy in groups where the cons equences of aggression lead to a loss or reduction in advantages derived from the relationship between the two individuals involved. Hence, Aureli et al. (2002) pred icted that if the advantages as sociated with an attempt at reconciling outweigh the costs of the risks of renewed aggression, peaceful post-conflict reunions are likely to occur. Schaffner and Caine (2000) did not find reconciliation in redbellied tamarins, Saguinus labiatus, but no loss of benefits seemed to arise from the preceding conflict between valuable relationship partners. After a conflict, baseli ne behavior patterns appeared to be restored without any obvious attempts at reconciliation (Schaffner et al., 2001). Assuming that one function of r econciliation is the repair of a damaged relationship (the Valuable Relationship hypothesis), one would pred ict that individuals wi thin dyads with more valuable relationships would display higher conciliatory rates th an those with less valuable relationships within the same group. Watts (1995) found that fe male gorillas reconcile with male gorillas, but not with each other. Since fe male gorillas only form valuable relationships or strong bonds with males, damage to a re lationship with another female would not necessarily result in a loss of benefits. Variation in Conciliatory Tendency Other factors likely to explain apparent with in-group variation in conciliatory tendencies include the security of a rela tionship (Cords & Aureli, 2000) a nd the compatibility of the two individuals (Cords & Aureli, 2000). The security of a relationship is defined by the consistency of the behavior of each individual in a dyadic relationship towards each other, which can be measured by observing the signs of tension during an approach, the presence of appeasement or friendly gestures during an appr oach, or the directness with which an approach is made (Cords & Aureli, 2000). Although this hypothesis has not been directly tested, there is
Mandrill Reconciliation 12 a small amount of evidence that is consistent with the idea. For example, Cords (1988) found that unrelated juvenile long-tailed macaques had a higher co nciliatory tendency than related juvenile individuals. She posited that relations hip security may be an important factor for conciliatory tendency, since all juveniles played with one another and were considered to have valuable and compatible relationships. Compatib ility of a relationship refers to how well the individuals get along and may be measured in many ways, including the Relationship Quality Index (RQI), which is a measure of affiliative beha vior relative to agonist ic behavior within a dyad (Weaver & de Waal, 2000). Differences in how behaviors are operati onalized are another cause of observed variability in conciliatory tendencies. Res earchers often vary in how they operationalize a conflict or affiliative behavior, and this may al so lead to the appearance of variation in conciliatory tendency both be tween and within groups and sp ecies (see Table 1). While studying chimpanzee reconciliation, Fuentes et al. (2002) required three or more non-contact agonistic behaviors to constitute a conflict, whereas de Waal and van Roosmalen (1979) and Preuschoft et al. (2002) only re quired one. de Waal and Re n (1988) defined conflicts among stumptailed macaques as interactions with faci al and vocal threats that are accompanied by a chase of at least two meters, whereas others have used all occurrences of aggression to indicate a conflict (e.g., Kappeler, 1993; Koyama, 2001). Preuschoft et al. (2002) did not include any vocalizations in their definitions of affiliative behavior, but Silk et al. (1996) did. In addition, Cords (1993) considered proximity to be affilia tion, whereas Palagi et al. (2005) did not. Finally, the analytical methods used to dete rmine the existence of reconciliation may be responsible for some variability. There are thre e methods that are typically used to determine whether dyads reconcile their conflicts within a group. The PC-MC method compares the
Mandrill Reconciliation 13 latencies to the first affiliative behavior between former opponents after a conflict (Post Conflict; PC) to those in a matched-control period (Matched Control; MC) (de Waal & Yoshihara, 1983). The rate method compares th e rate of affiliative behavior in the PC observation to rates during either the MC or baseline (Judge, 1991). Finally, the time rule compares the total number of first affiliative behaviors exchanged at each of ten one minute blocks in the PC to the total number of firs t affiliative contacts at the corresponding MC one minute blocks (Aureli et al., 1989). Research ers have obtained different results applying different methods to the same data. Fo r instance, Kappeler ( 1993) found evidence for reconciliation in brown lemurs using the time rule but not using the PC-MC method. Mandrills Expanding the array of group-li ving organisms in which rec onciliation is studied will further facilitate our ability to predict when rec onciliation is likely to oc cur and help illuminate its function. One species whose c onciliatory tendencies have yet to be examined are mandrills ( Mandrillus sphinx ). Mandrills are terrestrial, forest-d welling primates that reside throughout western Africa, including Gabon, Cameroon, Guinea and Congo (Grubb, 1973). Mandrills are one of the most sexually dimorphic primate sp ecies, with adult males being over three times the size of adult females (Wickings & Dixson, 1992) Adult males possess violet, red, and blue coloration on their snout, rump, and genitalia. The intensity of this pigmentation is highly positively correlated with dominance rank (Setche ll & Wickings, 2005) and serves as a social badge of status. The species forms female ph ilopatric groups (they remain in their native group) that may number as high as 600 individuals in the wild (Abernathy et al., 2002). The female philopatric groups have been reported to include one or more permanent adult males (Rogers et al., 1996), although male presence may only be seasonal (during breeding season,
Mandrill Reconciliation 14 Abernathy et al., 2002). Adult males vary in their group association, ranging from being solitary to living in the groups pe riphery to being intimately asso ciated with the group (Rogers et al., 1996; Setchell & Dixson, 2001). In the wil d, males leave their natal groups before they reach adulthood (9-10 years of age) (Abernathy et al., 2002). In captivity, male emigration is replaced by peripheralization (Setchell & Dixson, 2002). The intensity of male coloration is believed to affect the probability of violent aggression between unfamiliar conspecifics (Preuschoft & van Schaik, 2000) The association between coloration intensity, conf lict and reconciliation remains to be explored. Violent aggressive encounters between male mandrills are particularly risky because of the males large canines, up to 44 mm or 1.73 inches (Setchell & Dixs on, 2002). Not surprisingly, mandrills employ formal signals or indicators of submission, which include fleeing/avoi dance, screaming and presentation, and of dominance, which incl ude chasing and lunging (Setchell & Wickings, 2005). Do Mandrills Reconcile? Mandrills seem to be an ideal candidate in which to investigate reconciliation since they meet some but not necessarily all of the requisi tes proposed by de Waal (2000) and Aureli et al. (2002). Several lines of evidence suggest that mandrills are not lik ely to reconcile their conflicts, males in particular, which have been the focus of the majority of the previous research. First, adult males spend little time in each others company (Setchell & Wickings, 2005). Second, they do not seem to form coope rative alliances or coalitions (Setchell & Wickings, 2005), which is thought of as one of the most important aspects of a valuable relationship (van Schaik & Aureli, 2000). Third, in a despotic species such as mandrills, fear of approaching the dominant individual may reduce the likelihood of reconciliation (de Waal &
Mandrill Reconciliation 15 Ren, 1988). Preuschoft and van Schaik (2000) pred icted that in despot ic species, dominant individuals may not rely on cooperation from subordinate individuals, since if dominants needed the support of subordinates, the subor dinates would possess some leverage (e.g., withholding cooperation; Vehrencam p, 1983) that they could manipul ate to force the dominant individual to become more egalitarian. Fi nally, mandrills may not display enough aggression to warrant the development of post-conflict reunions; formal indicators of dominance and submission may mitigate open conflicts. In contrast, other evidence suggests that mandr ills would be likely to reconcile. First, juvenile and adolescent males have high rates of play and other affiliative interactions with one another (Charpentier, Peignot, Hossaert-McKey, & Wickings, 2004). Play may be a valuable aspect of mandrill relationships, considering the importance of play in developing social and survival skills. Second, mandrills are terrestrial foragers and primarily herbivorous (Mellen et al., 1981). They may be tolerant of conspecifi c proximity because resources are scattered, and this tolerance may be sufficient to constitute relationships with enough value to warrant their repair. Third, even though physical conflict is rare in adult male mandrills (Setchell & Wickings, 2005), adolescent males engage in a rela tively high rate of aggression (Charpentier et al., 2004). Fourth, females are philopatric and form matrilineal hierarchies among related females so they seem likely to form valuable relationships and reap the accompanying benefits (such as kin based agonistic support). de W aal and Ren (1988) did not find an effect for kinship on reconciliation rates in another female philopatric species (rhesus macaques, Macaca mulattta ) but that was probably because the effects for bond strength had been factored out. Thus, it is possible that female mandrills onl y form strong bonds with their kin. Fifth, mandrills are thought to be closely related to baboons (Stammbach, 1987) and mangabeys
Mandrill Reconciliation 16 (Disotell, 1994), and both of t hose species reconcile (Gust & Go rdon, 1993; Silk et al., 1996). Sixth, males and females may form valuable relationships, which could possibly warrant reconciliation between sexes. A male that has a close relationship with a female could benefit from the females reproductive receptivity. Due to the extreme sexual dimorphism in mandrills, having a close relationship with a male could be very valuable to a female because the male is capable of protecting her from ot her sexually harassing males or her infants from infanticidal males (Smuts, 1985). A relationship that offers protection of offspring is believed to be one of the most valuable relationships (a long with those that offer agonistic support) formed in non-human primates (van Schaik & Aureli, 2000). Seventh, the function of the silent bared-teeth face (SBTF) in mandrills, a sign al in which the animal retracts its lips in a horizontal figure eight shape and thereby displays its canines and premolars, has been poorly understood; its interpretation has ra nged from aggressive to aff iliative in nature (see Laidre & Yorzinski, 2005, for a short review). Laidre a nd Yorzinski (2005) have recently suggested that the silent bared-teeth face serves a conciliato ry role; mandrills were more likely to exchange silent bared-teeth faces after an agonistic interaction than befo re one. The silent bared-teeth face may signal benign intentions and thus may have evolved for use in peaceful post-conflict reunions. Additional work by Bout and Theirry ( 2005) demonstrated that the silent bared-teeth face was mostly exchanged in peaceful situations such as friendly contact, play, mating or socio-sexual interactions, and friendly following. But they also reported that mandrills are likely to produce the silent bare d-teeth face after an aggressive exchange. Importantly, usage of the signal did not covary with dominance, meaning that it is unlikely to communicate information about social status. Hypotheses
Mandrill Reconciliation 17 The purpose of this study was to examine c onflict and post-conflict interactions in a group of captive mandrills. This study was de signed to test the following hypotheses: (1) H 1 : Mandrills reconcile. Predictions: (a) The number of attracted pairs is greater than the number of dispersed pairs. (b) Mandrills displa y shorter latencies to peaceful signal exchange with their former opponents during post-conflict samples than during correspond ing matched-control samples. (c) The frequency of first peaceful signal exchanges between former opponents is greater in at least one one-minute block in the post-conflict observ ation than the mean rate of peaceful signal exchange during corresponding matched-control observations. (d) Former opponents display higher rates of peaceful signal exchange in post-conflict samples than during corr esponding matched-control samples. (e) The percentage of peaceful signals exchanged between former opponents is greater in post-conflict sample s than during corresponding ten minute matchedcontrol samples. (2) H 1 : The mandrill silent bared-teeth face (SBT F) serves a conciliatory role. Prediction: (a) The silent bared-teeth face will be more likely to be the first peaceful exchange between former opponents after a c onflict than during corresponding matchedcontrol observations.
Mandrill Reconciliation 18 Chapter Two Method Subjects Behavioral data were collected from a captive group of mandrills ( Mandrillus sphinx ) housed at the Lowry Park Zoo in Tampa, FL During the majority of the study, the group consisted of nine individuals: two adult female s, one adolescent female, one juvenile female, two adult males and three adolescent males (T able 2). The outdoor enclosure measures approximately 40 x 25, and the animals inha bit it from 0900 to 1700. They are indoors during all other times. The animals are fed at 0900 and 1700 and have ad libitum access to water. At the time of study, the group had b een intact for approximately six years, the exception being the youngest female, who was born in 2002. Due to a couple of severe aggressive encounters, the beta male (Milo) wa s permanently isolated from the alpha male and the females a little less than three weeks after data collection began. From that day forward, two groups were rotated on exhibit. On one day, the group consisted of Nestor, Miller, Mukobi, Moesha, Jalisa, Jinx, Jerome, and Jasper (all but Milo). On the following day, the (bachelor) group consisted of Milo, Mukobi, Jasper, and Jerome. This rotation was conducted on a daily basis for the rest of the study. Moreover, one and a half months later Mukobi was found dead at the bottom of the moat that separates the mand rill exhibit from the visitor viewing area. Thus, from that day forward the bachelor group consisted only of Milo, Jasper, and Jerome. In addition to the perpet ually changing nature of the mandrill groups, in
Mandrill Reconciliation 19 June of 2006 the three remaining adolescent male s were transferred out of the group to a new facility. Data Collection Three types of behavioral data were collected: baseline data, c onflict data, and postconflict data. Each drew on the behaviors from Table 3. Behavioral data were recorded on a Dell laptop computer with Noldus Observer 5.0 software. Data recording methods followed Preuschoft et al. (2000) and de Waal and Yoshih ara (1983) and were enti rely observational. Baseline data collection All occurrences of behavioral states and events in Table 3 were collected using 20-minute focal animal samples (Altmann, 1974). Focal individuals were randomly selected using a random numb er generator. Once a sample had been collected on all group members, the proce ss of random sampling began again. Conflict data collection Data collection on conflict s took priority over baseline observational recording. When conflicts occu rred during a focal animal sample, the focal sampling was discontinued and data were collect ed on the conflict. In this study, a conflict was characterized by particular non-physical and physical forms of aggression. Non-physical aggression was operationalized as ground slaps, head jerks and threat grunts (level 1 aggression; see Table 3) that were accompanie d by either lunging towards or chasing another animal (level 2 aggression). Physical aggr ession was defined as bi ting, hitting, or grabbing (level 3 aggression). When the conflict began, the identity of the aggressor, recipient, and level of aggression were record ed. The conflict was assumed to be ended immediately after the last aggressive excha nge (including level 1 aggre ssion) had terminated. Post-conflict (PC) data collection After the conflict ended, the distance between the opponents was immediately recorded. In addition, all occurrences of the behavior of one of
Mandrill Reconciliation 20 the opponents were recorded for ten minutes; this was the post-conflict or PC sample. If within two minutes of the PC s inception, further aggres sion between the individuals involved was observed, the PC obser vation started over. The majo rity of research indicates that increased affiliation in the PC period relativ e to the MC period is limited to the first two or three minutes after the observations inception (see Kapp eler & van Schaik, 1992). However, Rolland and Roeder (2000) had to use 60-minute PC observations in order to demonstrate reconciliation in ri ng-tailed lemurs. Therefore, a ten-minute PC duration was chosen as an intermediate duration (Aure li & van Schaik, 1991a). Which opponent was observed after a conflict depended upon how many PC-MC pairs the individual had previously been involved in, how many times each individual had been followed as an aggressor or recipient, and visibi lity of the former opponents. Pa rticular attention was paid to the frequency and timing of affiliative and peacef ul behavior, including the exchange of the silent bared-teeth face, lip-smacking, groom ing, head-shaking, playing, non-aggressive touching and peaceful proximity (Table 3). Pe aceful proximity was defined as any time an animal was within 2 meters of the focal anim al without exchanging a ny agonistic signals or performing any displacement behavior. This wa s typically characterized as two individuals sitting near one another without exchanging any overt signals, and seemed to be the most common affiliative behavior between male mandrills (personal observation). Matched-control (MC) data collection For each PC sample collected, a corresponding ten-minute matched-control (MC) sample was co llected from the same focal animal on the next day of observation and at approximately th e same time as the previous PC observation. The individual must not have been in a conflict in the ten minutes prior to the start of the MC observation. When possible, the researcher began a MC observation when the spatial
Mandrill Reconciliation 21 distance between the former opponents was appr oximately equal to the distance between opponents at the inception of the corresponding PC observation. In other words, if the researcher could not match the time of day and proximity of the former opponents from the post-conflict observation within the same week the initial conflict transpired, preference was given to matching the time of day (as an altern ative to throwing out the conflict). The aim of this criterion was to eliminate confounds due to initial proximity (Call, 1999).
Mandrill Reconciliation 22 Chapter Three Results Data were collected for a total of 51 dyadi c conflicts (see Table 4). Conflicts were recorded from 19 of a possible 36 dyads. 35 of the conflicts involved non-contact threats and chasing, whereas the remaining 16 conflicts involved physical aggression. Reconciliation data were analyzed in f our ways, each of which tests a different prediction concerning whether mandrills rec oncile. These include the PC-MC method (de Waal & Yoshihara, 1983), the time rule (Aureli et al., 1989), the rate method (Gust & Gordon, 1993), and selective attraction (de Waal & Yoshihara, 1983). The Corrected Conciliatory Tendency or CCT (Veenema, 1994) was used to determine the percentage of conflicts reconciled at group and dyadic levels. The data were analyzed for consolation using the PC-MC method. All analyses were two-tailed, and Wilcoxon matched pairs tests were used whenever possible in order to ensure that any group differences were not due to one or a few individuals. Results with probability levels of .05 or lower were considered significant, while those with probability le vels ranging from .06-.08 were considered tendencies or trends. All means are repor ted SE. Although there were 36 possible combinations of dyads, most analyses were lim ited to 32 of those dyads. This is because the beta male (Milo) was only in the group with the adult females for a very short time before the animals were separated into two groups (see Meth od). Milo did not interact with the females enough to provide any reliable data during th e limited time he had access to them.
Mandrill Reconciliation 23 Reconciliation Reconciliation with the PC-MC Method. One method that was used to determine whether the mandrills reconcile compares the latencies for former opponents to affiliate in the PC and MC samples. This is calle d the PC-MC method (de Waal & Yoshihara, 1983). If two former opponents exchange peaceful or affiliative behavior at an earlier time in the PC observation than in the MC obs ervation, they are labeled as an attracted pair. If they exchange peaceful or affiliative behavior earlier in the MC than in the PC, the pair is labeled as dispersed. Finally, if the former opponents exchange peaceful or affiliative behavior at the same time in each observation period (including if they fail to exchange any affiliative behavior in either sample) they are labeled neutral pairs (de Waal & Yoshihara, 1983). Using this met hod, the animals demonstr ate reconciliation if the number of attracted pairs is significantly greater than the number of dispersed pairs according to a Wilcoxon matched-pair signed ranks test. Figure 1 reveals that the number of attracted pairs ( N = 29) was significantly greate r than the number of dispersed pairs ( N = 6), Z = -2.68, N = 9, p= .008. This finding supports prediction 1a. While PC-MC pairs were labeled as attracted if the latency to affiliate was shorter in the PC than in the MC, the latency differences may not be significantly different statistically from one another. Thus, a second comparison was made between the mean latency to affiliate in the PC and the mean latency to affiliate in the MC using a two-tailed betweensubjects t-test. Former opponents exchanged pe aceful signals much ear lier after a conflict (62.94s 16.18s) than during corresponding ma tched-control sessions (270.97s 51.39s), t (19.28) = 3.86 p=.001 (rerun) This finding is consiste nt with prediction 1b.
Mandrill Reconciliation 24 Reconciliation with the Time Rule. The time rule was also implemented to determine whether there was evidence of reconciliation (Aureli et al., 1989; Kappeler, 1993; Veenema et al., 1994). This approach involves parsing the 10-minute PC and MC observations into ten one-minute blocks (ten for the PC, ten for th e MC) and recording the total number of first affiliative behaviors that occur within each mi nute block lumped across all PC-MC pairs. Next, the total number of first affiliative behaviors within each PC block is compared to the number within each corresponding MC block (e.g., total number of first affiliative behaviors exchanged in the PC from 0-1 minutes versus the per-minute rate of affiliative behaviors exchanged in the MC) using a Wilcoxon matched pa irs test (Aureli & van Schaik, 1991a). If the frequency of affiliative behavior in any of the one-minute PC sample blocks was greater than the per-minute MC rate, reconciliation was considered to have occurred. Figure 2 shows that the frequency of peaceful signa ls exchanged between former opponents was significantly elevated in the first ( Z = -2.67, N = 9, p = .008) and second ( Z = -1.96, N = 9, p = .050) minutes after a conflict relative to those in the matc hed controls. This finding corroborates prediction 1c. Reconciliation with the Rate Method. In contrast to the PC-MC method, the rate method compares rates of affilia tive behavior between former opponents in the PC sample to the rates of affiliative behavior derived from the MC sample. If the rate of affiliative behavior is significantly higher in the PC than in the MC using a Wilcoxon matched-pairs signed ranks test (Gust & Gordon, 1993), this group of mandrills will ha ve provided evidence of reconciliation. This analysis was limited to the rates of silent bare d-teeth face exchanges, since two-thirds of the first affiliative exchanges between former opponents after a conflict were silent bared-teeth faces (see below). Figure 3 shows that the per minute rate of silent
Mandrill Reconciliation 25 bared-teeth face exchanges was significantly greater between former opponents in the PC (.29 .11) than in the MC (.03 .01) periods ( Z = -2.67, N = 9, p = .008), which substantiates prediction 1d. Selective Attraction. To determine whether former opponents preferentially contact each other after a conflict and do not simply display overall higher interaction rates (which is an alternative explanatio n of greater observed attracted than dispersed pairs), the percentage of peaceful behavior exchanged between form er opponents in the PC sample was compared to the percentage in the MC sample. The percentage of peaceful signal exchange was calculated by dividing the cumulative freque ncy of peaceful signal exchanges with the former opponents by the cumulative frequency of peaceful signal exchanges with any group member. A Wilcoxon matched pairs test was used to test the pr ediction that the percentage of peaceful signals exchanged with the former opponent is greater in the PC than in the MC across all opponent pairs. If the percentage of peaceful signals exchanged with a former opponent was higher in the PC observation than in the corresponding MC observation, the pair was considered to be sele ctively attracted. This analysis was limited to the exchange of silent bared-teeth faces, since this was by far the most common peaceful signal exchanged after a conflict. Figure 4 indi cates that former opponents di rected a significantly greater percentage of silent bared-teeth faces towards one another in PC (.50 .08) than in MC (.16 .08) periods (Z = -2.429, N = 9, p = .015). This result is cons istent with prediction 1e. Silent bared-teeth face as conciliatory. To test the hypothesis that silent bared-teeth face exchanges serve a conciliatory role, the numbe r of silent bared-teeth face exchanges that were observed as the first exchange between former opponents after a conflict was compared to the number of silent bared-teeth face ex changes exchanged between former opponents in
Mandrill Reconciliation 26 matched-control observations using a chi-square goodness of fit test (de Waal & Yoshihara, 1983). If the frequency of silent bared-teeth face exchanges that comprise the first exchange between former opponents after a c onflict is found to be significan tly greater than the latter two frequencies, it would suggest that silent bared-teeth face exchanges are used as a conciliatory gesture. Figure 7 shows that silent baredteeth face exchanges comprised 64.5 % of the first peaceful exchanges between fo rmer opponents in the PC, relative to only 12.5 % in the MC. A chi-square analysis corrobo rated that this diffe rence was significant, X 2 ( N = 47, df = 1) = 11.47, p <.05. This finding substantiates prediction 2a. Additional Analyses Conciliatory Tendency. Since the data are consistent with the idea that mandrills reconcile, additional analyses were conducted in order to more closely inspect the distribution of peaceful post-conf lict signals. Thus, a conciliatory tendency was calculated for the entire group, each dyad, kin and non-kin, and intraand intersexual dyads using a version of the original concil iatory tendency formula that co rrects for increased observation duration and for baseline levels of affiliativ e behavior, both of which may result in an inflation of attracted (and di spersed) pairs relative to neut ral pairs (CCT, Veenema et al., 1994). This allowed comparisons of the tende ncy to reconcile betw een various subgroups (kin vs. non-kin, males vs. females, etc). Corrected Conciliatory Tendenc y (CCT) = 100 x (# attracted pairs# dispersed pairs)/ (total number of all pairs) There was a total of 29 attracte d pairs, 6 dispersed pairs, a nd 16 neutral pairs. Using the above formula, the CCT for the entire group was 45.10%.
Mandrill Reconciliation 27 Sex effects. Previous studies have noted differences in conciliatory tendencies between the sexes (Watts, 1995). These differences have often been linked to differences in relationship value (Aureli et al., 2002). However, a Mann-Wh itney U test indicated that there were no differences in CCTs between sexes (male-male= 50%, female-female = 50%). However, when limiting the analysis to dyads that had fought at leas t twice, intrasexual dyads (46.60 5.60) tended to have higher CCTs than intersexual dyads (28.76 8.02), U = 12.5, N =15, p = .071. In addition, a chi-square analysis revealed that conflicts were more frequent in intrasexual dyads than in intersexual, X 2 ( N = 51, df = 1) = 8.82, p < .010. (Note that the chi-square analysis should be interpre ted with caution because it appears to violate the independence assumption; each individual can potentially contribute to each cell). Kinship effects. It is often reported that related indi viduals are more likely to reconcile than unrelated individuals (Aure li et al., 2002). This variance is believed to be the result of kin relationships being more valu able than relationships with non-kin. However, kin are also considered to have more secure relations hips than non-kin (Cords & Aureli, 2000). Consequently, if relationship security and not value mediates conciliatory tendencies, one would expect non-kin to reconc ile more often than kin. There were no differences in CCTs betw een kin (31.75 18.75) and non-kin (47.39 8.73), t (13) = .845, n.s. However, a chi-square test revealed that non-kin engaged in more conflicts than kin, X 2 ( N = 51, df = 1) = 12.25, p < .05. According to a between subjects ttest, there were no differences in the proportion of time spent in proximity between kin (.10 .03) and non-kin (.06 .02) when all dyads were included in the analysis, t (30) = -1.214, n.s. However, when the data from dyads containing the alpha male and any of the females were excluded, a between subjects t-test showed that kin tended to spend a greater proportion of
Mandrill Reconciliation 28 their time in peaceful proximity ( .10 .03) than did non-kin (.03 .01), t (26) = -2.082, p = .055. Finally, when all dyads were included, a twotailed between subjects t-test failed to find a difference between the rate (per 10 mins) of silent bared-teeth exchange in non-kin (.64 .21) and kin (.26 .06), t (30) = 1.713, n.s. However, if the analysis is limited only to dyads that do not include the alpha male, non-ki n tended to exchange the silent bared-teeth face at a higher rate (.59 .23) than did kin (.12 .05), t (22) = 2.003, p = .064. Effects of percentage of time in peaceful proximity. According to the Valuable Relationship hypothesis, animals r econcile in order to repair da mage to relationships that provide fitness benefits to the individuals. Although re lationship value has been characterized in many ways, some consider rates of affiliative behavior to be indicative of the quality, compatibility, or value of a relationshi p. Thus, the Valuable Relationship hypothesis would predict that dyads with high rates of affiliative behavior would reconcile more often than those with low rates of affiliative beha vior. Hence, an analysis was conducted to determine whether there was a relationship betw een the percentage of time individuals spend in peaceful proximity with one another and their propensity to reconcile. CCTs did not appear to differ as a function of the per centage of time spent in peaceful proximity, U= 35.00, N = 18, p = .617. The dyads were parsed into tw o groups using a median split of the percent proximity values to demarcate the groups. The gr oup that spent more time in peaceful proximity had a CCT of 46.85 23.09, whereas dyads that spent less time in peaceful proximity had a CCT of 43.67 13.19. A between subjects t-test and Figure 5 show that dyads that spent less time in proximity had a higher rate of conflict per 10 minutes (.035 .010) than those who spent more time in peaceful proximity ( .010 .004), t (19.82) = 2.38, p = .027.
Mandrill Reconciliation 29 Effects of baseline silent bared -teeth face exchange rate. It is plausible that relationship security and not valu e per se is responsible for th e pattern of r econciliation in some cases. One way to operationalize the security of a relationship is to examine the rate of appeasement signals or signals of benign intentions between individua ls. Dyads with high rates of appeasement or benign signal exchange are considered less secure than those with low exchange rates. This follows from the no tion that an encounter be tween individuals with insecure relationships, such as an approach, produces uncertainty in the animal approached about the intentions of the approacher. Thus, individuals within these insecure dyads would have a greater need to signal their peaceful inte ntions prior to approaching one another than those with secure rela tionships. There is considerable evidence from this study and others that is consistent with the idea that the silent bared-teeth face of the ma ndrill is a signal of benign intent. Mandrill dyads with high baseline rates of silent bared-teeth face exchange are considered insecure, and those with low baselin e rates are considered secure. Hence, an analysis was conducted to determine whether ther e was a relationship between the security of a relationship and the likelihood of reconciliation. Dyads were parsed into two groups (high and low silent bared-teeth face exchange) using the median baseline silent bared-teeth face rate as a cutoff. When the analysis was lim ited to dyads that engaged in at least two conflicts, a between subjects t-te st showed that dyads with hi gher baseline rates of silent bared-teeth face exchange had a higher CCT (76.33 11.41) than dyads with lower rates of silent bared-teeth face exchange (16.6 10.51), t (7.948) = -3.850, p = .005 (see Table 5 for individual data). Due to the fact that each mandrill CCT was based on a limited number of conflicts, CCTs were also calculated using the total number of conflicts in each group (higher and lower baseline SBTF rates). Table 6 show s that the dyads with lower baseline SBTF
Mandrill Reconciliation 30 rates had a CCT of 16.7, whereas the dyads with higher baseline SBTF rates had a CCT of 55.3. In addition, a between-subjects t-test revealed that the rate of conflict per 10 minutes tended to be higher for dyads who had higher base line silent bared-teeth face exchange rates (.03 .010) than for dyads with lower baseline silent bared-teeth face exchange rates (.01 .004), t (18.995)= -1.86 p = .079 (Figure 6). This finding was also substantiated by a positive correlation between baseline silent bared-teeth face exchange rate and conflict rate, r (32) = .48, p = .005. A Pearson product moment correl ation between the percentage of time spent in peaceful proximity and rate of silent bared-teeth face exchange was not significant, r (32) = .017, n.s. When the data from the dya ds including the alpha male and any of the females were removed, the result was still not significant, r (28)= -.241, n.s. Finally, a MannWhitney U did not reveal a difference in the rate of silent bared-teeth face exchange per 10 minutes for mandrills that spent more time in peaceful proximity (.31 .11) relative to those who spent less time in peacef ul proximity (.59 .20), U = 89, N = 32, p = .14. However, when the data from the dyads that included th e alpha male and the females were excluded, mandrills that spent less time in peaceful proxim ity to one another had a higher rate of silent bared-teeth face exchange (.65 .22) than thos e who spent more time in peaceful proximity to one another (.16 .05), t (26) = 2.12, p = .05. The groups (higher and lower time spent in peaceful proximity) were created using a median split. Effects of peaceful signa l exchange after a conflict Displacement behavior. According to the Uncertainty Reduction hypothesis, anxiety is a mediator of reconciliation. In other words, animals with higher levels of post-conflict anxiety should reconcile more than those with lower levels of post-conflict anxiety. Exchanging peaceful post-conflict signals is thought to function to restore anxiety levels to
Mandrill Reconciliation 31 baseline conditions. One way to measure anxiety in non-human primates is to record rates of displacement behavior, such as scratching, self-grooming, yawning, and body shaking. In order to explore the effects of aggressive conf licts and peaceful post-conflict signals on the mandrills anxiety levels, their rates of displace ment behaviors were recorded and compared over time. More specifically, each individuals rate of displa cement behavior was calculated for each of the ten one-minute blocks for pos t-conflict observations that were followed by peaceful signal exchange and for post-confli ct observations that were not followed by peaceful signal exchange. The post-conflict rate s of displacement behavior were compared to the mean rate of displacement behavior in eac h individuals matched-control sessions. There was no evidence supporting the notion that the displace ment behavior of former opponents was elevated after a c onflict. A Wilcoxon matched pair s signed ranks test did not show a difference between rates of displacement behavior in post-confli ct observations that were not followed by peaceful signals rela tive to their corresponding matched-control observations (Table 7). However, the Wilcoxon matched pairs te st did reveal that individuals who exchanged peaceful signals af ter a conflict had a lower per minute rate of displacement behavior in the first (.36 .12), second (.32 .12), and nint h (.36 .19) minutes after the signal exchange than they did in their corresponding matchedcontrol sessions (see Table 7) Relationship disturbance Aureli et al. (2002) posited th at, in order for animals to reconcile, they must have aggr essive conflicts that disturb the relationship between the two individuals involved. If the aggressive conflicts do not ha ve an effect on the former opponents, there would not be any need to reconc ile. Schaffner et al. (2005) found that red bellied tamarins do not reconcile, but also f ound that former opponents were just as likely to be in proximity to one anot her after a conflict as during corresponding matched-control
Mandrill Reconciliation 32 observations. Because peaceful post-conflict behavior among mandrills is more frequent between animals with insecure relationships, it is plausible that what the mandrills are doing is not reconciling per se. Hence, an analys is was conducted to examine the effects of aggressive conflicts and peacef ul post-conflict signals on the likelihood of proximity. The proportion of observations in which the former opponents were in proximity to one another after a conflict that was not followed by peaceful signal exchange did not differ from that of the matched-control observations, X 2 ( N = 67, df = 1) = 4.42, n.s. In contrast, former opponents were more likely to be in proximity after conflicts followed by peaceful signal exchange than after conflicts with no peaceful exchange, X 2 ( N = 40, df = 1) = 8.21, p <.05. In addition, former opponents were more likely to be in proximity after a conflict followed by peaceful signal exchange than dur ing matched-control observations, X 2 ( N = 71, df = 1) = 4.41, p <.05.
Mandrill Reconciliation 33 Chapter Four Discussion Mandrills exchanged peaceful signals with one another sooner and more frequently after they engaged in an aggr essive conflict than during matched-control observations. In addition, mandrills with higher base line rates of silent bared-t eeth face exchange were more likely to fight than those with lower baseline ra tes of silent bared-teeth face exchange. The mandrills who had higher rates of conflict also had lower rates of affiliative behavior. Finally, mandrills with higher baseline rates of silent bared-teeth face exchange were more likely to reconcile than those with lower baseli ne rates of silent bare d-teeth face exchange. The silent bared-teeth face constituted two-th irds of the first peaceful exchanges between former opponents. The data indicated that former opponents reconciled regardless of whether the PC-MC method, time rule, or rate method was used. According to the time rule, reconciliation was limited to the first two minutes after a conflict. In the majority of gregarious animals that reconcile, peaceful signals are ex changed within three minutes of the preceding conflict (Kappeler & van Schaik, 1992). It is likely that one function of the mandrill silent ba red-teeth face is to signal an individuals benign intentions. Signals of benign intent are similar to appeasement signals, which are directed from subordinates to dom inants in order to reduce the likelihood of receiving aggression. However, unlike appeasemen t signals, signals of benign intent can also be used by dominants to reduce the fear of subor dinates. Data collected during this and other
Mandrill Reconciliation 34 studies are consistent with the notion that the silent bared-teet h face has this function. First, the silent bared-teeth face is not a unidirectional signal; it is performed by both the dominant and subordinate members of a dyad (Setchell & Wickings, 2005). Appeasement signals, in contrast, are usually directed fr om subordinates to dominants. Second, the silent bared-teeth face is more likely to precede peaceful inte ractions than it is to precede aggressive interactions (Bout & Thierry, 2005; Laidre & Yorzinski, 2005) Finally, the mandrills from this study who had higher rates of conflict also exchanged the silent bared-teeth face at higher rates during baseline. In dividuals who have high rates of aggression would have a greater need to signal their pe aceful intentions prior to an approach than those who do not frequently engage in aggressive interactions. The baseline rate of silent bared-teeth face exchange may be a reliable indicator of the security of a relationship. Cords and Aureli (2000) reported that one way to operationalize the security of a relationship between individuals is to compare the likelihood that an approach is preceded by or appears simultaneous ly with an appeasement signal. Although the silent bared teeth face is not an appeas ement signal per se, both appeasement signals and signals of benign intent would be expected to be elevated in insecure dyads. Not only did mandrills with higher rates of silent bared-teeth face exchange have higher rates of conflict, but mandrills with higher rates of conflict also spent less time in non-aggressive proximity than those with lower rates of conflict. A combination of high rates of aggression and little time spent near one another without behaving ag gressively seems like an idea description of an insecure relationship. Furthermore, it is commonly assumed that kin have more secure relationships than nonkin (Cords & Aureli, 2000). Among the mandrill s, kin engaged in fewer conflicts and tended
Mandrill Reconciliation 35 to have lower baseline rates of silent bare d-teeth face exchange than non-kin. Moreover, with the exception of the dyads that included the alpha male and any of the females, kin tended to spend more time in peaceful proximity than did non-kin. Animals with secure relationshi ps might not have as great a need to signal their benign intentions while approaching one another. This is most likely because, within a secure relationship, an approachers intentions are more predictabl e. Animals with insecure relationships benefit from si gnaling their benign intentions while approaching to avoid aggression or supplantation. Subordinates coul d benefit from signaling peaceful intentions by reducing the probability of a ttack or increasing the dominants tolerance of them. Dominants could benefit from directing signals of benign intent to subordinates by promoting group cohesion, preventing group di spersal, reducing the uncerta inty of a subordinate, and facilitating peaceful interaction in some cases (Preuschoft & van Schaik, 2000). The function of the silent bared-teeth face, therefor e, seems analogous to the function of grunts in baboons, who are a close relative of mandrills (Cheney et al., 1995). Grunts are primarily used to signal a dominant females peaceful intentions towards an unrelated subordinate female. This is supported by the fact that during baseline, grun ts are most commonly directed from a dominant female to an unrelat ed subordinate female. Cheney et al. (1995) reported that an approach by a dominant was not as likely to supplant a related subordinate relative to an approach to an unrelated subordinate. Interestingly, the best predictor of mandr ill conciliatory tende ncy was the baseline silent bared-teeth face rate of the dyad. Mandrills that had higher baseline rates of silent bared-teeth face exchange had higher conciliatory tendencies than those who exchanged the silent bared-teeth face at lower rates during baseline. In light of the present evidence which
Mandrill Reconciliation 36 suggests that mandrills with high silent bare d-teeth face exchange rates have insecure relationships, it seems that the mandrills who were most likely to reconcile were those with insecure relationships. Research conducted with other human and non-human animals suggests that relationship security plays an im portant role in the likelihood of reconcilia tion. For instance, Russian children are more likely to reconcile c onflicts with their acquaintances than with their friends (Butovskaya, Verbeek, Ljungberg, & Lunardini, 2000). Acquaintances typically do not know as much about each other and engage in more intense aggressive encounters relative to friends. Therefore, one of the primary differences between acquaintances and friends may very well be relationship security. Furthermore, in human children, friends are often more likely to reconcile than siblings (Dunn, 2004, p. 37). Once again, it is plausible that one of the main differences between friends and siblings is the security of the relationship. In addition, nonkin reconcile more frequently than kin in spotted hyenas (Wahaj et al., 2001). As with primates, hye na kin spend more time with one another, exchange more affiliative behavior, and are much more likely to form alliances than are nonkin. Moreover, as with the mandrills in this study, conflicts between the hyenas were much more frequent between non-kin th an between kin. Therefore, it is likely that the hyenas with insecure relationships also have the highest conciliatory tendencies Since mandrills with insecure relationships were the most likely to reconcile, the pattern of reconciliation in mandrills was in dire ct contrast to predictions derived from the Valuable Relationship Hypothesis. Because in secure dyads had higher rates of aggression, and dyads with higher rates of aggression had lo wer rates of affiliative behavior, it is highly unlikely that these are the most valuable relati onships formed within the group. Furthermore,
Mandrill Reconciliation 37 no correlation was found between affiliative behavior and concil iatory tendency. The results from other species also pose challenges to the Valuable Relationship Hypothesis. For instance, hyenas are characterized by female philopatry, and females are more likely to form alliances than males (Wahaj et al., 2001). Hence, the Va luable Relationship Hypothesis would predict that females should have higher co nciliatory tendencies th an males. However, there was not a difference in conciliatory te ndencies between male and female dyads. Finally, in chimpanzee societies, males are thoug ht to have the most valuable relationships because they are philopatric and cooperate more often than do females. The presence of male philopatry and cooperativ e inclination has been used to explain why male dyads reconcile a greater percentage of their conflicts than do fema les (Cords & Aureli, 2000). However, Preuschoft et al. (2001) reported that agonistic s upport, a valuable form of cooperation, was not a reliable predictor of conciliatory tendency among chimpanzees. It is plausible that relationship security, and not value per se, explains the differences in chimpanzee conciliatory tendencies. For instance, both dominance rank and coalition partners can change very rapidly in ma le chimpanzees, whereas female dominance hierarchies tend to remain relatively stable over time (de Waal, 1982). Furthermore, in contrast to males, female chimpanzees do not sh ow elevated rates of displacement behavior after a conflict with a member of the same sex (Koski, Koops, & Sterck, 2007). Accordingly, males may experience more post-conflict anxiety and have a stronger propensity to reconcile because their relationshi ps are less secure and thus more susceptible to damage from an ensuing conflict. Alternatively, it is possible that insecure relationships are more likely to be reconciled than valuable relationships if the signals used for reconciling are more common between
Mandrill Reconciliation 38 rivals than between friends during baseline. Unlike many other primates, mandrills do not use affiliative signals to reconcile. This behavi or pattern is somewhat consistent with Silks Benign Intent Hypothesis (Silk, 2000). This hyp othesis posits that peaceful post-conflict behavior does not function to repair the damage inflicted upo n valuable relationships, but instead serves to convey that the conflict is over and that the intentions of the signaler are no longer malicious. Silk reported that female ba boons selectively direct post-conflict grunts to other females who have infants. This inclin ation decreases as the infant ages. Silk concluded that such post-conflict signals function to facilitate infant handling and in no way function to repair damage to a relationship. Instead, the signals are indicative of an individuals temporary benign in tentions that sometimes help them obtain a desired resource (e.g., handling an infant). Although there is no evidence that mandrills reconciled to promote friendly interactions, their post-c onflict interactions do seem to consist of short-term signals of peaceful intentions and do not seem to be oriented towards repairing valuable relationships. This is evidenced by the fact that there were not many affiliative behaviors that followed conciliatory signa ls and that post-conflict signals were not directed towards good or valuable relationship partners. In addition, one of the primary means by which hyenas reconcile is engaging in greeting displays. Since gree ting displays seem to reduce uncertainty in tense situations, it is plausibl e that they are analogous to the mandrill silent bared-teeth face (Setchell & Wickings, 2005). It would be interesting to determine whether the distribution of hyena greeting displays during baseline would be similar to the distribution of the mandrill silent bared-teeth face. It is important, however, to note that East et al. (1993) reported that the distribution of hyena greeting displays was similar to the distribution of grooming interact ions in primates. Finally, it is often noted in the human
Mandrill Reconciliation 39 conflict resolution literature that there is a qualitative difference between the way that children reconcile with friends and the way they reconcile with others, such as siblings or non-friends. With friends, children are more likel y to utilize conciliatory strategies, such as negotiation, that allow them to overcome the preceding confli ct and resume their social encounter (Verbeek, Hartup, & Collins, 2000). In contrast, resolution strategies with nonfriends or siblings are typically not geared towards continuing a social interaction. This may be analogous to what is seen in non-human pr imates. Species who selectively reconcile valuable relationships may be more likely to us e physical contact as a conciliatory gesture, which could increase the chances of former opponents grooming one another or engaging in a play bout. In contrast, species that are more likely to reconc ile insecure relationships might be more likely to use peaceful signals that do not necessarily ai m to promote social interaction. For instan ce, the mandrills hardly ever enga ged in friendly in teractions after exchanging peaceful post-conflict signals. It is interesting that mandrills reconcile in light of the fact that aggressive conflicts do not seem to have the same disruptive effects or distribution that is characteristic of the majority of animals that reconcile. There are three pieces of evidence consistent with this view. First, the present study did not find th at the rates of displacement behavior were elevated after conflicts that were not follo wed by any peaceful behavior. Hence, it is possible that aggressive conflicts in mandrills do not produce a significant amount of uncertainty in the former opponents. However, it might not be relevant that the rates of displacement behavior were not elevated after a conflict. For instance, Manson, Perry, and Stahl (2005) reported that wild white-faced capuchins reconcile, but their rates of displacement behavior did not increase after a conf lict. Perhaps anxiety does not manifest in
Mandrill Reconciliation 40 the same ways in all non-human primates, or it may be that anxiety is not always a mediator for reconciliation. It should be noted, however, that there is a fair amount of support for the Uncertainty Reduction Hypothesis, which posits that post-conflict anxiety levels determine the likelihood of reconciliation. Second, Schaffn er et al. (2001) repor ted that red-bellied tamarins do not reconcile because the preceding aggressive conflicts did not seem to alter typical interaction patterns. The present data from mandrills indicate that aggressive conflicts that were not followed by peaceful si gnals did not reduce the probability that the former opponents were in proximity to one anot her. Finally, Kappele r (1993) failed to find evidence that ringtailed lemurs reconcile, bu t noted that conflicts did not occur between animals with high rates of agonis tic support or tolerance. Hen ce, Aureli et al. (2002) have postulated that conflicts must occur between indi viduals with valuable relationships in order for animals to reconcile. Similar to the ring-tailed lemurs, mandrills who spent lots of time together were much less likely to fight than those who spent little time together. It is also noteworthy that ma ndrills exhibit a conciliatory te ndency that is higher than those reported for the majority of species that ha ve been studied. This is in contrast to what would be predicted of the mandrills. For instance, Thierry (2000) reported that among macaque species, those in which unidirectional conflicts are common ha ve low conciliatory tendencies. The majority of conflicts witne ssed in the present study were unidirectional, comprised of one animal pursuing and occasiona lly physically attacking another animal, who would flee (personal observation). Additionally, mandrills are fairly despotic animals. Despotic animals typically have much lower conc iliatory tendencies relative to those living in egalitarian societies (de Waal & Luttrell, 1989).
Mandrill Reconciliation 41 Signals of benign intent do seem to have the same uncertainty reduction effects as affiliative signals when exchanged after a conf lict. Cheney et al. (1995) found that when dominant female baboons emit a grunt towards unr elated subordinate females after a conflict, the subordinates were less likely to be s upplanted in response to an approach by the dominants. Likewise, mandrill former opponents were more likely to be in proximity to one another after the exchange of peaceful post-c onflict signals relative to both conflicts not followed by peaceful signals and to matched-contro l periods. However, it is not quite clear that this indicates greater to lerance after a peaceful exchange It seems that there was a qualitative difference between proximity duri ng baseline and post-conflict encounters. Whereas proximity during baseline was usually characterized by two individuals sitting near one another, post-conflict proximity mainly involved the former opponents standing next to one another after they had exchanged the silent bared-teeth face. It was almost as if they remained next to one another after exchanging silent bared-teeth faces to confirm that the others intentions were peaceful. More spec ifically, when two former opponents were in proximity after a conflict, it ofte n followed a particular pattern. One animal would attack the other and then terminate the attack. The animal that was formerly being attacked would stop, turn and face its pursuer. Many times the two animals would exchange the silent bared-teeth face and take a few steps towards one another, which resulted in the animals being within two meters of each other (and hence in proximity). Often, the aggressive encounter would not lead to a true dispersal that was then followed by an approach. Instead, former opponents were already very close to tw o meters from one another imme diately after the aggressive encounter terminated. Therefore, measuring pr oximity might have been more meaningful if the animals had already been dispersed due to th e initial aggressive encounter and then later
Mandrill Reconciliation 42 approached one another. On a related note, the present study revealed that the rate of displacement behavior of mandrills was lower fo r the first two minutes after the post-conflict exchange of a signal of benign in tent relative to baseline. Th us, it is plausible that these signals do reduce postconflict anxiety due to uncertainty. Limitations and future directions The most notable limitation of this study is the small number of conflicts that were recorded. This was due to the zoo administ rations frequent shifti ng and manipulation of group compositions in order to mitigate the aggressive conflicts that transpired during the study period. Consequently, it was difficult to demonstrate variation in conciliatory tendencies, since not all animal s fought and many who did fight only were observed to do so once. Although mandrill dyads with higher baseli ne rates of silent bared-teeth face exchange tended to reconcile more often th an those with lower silent ba red-teeth face rates, the number of conflicts was low enough to potentially bias conciliatory tendency values. Since dyads that have low baseline rates of silent bared-teeth face exchange also have low rates of conflict, their conciliatory tendencies in particul ar are based on a limited number of conflicts. Thus, it is possible that the conciliatory tende ncy values would change as more data are collected. In addition, it was difficult to determine the effects of exchanging peaceful post-conflict signals due to the relatively small number of conflicts observed and because there was only one observer collecting the data. It was not feasible to videot ape the mandrills. When the camera was zoomed our far enough to capture th e mandrill enclosure, it was too difficult to ascertain the identity of the individuals. C onversely, when the camera was zoomed in close enough to determine individual identities, it was very difficult to follow the fast-moving
Mandrill Reconciliation 43 mandrills with the camera. Moreover, although interobserver agreement could be calculated if there was another person observing the interac tions, it was not feasible to recruit and train an independent observer due to the time constraints on the study (since the adolescents were leaving in June). Nevertheless, the pattern of results is noteworthy and different from the vast majority of research concerning reconciliat ion in non-human animals. Indeed, it is more difficult to demonstrate a trend or significance with a small sample size; the pattern of results from this study is robust enough to overcome the low statistical power of the analyses. Finally, if the CCTs are calculated for each grou p as a whole (instead of individual dyads within the group), which bases the CCT on more conflicts, dyads with hi gher baseline silent bared-teeth face exchange rates still had a mu ch greater conciliatory tendency relative to dyads with lower baseline silent bared-teeth face rates. Therefore, although it may be premature to make strong inferences from th ese data, the results at least warrant further investigation of how and w hy animals reconcile. Another drawback of this study is that so me of the results are inconsistent. For instance, although mandrills with higher rates of silent bared-teeth face exchange had higher conflict rates than those with lower silent ba red-teeth face exchange rates and mandrills with higher conflict rates spent less time in non-a ggressive proximity, there was not a direct relationship between silent bare d-teeth face exchange rates an d time spent in non-aggressive proximity. In addition, kin di d not spend more time in non-aggressive proximity to one another, nor did they exchange the silent ba red-teeth face at a lowe r rate, than did non-kin when all dyads in the group were considered. Ho wever, this pattern of results is in part due to the fact that the alpha male used the silent bared-teeth face for more than just a signal of his benign intentions. The alpha male also fr equently used the silent bared-teeth face to
Mandrill Reconciliation 44 solicit copulation from the females in the group; he did not simply direct the silent baredteeth face towards animals with which he had in secure relationships. It has been reported elsewhere that male mandrills use the silent bared-teeth face during sexual encounters (see Dixson, 1998). Since the silent bared-teeth face involves revealing ones large canines, it is possible that the signal and th e large canines first evolved be cause they facilitated female mate assessment. Also consistent with the idea that the large canines of the adult male mandrills evolved via sexual selection is the fact that mandrills do not seem to use these canines for food-related purposes. Furthermore, the alpha male performed more silent bared-teeth face displays than any other group member. When he directed the display towards other group members, its distribution did not vary with conflict rates or time in non-ag gressive proximity. This is probably due to the fact that mandrills are despotic animals, and ther efore all of the other group members are wary of the alpha male, who is far and away the largest and most powerful individual in the group. Since it is like ly that all of the group members feared the alpha male, he probably directed the silent baredteeth face to all of them in order to put them at ease. Thus, both the dual function of the silent bared-teeth face (courtship and peace signaling) as well as the more non-selective usag e of the gesture from the alpha male make it more difficult to show a direct relationship be tween relationship security and the silent baredteeth face exchange rate. Moreover, the alpha male also spends most of his time in non-aggressive proximity with the adult females, neither of which were his kin. Consequently, the mean percentage of time spent in proximity with non-kin is inflated and misrepresents the behavior of the other
Mandrill Reconciliation 45 mandrills in the group, which is that kin t ypically spend more time in non-aggressive proximity than do non-kin. Future research should examine whether it is possible to predict the distribution of peaceful post-conflict signals based on whether signals of benign intent or affiliative signals are exchanged after a conflict. For instance, ma ndrills exchange signals of benign intent and not affiliative signals after a conflict, and the mandrills that are more likely to reconcile are those that have insecure relationships. Mandr ills that have insecu re relationships are characterized by having higher rates of conflict and benign in tent signal exchange during baseline, and they are also more likely to be unrelated. Similarly, hye nas are more likely to reconcile with non-kin than kin and are t hought to have more valuable and secure relationships with kin than with non-kin. It is likely that the signals they use for reconciliation (greetings) are also signals of benign intent and not necessarily affiliative signals (but see East et al., 1993). Finally, baboons use signals of benign intent signals to reconcile, and it is plausible that they are more likely to direct these signals towards those with whom they have an insecure relations hip. Thus, it is possible that animals who exchange signals of benign intent after a conflict will be more likely to reconcile with insecure relationship partners, and that animals who exchange friendly or affiliative signals after a conflict are more lik ely to reconcile with valuab le relationship partners.
Mandrill Reconciliation 46 List of References Abegg, C., Thierry, B., & Kaumanns, W. (1996) Reconciliation in three groups of liontailed macaques. International Journal of Primatology 17, 803-816. Abernathy, K. A., White, L. J. T., & Wickings, E. J. (2002). Hordes of mandrills ( Mandrillus sphinx) : extreme group size and seasonal male presence. Journal Zool. Lond., 258, 131-137. Arnold, K., & Barton, R. A. (2001). Post -conflict behaviour of leaf monkeys ( Tachypithecus obscurus ). I. Reconciliation. International Journal of Primatology 22, 243-266. Aureli, F. (1992). Post-conflict beha vior among wild long-tailed macaques (Macaca fascicularis ). Behavioral Ecology and Sociobiology 31, 329-337. Aureli, F. (1997). Post-conflict anxiety in non-human primates: the mediating role of emotion in conflict resolution. Aggressive Behavior 23, 315-328. Aureli, F. & van Schaik, C.P. (1991a). Po st-conflict behavior in long tailed macaques ( Macaca fascicularis ): I. The social events. Ethology, 89, 89-100. Aureli, F., & van Schaik, C.P. (1991b). Po st-conflict behavior in long tailed macaques ( Macaca fascicularis ): II Coping with the uncertainty. Ethology, 89, 101-114. Aureli, F., van Schaik, C.P., & van Hooff, J.A.R.A.M. (1989). Functional aspects of reconciliation among captive long-tailed macaques ( Macaca fascicularis ). American Journal of Primatology 19, 39-51. Aureli, F., Cords, M. & van Schaik, C.P. (2002). Conflict resolution following aggression in gregarious animals: a predictive framework. Animal Behaviour, 64, 325-343 Aureli, F., Das, M., & Veenema, H. C. (1997). Differential kinship effects on reconciliation in three species of macaques ( Macaca fascicularis, M. fuscata, and M. sylvanus ). Journal of Comparative Psychology 111, 91-99. Aureli, F., Das, M., Verleur, D., & van Hooff, J. A. R. A. M. (1994). Post-conflict social interactions among Barbary macaques (Macaca sylvanus ). International Journal of Primatology 15, 471-485. Aureli, F., van Schaik, C. P., & van Hooff, J. A. R. A. M. (1989). Functional aspects among captive long-tailed macaques ( Macaca fascicularis ). American Journal of Primatology 19, 39-51. Bertram, B.C.R. (1975). Social factor s influencing reproduction in wild lions. Journal of Zoology, 177, 463-482. Bjornsdotter, M., Larsson, L., & Ljungberg, T. (2000). Post-conflic t affiliation in two captive groups of black-and-white guereza Colobus guereza Ethology, 196, 289300. Bout, N., & Thierry, B. (2005). Peaceful mean ing for the silent bare d-teeth displays of mandrills. International Journal of Primatology 26 (6), 1215-1228. Butovskaya, M., Verbeek, P., Ljungberg, T., & Lunardini, A. (2000). A multicultural
Mandrill Reconciliation 47 view of peacemaking among young children. In F. Aureli & F. B. M. de Waal (Eds.), Natural Conflict Resolution (pp. 243-258). Berkeley, CA: University of California Press. Call, J. (2002). Distance regulation in macaque s: a form of implicit reconciliation? In F. Aureli & F. B. M. de Waal (Eds.), Natural Conflict Resolution (pp. 191-193). Berkeley, CA: University of California Press. Call, J. (1999). The effect of inter-opponent distance on the occurren ce of reconciliation in stumptail (Macaca arctoides) and rhesus macaques (Macaca mulatta ). Primates 40, 515-523. Call, J., Aureli, F., & de Waal, F. B. M. (2002). Postconflict thir d-party affiliation in stumptailed macaques. Animal Behaviour, 63, 209-216. Caro, T.M. (1986). The functions of stotting: a review of the hypotheses. Animal Behaviour, 34, 649-662. Castles, D. L., & Whiten, A. (1998a). Post-c onflict behavior of wild olive baboons. I. Reconciliation, redirec tion, and consolation. Ethology 104, 126-147. Castles, D. L., & Whiten, A. (1998b). Post-conflict behavior of wild olive baboons. II. Stress and self dire cted behavior. Ethology, 104, 148-160. Charpentier, M., Peignot, P., Hossaert-Mckey, M., & Wickings, E. J. 2004. Changes in social interaction during adol escence in male mandrills ( Mandrillus sphinx ). American Journal of Primatology 73, 63-73. Cheney, D. L., & Seyfarth, R. M. (1989). Redirected aggression and reconciliation among vervet monkeys, Cercopithecus aethiops. Behaviour, 110, 258-275. Cheney, D. L., Seyfarth, R. M., & Silk, J. B. (1995). The role of grunts in reconciling opponents and facilitating interactions among adult female baboons. Animal Behaviour, 50, 249-257. Cords, M. (1988). Resolution of aggressi ve conflicts by immature long-tailed macaques Macaca fascicularis Animal Behaviour, 36, 1124-1135. Cords, M. (1992). Post-conflict reunions a nd reconciliation in l ong-tailed macaques. Animal Behaviour, 44, 57-61. Cords, M. (1993). On operati onally defining reconciliation. American Journal of Primatology 29, 255-267. Cords, M., & Aureli, F. (1996). Reasons for reconciling. Evolutionary Anthropology 5, 42-45. Cords, M., & Aureli, F. (2000). Reconciliation and relationship qualities. In F. Aureli & F. B. M. de Waal (Eds.), Natural Conflict Resolution (pp. 177-198). Berkeley, CA: University of California Press. Cords, M., & Thurnheer, S. (1993). Reconciliation with valuable partners by long-tailed macaques. Ethology, 93, 315-325 Creel, S., & Creel, N.M. (1995). Communal hun ting and pack size in African wild dogs, Lycaon pictus. Animal Behaviour 50, 1325-1339. de Waal, F. B. M. (1979). Reconci liation and consolation among chimpanzees. Behavioral Ecology and Sociobiology 5, 55-66. de Waal, F.B.M. (1982). Chimpanzee politics: Power and sex among apes New York: Harper & Row.
Mandrill Reconciliation 48 de Waal, F.B.M. (1986). The integration of dominance and social bonding in primates. Quarterly Review of Biology 61, 459-479. de Waal, F. B. M. (1987a). Dyna mics of social relationships. In B.B. Smuts, D.L.Cheney, R. M. Seyfarth, R. W. Wrangham, & T. T. Struhsaker (Eds.), Primate Societies (pp. 421-429). Chicago: Univer sity of Chicago Press. de Waal, F. B. M. (1987b). Tension regula tion and non-reproductive functions of sex in captive bonobos (Pan paniscus ). National Geographic Research 3, 318-335. de Waal, F. B. M. (1996). Conflict as negotia tion. In W. C. McGrew, L. F. Marchant, & T. Nishida (Eds.), Great Ape Societies (pp. 159-172). Cambridge: Cambridge University Press. de Waal, F.B.M. (2000). The first kiss: f oundations of conflict reso lution research in animals. In F. Aureli & F. B. M. de Waal (Eds.), Natural Conflict Resolution (pp. 15-33). Berkeley, CA: Universi ty of California Press. de Waal, F. B. M., & Luttrell, L. (1985). The formal hierarchy of rhesus monkeys: An investigation of the bared-teeth display. American Journal of Primatology 9, 7385. de Waal, F.B.M., & Luttrell, L. (1989). Toward a comparative socioecology of the genus Macaca : different dominance styles in rhesus and stumptail macaques. American Journal of Primatology 19, 83-109. de Waal, F. B. M., & Ren, R. (1988). Comparison of the reconciliation behavior of stumptail and rhesus macaques. Ethology 78, 129-142. de Waal, F. B. M., & Yoshihara, D. (1983). Reconciliation and redirected affection in rhesus monkeys. Behaviour, 85, 224-241. Demaria, C., & Thierry, B. (2001). A compar ative study of reconci liation in rhesus and Tonkean macaques. Behaviour, 138, 397-410. Disotell, T. R. (1994). Generic le vel relationships of the Papionini ( Cercopithecoidea ). American Journal of Physiological Anthropology 94, 47-57. Dixson, A.F. (1998). Primate sexuality: Comparativ e studies of the prosimians, monkeys, apes, and human beings Oxford, UK: Oxford University Press. Dunn, J. (2004). Childrens friendships: th e beginnings of intimacy Malden, MA: Blackwell Publishing. East. M.L., Hofer, H., & Wickler, W. (1993). The erect penis is a flag of submission in a female-dominated society: greetings in Serengeti spotted hyenas. Behav. Ecol. Sociobiol., 33, 355-370. Estes, R. D. (1991). The behavior guide to African mammals. Berkeley, California: University of California Press. Fels, D., Rhisiart, A.A., & Vollrath, F. (1995). The selfish crouton. Behaviour, 132, 4955. Fuentes, A., Malone, N., Sans, C., Mathes on, M., & Vaughan, L. (2002). Conflict and post-conflict behavior in a small group of chimpanzees. Primates 43 (2), 223-235. Grubb, P. (1973). Distribution, divergence and speciation of the drill and mandrill. Folia Primatologia 20, 161-177. Gust, D. A., & Gordon, T. P. (1993). Conflict resolution in sooty mangebeys. Animal Behaviour, 46, 685-694.
Mandrill Reconciliation 49 Henry, J.P. (1982). The relation of social to biological proce sses in disease. Social Science and Medicine, 16, 369-380. Hinde, R.A. (1985). Was The Expression of the Emotions a misleading phrase? Animal Behaviour, 33, 985-992. Hofer, H., & East, M. L. (2000). Conflic t management in female-dominated spotted hyenas. In F. Aureli & F. B. M. de Waal (Eds.), Natural Conflict Resolution (pp. 232-234). Berkeley, CA: University of California Press. von Holst, D. (1985). Coping behavior a nd stress physiology in male tree shrews ( Tupaia belangeri ). Fortschrite der Zoologie 31, 461-469. Johnson, R. N. (1972). Aggression in Man and Animals Philadelphia: Saunders. Judge, P. G. (1991). Dyadic and triadic reconciliation in pigtail macaques ( Macaca nemestrina ). American Journal of Primatology, 23, 225-237. Kaplan, J.R. (1986). Psychol ogical stress and behavior in non-human primates. In G Mitchell & J Erwin (Eds.), Comparative Primate Biology, Vol 2A: Behavior, Conservation, and Ecology (pp.455-492). New York: Alan R. Liss. Kappeler, P. M. (1993). Reconciliation and post-conflict behavior in ringtailed lemurs, Lemur catta and redfronted lemurs, Eulemur fulvus rufus. Animal Behaviour 45, 901-915. Koski, S.J., Koops, K., & Sterck, E.H.M. (2007). Reconciliation, relationship quality, and postconflict anxiety: Testing the inte grated hypothesis in captive chimpanzees. American Journal of Primatology, 69, 158-172. Koyama, N. F. (2001). The long-term eff ects of reconciliation in Japanese macaques ( Macaca fuscata ). Ethology,107, 975-987. Krause, J., & Ruxton, G.D. (2002). Living in groups. Oxford: Oxford University Press. Laidre, M. E., & Yorzinski, J. L. (2005). The silent bared-teeth face and the crest-raise of the mandrill ( Mandrillus sphinx ): a contextual analysis of signal function. Ethology, 111, 143-157. Lorenz, K. (1966). On Aggression New York: Harcourt, Brace, and World, Inc. Maestripieri, D., Schino, G., Aureli, F., & Troisi, A. (1992). A modest proposal: displacement activities as indicato rs of emotions in primates. Animal Behaviour, 44(5), 967-979. Manson, J.H., Perry, S., & Stahl, D. ( 2005). Reconciliation in wild white-faced capuchins ( Cebus capucinus ). American Journal of Primatology 65, 205-219. Matsumara, S. (1996). Post-conflict cont acts between former opponents in wild moor macaques (Macaca maurus ). American Journal of Primatology 38, 211-219. Mellen, J. D., Littlewood, A. P., Barrow, B. C., & Stevens, V. J. (1981). Individual and social behavior in a captive troop of mandrills ( Mandrillus sphinx ). Primates, 22 (2), 206-220. Menzel, C.R. (1993). Coordination and conflict in Callicebus social groups. In W.A. Mason & S.P. Mendoza (Eds.), Primate Social Conflict (pp 253-290). Albany, New York: State University of New York Press. Nishida, T., & Hiraiwa-Hasegawa, M. (1987). Chimpanzees and bonobos: Cooperative relationships among males. In B.B. Smuts, D.L.Cheney, R. M. Seyfarth, R. W. Wrangham, & T. T. Struhsaker (Eds.), Primate Societies (pp. 165-178). Chicago: University of Chicago Press.
Mandrill Reconciliation 50 Palagi, E., Paoli, T., & Tarli, S. B. (2004) Reconciliation and consolation in captive bonobos ( Pan paniscus ). American Journal of Primatology, 62, 15-30. Palagi, E., Paoli, T., & Tarli, S. B. (2005) Aggression and reconc iliation in two captive groups of Lemur catta International Journal of Primatology 26 (2), 279-294. Part, T., & Qvarnstrom, A. (1997). Badge si ze in collared flycatchers predicts outcome of male competiti on over territories. Animal Behaviour, 54, 893-899. Petit, O., & Thierry, B. (1994a). Reconcilia tion in a group of Guinea baboons. In J. J. Roeder, B. Thierry, J. R. Anderson, & N. Herrenschmidt (Eds.), Current Primatology, Vol II: Social Development, Learning, and Behaviour (pp. 137-145). Strasbourg: Presses de l'Un iversite Louis Pasteur. Petit, O., & Thierry, B. (1994b). Reconc iliation in a group of black macaques. Dodo, Journal of the Wildlife Preservation Trust 30, 89-95. Pfeffer, P. (1967). Le mouflon de Corse ( Ovis ammon musimom ); position systematique, ecologie et ethologue comparees. Mammalia (Supplement) 31, 1-262. Preuschoft, S., & van Schaik, C.P. (2000) Dominance and co mmunication: conflict management in various social settings. In F. Aureli & F. B. M. de Waal (Eds.), Natural Conflict Resolution (pp. 77-105). Berkeley, CA: Un iversity of California Press. Preuschoft, S., Wang, X., Aureli, F., & de W aal, F. B. M. (2001). Reconciliation in captive chimpanzees: a re-evaluation with controlled methods. International Journal of Primatology 23, 29-50. Rasa, O. A. E. (1977). The etholog y and sociology of the dwarf mongoose ( Helogale undulatta rufula ). Zeitschrift fur Tierpsychologie 43, 337-406. Ren, R., Yak, K., Su, Y., Qi, H., Liang, B., Ba o, W., & de Waal, F. B. M. (1991). The reconciliation behavior of golden monkeys ( Rhinopithecus roxe llanae roxellanae ) in small breeding groups. Primates 32, 321-327. Rogers, M. E., Abernathy, K. A., Fontaine, B ., Wickings, E. J., White, L. J. T., & Tutin, C. E. G. (1996). Ten days in the life of a mandrill horde in the Lope Reserve, Gabon. American Journal of Primatology 40, 297-313. Rohwer, S. (1982). The evolution of reliab le and unreliable badges of fighting ability. American Zoologist, 22, 531-546. Rowell, T. E., & Rowell, C. A. (1993). The social organization of feral Ovis aries ram groups in the pre-rut period. Ethology 95, 213-232. Sapolsky, R.M. (2005). The influence of social hierarchy on primate health. Science, 308, 648-652. Schaffner, C. M., Aureli, F., & Caine, N. G. (2001). Following the rules: why tamarins don't reconcile. American Journal of Primatology, 54 (supplement 1), 49-50. Schaffner, C. M., Aureli, F., & Caine, N. G. (2005). Following the rules: why tamarins don't reconcile. Folia Primatol 76, 67-76. Schaffner, C. M., & Caine, N. G. (2000). The peacefulness of cooperatively breeding primates. In F. Aureli & F. B. M. de Waal (Eds.), Natural Conflict Resolution (pp. 155-169). Berkeley, CA: University of California Press. Schaller, G. B. (1972). The Serengeti Lion Chicago: University of Chicago Press. Schino, G. (1998). Reconciliation in domestic goats. Behaviour, 135, 343-356. Schino, G., Schucchi, S., Maestripieri, D., & Tu rillazzi, P. G. (1988). Allogrooming as a
Mandrill Reconciliation 51 tension reduction mechanism: A behavioral approach. American Journal of Primatology, 16, 43-50. Setchell, J.M. (1999). Socio-sexual development in the male mandrill ( Mandrillus sphinx ). Unpublished doctoral dissertation, University of Cambridge, Cambridge, UK. Setchell, J. M., & Dixson, A. F. (2001). Arrested development of secondary sexual adornments in subordinate adult male mandrills ( Mandrillus sphinx ). American Journal of Physiological Anthropology 115, 245-252. Setchell, J. M., & Dixson, A. F. (2002). Developmental variables and dominance rank in male mandrills ( Mandrillus sphinx ). American Journal of Primatology 56, 9-25. Setchell, J. M., & Wickings, E.J. (2005). Dominance, status signa ls and coloration in male mandrills ( Mandrillus sphinx ). Ethology, 111, 25-50. Silk, J. (1987). Social behavior in evolutionary perspectiv e. In B. B. Smuts, D. L. Cheney, R. M. Seyfarth, R. W. Wra ngham, & T. T. Struhsaker (Eds.), Primate Societies (pp.318-329). Chicago: Univers ity of Chicago Press. Silk, J. B. (1996). Why do primates reconcile. Evolutionary Anthropology, 5, 39-42. Silk, J.B. (2000). The function of peaceful post-conflict interactions: an alternate view. In F. Aureli & F. B. M. de Waal (Eds.), Natural Conflict Resolution (pp. 179-181). Berkeley, CA: University of California Press. Silk, J. B., Cheney, D. L., & Seyfarth, R. M. (1996). The form and function of postconflict interactions be tween female baboons. Animal Behaviour, 52, 259-268. Smuts, B.B. (1985). Sex and friendship in baboons Hawthorne: Aldi ne Publishing Co. Stammbach, E. (1987). Desert, forest, and montane baboons: multilevel societies. In B. B. Smuts, D. L. Cheney, R. M. Seyfart h, R. W. Wrangham, & T. T. Struhsaker (Eds.), Primate Societies (pp.112-120). Chicago: Univ ersity of Chicago Press. Swedell, L. (1997). Patterns of r econciliation among wild gelada baboons ( Theropithecus gelada) : a brief report. Primates, 38, 325-330. Templeton, J. J., & Giraldeau, L. A. (1995). Patch assessment in foraging flocks of European starlingsevidence for the use of public information. Behavioral Ecology, (6), 65-72. Thompson, C.W., & Moore, M.C. (1991). Throat colour reliably sign als status in male tree lizards, Urosaurus ornatus Animal Behaviour, 52, 745-753. Uno, H., Tarara, R., Else, J.G., Suleman, M. A., & Sapolsky, R.M. (1989). Hippocampal damage associated with prolonged and fatal stress in primates. Journal of Neuroscience, 9, 1705-1711. van Schaik, C. P. (1983). Why are diurnal primates living in groups? Behaviour, 87 (12), 120-144. van Schaik, C. P., & van Noordwijk, M. A. (1986). The hidden costs of sociality: Intragroup variation in feeding strategies in Sumatran long-tailed macaques ( Macaca fascicularis ). Behavior, 99, 296-315. van Schaik, C. P., & Aureli, F. (2000). The natural history of valu able relationships in primates. In F. Aureli & F. B. M. de Waal (Eds.), Natural Conflict Resolution (pp. 307-333). Berkeley, CA: University of California Press. Vehrencamp, S. (1983). A model for the ev olution of despotic versus egalitarian societies. Animal Behaviour 31, 667-682.
Mandrill Reconciliation 52 Verbeek, P., & de Waal, F. B. M. (1997). Post-conflict behavior in a group of brown capuchins in the presence a nd absence of attractive food. International Journal of Primatology 18, 703-725. Verbeek, P., Hartup, W.W., & Collins, W.A. (2000). Conflict management in children and adolescents. In F. Aureli & F. B. M. de Waal (Eds.), Natural Conflict Resolution (pp. 34-53). Berkeley, CA: University of California Press. van Vuren, D. (1996). Ectoparasites, fitness, and social behaviour in yellow-bellied marmots. Ethology, 102, 686-694. Wahaj, S. A., Guse, K. R., & Holekamp, K. E. (2001). Reconciliation in spotted hyena ( Crocutta crocutta ). Ethology 107, 1057-1074. Watts, D. P. (1995). Post-conflict social events in wild mountain gorillas ( Mammalia, Hominoidea). I. Social interactions between opponents. Ethology 100, 139-157. Weaver, A. (2003). Conflict and reconciliation in captive bo ttlenosed dolphins, Tursiops truncatus. Marine Mammal Science, 19 (4), 836-846. Weaver, A., & de Waal, F. B. M. (2000). The mother-offspring relationship as a template in social development: reconciliation in captive brown capuchins ( Cebus apella ). Journal of Comparative Psychology, 117 (1), 101-110. Wickings, E. J., & Dixson, A. F. (1992) Testicular function, secondary sexual development, and social status in male mandrills ( Mandrillus sphinx ). Physiology & Behavior 52, 909-916. Wiley, R. H. (1991). Lekking in birds a nd mammals: behavioral and evolutionary issues. Advances in the Study of Behavior 20, 201-291 York, A. D., & Rowell, T. E. (1988). Reconciliation following aggression in patas monkeys, Erythrocebus patas. Animal Behaviour, 36, 502-509.
Mandrill Reconciliation 53 Table 1: Typical Approaches to Analyzing Reconciliation Data, Operationalization of Conflicts and Operationalization of Affilia tion Animal Method Conflict Aff iliation Reconciled? Author Lemur catta PC-MC staring, spat calls, lunge, greeting, grooming, huddle-with L. catta : No Kappeler (1993) Eulemur fulvus rufus Time rule chase, grab, bite E. fulvus : PC-MC: No Time rule: Yes Lemur catta PC-MC staring, spat calls, l unge, body-body and olfactory contact, Yes Palagi et al. (2005) chase, grab, bite grooming, greeting Macaca arctoides PC-MC facial and vocal threats only non-agonistic body contact Yes de Waal & Ren (1988) accompanied by lunges or pursuits > 2m Macaca fascicularis PC-MC threats and submissive behavior proximity (50 cm or less), Yes Cords (1993) Time rule body contact, exchange of affiliative signals Pan troglodytes PC-MC 3 or more non-contact agonistic no def inition provided Yes Fuentes et al. (2002) events, agonistic contact Pan troglodytes PC-MC tug, brusque rush, trample, kiss, embrace, grooming, Yes Preuschoft et al. (2002) bite, grunt-bark, shrill-bark, gentle touch, finger-in-mouth, flight, crouch, shrink/flinch, sexual behavior, social play, bared-teeth scream contact sitting Papio cynocephalus Rate head bobs, eye threats, vocal approach, grooming, groom Yes Silk et al. (1996) ursinus threats, lunges, chases, bites, present, touching, embracing, attacks grunts
Mandrill Reconciliation 54 Table 2: Demographics of Mandrills in Sample Animal Sex/ age class Age Relative Relationship Coefficient of relatedness Nestor (N) AM 16 Milo Son 0.5 Jasper Son 0.5 Moesha Daughter 0.5 Mukobi Son 0.5 Jerome Son 0.5 Jinx Daughter 0.5 Miller (R) AF 13 Milo Son 0.5 Moesha Daughter 0.5 Mukobi Son 0.5 Milo (M) AM 9 Moesha Sister ~.5 Mukobi Brother ~.5 Miller Mother .5 Moesha (O) AdF 7 Milo Brother ~.5 Mukobi Brother ~.5 Miller Mother .5 Mukobi (P) AdM 6 Milo Brother ~.5 Moesha Sister ~.5 Miller Mother .5 Jalisa (A) AF 13 Jasper Son .5 Jerome Son .5 Jinx Daughter .5 Jasper (E) AdM 8 Jinx Sister ~.5 Jalisa Mother .5 Jerome Brother ~.5 Jerome (J) AdM 6 Jasper Brother ~.5 Jinx Sister ~.5 Jalisa Mother .5 Jinx (X) AdF 3 Jasper Brother ~.5 Jerome Brother ~.5 Jalisa Mother .5
Mandrill Reconciliation 55 Table 3: Mandrill Ethogram (Based on Setchell, 1999) Affiliative/peaceful behaviors Silent bared-teeth face: mouth retracted horizontally and vertically at the corners so the canines are partly exposed. Mouth remains closed in the center and the incisors are co vered by the lips and only partly visible, resulting in a figure-eight shape. Lip-smacking: lips quickly opened and closed together audibly. Head-shake: head is rotated from one side to the other, sometimes repeated back to original position. Typically accompanies Silent bared-teeth face Non-aggressive touching: making physical contact with another without any signs of threat or agitation (such as yawning, etc) Playing: engaging in relaxed chasing, biting, wrestli ng that is almost always accompanied by a relaxed, open-mouthed play face (teeth are usually covered). Allogrooming: picking through the fur of another individual with either hands or mouth Peaceful proximity: individuals are within ~ two mete rs of one another without exc hanging any agonistic signals. Further, both individuals must not perfor m any of the following anxiety indicating behaviors: yawn, scratch, body shake Submissive behaviors Scream: sharp, occasionally repeated vocalization with open-mouthed bared teeth expression. Present: directing rear end towards another indi vidual, usually while looking back at the animal Crouch: making itself smaller and closer to the ground, usually in response to another's approach or threat Look away: abruptly directing gaz e away from another
Mandrill Reconciliation 56 Table 3: (Continued) Avoid: leaving another when it approaches or moves out of proximity to another animal by walking at a normal pace Flee: avoiding another by running away from it Level 1 aggression Head Jerk: staring at another and emphatically nodding the head down and forward with mouth closed Threat Grunt: a short bark directed at another group member Ground Slap: Striking one or both hands on the ground quickly and with force. Level 2 aggression Lunge: moving towards an animal rapidly for a distance of less than 2m in an aggressive context (e.g. not accompanied by a play face and usually preceded or followed by level 1 or level 3 aggression) Chase: moving rapidly towards another animal for a distance of greater than 2m in an aggressive context Level 3 aggression Hit: striking another using its extremities Grab: clasping or attempting to clasp another animal quickly in an aggressive context Bite: placing its mouth on the body of another and clamping down in an aggressive context
Mandrill Reconciliation 57 Table 3: (Continued) Distance Proximity: within 2m of the focal individual. 0-1m = near, 1-2m = far Anxiety related behaviors Yawning : an animal opens its mouth and exposes all of its canines Half yawn: a yawn where the teeth remain covered by the lips Body shake: trembling in a vigorous shaking motion Scratch: vigorous, repeated raking of the skin/hair with fingers Auto-grooming: manual or oral manipulation of own fur
Mandrill Reconciliation 58 Table 4: Conflict information Dyad Date Time Aggression level Aggressor A, D, or N? Jerome-Miller 1/24/06 2:46p 1 + 2 Miller NEUTRAL Jerome-Jasper 1/25/06 11:53a 1 + 2 Jerome ATTRACTED Nestor-Milo 2/6/06 4: 56p 2 Nestor NEUTRAL Moesha-Jinx 2/7/06 4:50p 1 + 2 Moesha ATTRACTED Jinx-Moesha 2/8/06 1:55p 1 + 2 Moesha NEUTRAL Mukobi-Jasper 2/8/06 2:14p 2 + 3 Mukobi ATTRACTED Milo-Nestor 2/10/06 12:02p 2 Nestor NEUTRAL Nestor-Milo 2/10/06 1: 57p 2 Nestor NEUTRAL Jinx-Moesha 2/10/06 2: 16p 2 Moesha ATTRACTED Mukobi-Jasper 2/14/06 2:23p 2 + 3 Mukobi NEUTRAL Jalisa-Moesha 2/22/06 4:30p 2 Moesha NEUTRAL Jinx-Mukobi 2/22/06 4:41p 1 + 2 Mukobi DISPERSED Jasper-Mukobi 2/23/06 11:08a 3 Mukobi ATTRACTED Mukobi-Jasper 2/24/06 10:37a 2 Jasper DISPERSED Jasper-Mukobi 2/24/06 11:20a 3 Jasper ATTRACTED Jasper-Mukobi 2/24/06 11:15a 3 Jasper ATTRACTED Mukobi-Jasper 2/24/06 11:33a 3 Jasper ATTRACTED Jasper-Mukobi 2/24/06 10:29a 2 + 3 Jasper DISPERSED Jasper-Mukobi 2/27/06 12:48p 3 Mukobi ATTRACTED Miller-Jasper 2/27/06 4:02p 2 + 3 Miller ATTRACTED Milo-Jasper 3/6/06 2:12p 3 Milo ATTRACTED Jasper-Jerome 3/8/06 4:52p 3 Jasper ATTRACTED Mukobi-Jasper 3/10/06 4:30p 2 + 3 Jasper ATTRACTED Mukobi-Jasper 3/14/06 4:52p 2 Jasper ATTRACTED Jasper-Mukobi 3/14/06 3:58p 1 + 2 Jasper DISPERSED Moesha-Jalisa 3/21/06 2:00p 2 + 3 Moesha ATTRACTED Jasper-Nestor 3/27/06 3:37p 2 Nestor ATTRACTED Miller-Jerome 3/29/06 4: 57p 1 + 2 Miller NEUTRAL Jerome-Milo 4/3/06 2:30p 2 Milo ATTRACTED Jasper-Miller 4/10/06 4: 46p 2 Miller ATTRACTED Jasper-Miller 4/10/06 5:10p 1 + 2 Miller ATTRACTED Jerome-Jinx 4/21/06 11:25a 3 Jinx ATTRACTED Miller-Jasper 4/ 21/06 11:44a 1 + 2 Miller DISPERSED Jasper-Miller 5/1/06 4:10p 2 Miller ATTRACTED Jinx-Moesha 5/3/06 11:27a 1 + 2 Moesha ATTRACTED Jerome-Jasper 5/11/06 3:43p 1 + 3 Jerome ATTRACTED Jinx-Jasper 5/17/06 11:41a 2 Jasper ATTRACTED Jerome-Jalisa 5/19/06 3:00p 2 Jalisa NEUTRAL Moesha-Jalisa 5/21/06 4: 39p 2 Moesha ATTRACTED Jinx-Jalisa 5/21/06 3:50p 3 Jalisa DISPERSED Jerome-Nestor 5/23/06 4: 32p 2 Nestor NEUTRAL Nestor-Miller 5/23/06 2: 54p 2 Nestor NEUTRAL
Mandrill Reconciliation 59 Table 4: (Continued) Dyad Date Time Aggression level Aggressor A, D, or N? Miller-Jerome 5/23/06 4: 14p 2 Miller ATTRACTED Miller-Nestor 5/25/06 1: 16p 2 Nestor NEUTRAL Nestor-Jasper 5/ 25/06 2:30p 1 + 2 Nestor ATTRACTED Jasper-Moesha 6/8/06 11: 10a 1 + 2 Moesha NEUTRAL Jerome-Moesha 6/10/06 12:20p 1 + 2 Moesha ATTRACTED Moesha-Jalisa 6/16/06 11:08a 2 Moesha NEUTRAL Moesha-Jinx 6/24/06 11:22a 2 Moesha ATTRACTED Jerome-Jalisa 6/26/06 12:04p 2 Jalisa NEUTRAL Jerome-Moesha 6/26/06 12: 04p 1 + 2 Moesha NEUTRAL
Mandrill Reconciliation 60 Table 5: Baseline Silent Bared-Teeth Face Rate, CCT, and Number of Conflicts by Dyad Dyad SBTF rate CCT Number of conflicts Jinx-Jasper 0 1 1 Jinx-Mukobi 0 -1 1 Jinx-Jalisa 0.013396 -1 1 Jerome-Jinx 0.025124 1 1 Jasper-Moesha 0.140008 0 1 Milo-Jasper 0.264601 1 1 Jerome-Nestor 0.362674 0 1 Jerome-Milo 0.363141 1 1 Jerome-Jalisa 0.027319 0 2 Jerome-Moesha 0.123854 0.5 2 Nestor-Jasper 0.770125 1 2 Nestor-Miller 0.306443 0 2 Miller-Jerome 0.393546 0.3 3 Jerome-Jasper 0.45076 1 3 Milo-Nestor 0.369165 0 3 Moesha-Jalisa 0.448638 0.5 4 Jinx-Moesha 0.608408 0.8 5 Jasper-Miller 0.613595 0.6 5 Mukobi-Jasper 2.904901 0.4 12
Mandrill Reconciliation 61 Table 6: CCTs as a Function of Baseline SBTF Rate with CCTs Calculated for Entire Groups (LOWER SBTF) SBTF RATE CCT CONFLICTS A D N CCT Jinx-Jasper 0 1 1 1 Jinx-Mukobi 0 -1 1 1 Jinx-Jalisa 0.013396 -1 1 1 Jerome-Jinx 0.025124 1 1 1 Jerome-Jalisa 0.027319 0 2 2 JeromeMoesha 0.123854 0.5 2 1 1 JasperMoesha 0.140008 0 1 1 Milo-Jasper 0.264601 1 1 1 Nestor-Miller 0.306443 0 2 0 0 2 TOTAL 4 2 6 0.167 (HIGHER SBTF) SBTF RATE CCT CONFLICTS A D N CCT Jerome-Milo 0.363141 1 1 1 Milo-Nestor 0.369165 0 3 3 Miller-Jerome 0.393546 0.33 3 1 2 Moesha-Jalisa 0.448638 0.5 4 2 2 Jerome-Jasper 0.45076 1 3 3 Jinx-Moesha 0.608408 0.8 5 4 1 Jasper-Miller 0.613595 0.6 5 4 1 Nestor-Jasper 0.770125 1 2 2 Mukobi-Jasper 2.904901 0.416667 12 8 3 1 TOTAL 25 4 9 0.553
Mandrill Reconciliation Table 7: Rates of Displacement Behavior A fter Conflicts Followed by Peaceful Signals (a) and After Conflicts not Followed by P eaceful Signals (b) Between Former Opponents (a) (b) Minute PC Baseline Z Minute PC Baseline Z 1 0.363636 0.557656 -1.969* 1 1.705882 0.628431 1.161 2 0.318182 0.557656 -2.559* 2 1.117647 0.628431 0.876 3 0.681818 0.557656 -.699 3 1.058824 0.628431 0.308 4 0.590909 0.557656 -.732 4 1.411765 0.628431 0.734 5 0.727273 0.557656 -.146 5 1.764706 0.628431 1.113 6 0.5 0.557656 -.765 6 1.764706 0.628431 0.592 7 0.954545 0.557656 -.601 7 1.117647 0.628431 1.492 8 0.5 0.557656 -1.024 8 0.647059 0.628431 0.45 9 0.363636 0.557656 -2.457* 9 0.647059 0.628431 1.018 10 0.545455 0.557656 -1.739 10 0.529412 0.628431 1.208 = p<.05 62
Mandrill Reconciliation Figure 1: Patterns of post-conflict behavior0 5 10 15 20 25 30 35 Attracted DispersedFrequency Figure 2: Peaceful signal exchange over time in post-conflict and matched-control observations0 5 10 15 20 25 12345678910 Minute block Frequency PC MC 63
Mandrill Reconciliation Figure 3: Rate of silent bared-teeth face exchange in post-conflict and matched-control observations0 0.1 0.2 0.3 0.4 0.5 PC rate MC rateRate/10 minutes Figure 4: Percentage of silent baredteeth face exchange with former opponents in PC and MC0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 PC MC% of SBTF 64
Mandrill Reconciliation Figure 5: Conflict rate as a function of % time in peaceful proximity0 0.01 0.02 0.03 0.04 0.05 Low % prox High % proxConflict rate/10 mins Figure 6: Conflict rate as a function of baseline SBTF exchange 0 0.01 0.02 0.03 0.04 0.05 Low SBTF High SBTFConflict Rate per 10 mins 65
Mandrill Reconciliation Figure 7: % of 1st peaceful contacts in post-conflict and matched-control0 10 20 30 40 50 60 70 80 90 100PC MC% of behavior BARED TEETH OTHER PROX 66