Behavioral Ecology Advance Access originally published online on October 3, 2006
Behavioral Ecology 2007 18(1):115-120; doi:10.1093/beheco/arl045
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Grooming and agonistic support: a meta-analysis of primate reciprocal altruism
Istituto di Scienze e Tecnologie della Cognizione del Consiglio Nazionale delle Ricerche, Rome, Italy
Address correspondence to G. Schino. E-mail: gschino{at}casaccia.enea.it.
Received 3 March 2006; revised 22 August 2006; accepted 1 September 2006.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Grooming and agonistic support are 2 common primate behaviors that have been hypothesized to constitute examples of reciprocal altruism. In particular, because primates often direct their grooming up the dominance hierarchy, it has been suggested that they may exchange grooming for agonistic support. Empirical tests of this hypothesis have resulted in highly inconsistent findings. I synthesized the published literature on the relation between grooming and agonistic support in primates using modern meta-analytical techniques. A meta-analysis of 36 studies carried out on 14 different species showed that a significant positive relation exists between grooming and agonistic support (weighted average r = 0.154, corrected for publication bias). These findings suggest that grooming and agonistic support may have evolved as part of a system of low-cost reciprocal altruism. They also highlight the potential of meta-analysis in tackling the study of behavioral phenomena characterized by low overall frequency and small effect sizes.
Key words: agonistic support, grooming, meta-analysis, primates, reciprocal altruism.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Altruistic behaviors, that is, behaviors that benefit the recipient at some cost to the actor, constituted an evolutionary paradox until Hamilton (1964)
and Trivers (1971) provided explanations for their evolution among related and unrelated animals, namely, kin selection theory and reciprocal altruism theory.
Although a role for kin selection in the evolution of altruistic behaviors among related animals is now widely accepted, evidence of reciprocal altruism is still mixed. Relatively few examples of reciprocal altruism have been observed, and these observations were often difficult to replicate (Packer 1977; Wilkinson 1984; Milinski 1987; Bercovitch 1988; Dugatkin 1988). In particular, many of the examples of altruistic behavior that were initially identified were later demonstrated to involve little or no cost for the actor, thus needing to be redefined as mutualistic rather than altruistic (e.g., Clutton-Brock 2002).
Behaviors that benefit the recipient at little or no cost to the actor ("services" hereafter, following de Waal 1997) pose a slightly different problem to the evolutionist. Being advantageous to the actor, they do not constitute an evolutionary paradox. Because they involve net benefits for the recipient, they can be exchanged with similar acts in a sort of low-cost reciprocal altruism or "service economy" (de Waal 1997). The focus of attention of evolutionists has thus shifted from demonstrating the occurrence (and reciprocation) of altruistic behaviors to investigating the rules that underlie partner choice, that is, the deployment of services among different group mates. The biological market theory (Noë and Hammerstein 1995), which emphasized bargaining and outbidding among multiple partners, constituted a natural development of reciprocal altruism theory.
Primates have played a preeminent role in this scenario. Indeed, their cognitive capacities made them good candidates in a search for altruistic behaviors. Primates (especially Catharrine primates) spend a relatively large part of their time carefully inspecting and cleaning the fur of other individuals, a behavior named allogrooming (grooming, hereafter). Also, in several primate species, individuals often support each other during agonistic encounters, meaning that a third individual may intervene into an ongoing fight to help one of the 2 opponents. These 2 behaviors were thus readily identified as good examples of social services.
Both grooming and agonistic support are preferentially directed to related individuals, so that a role of kin selection in their evolution is very likely. Nevertheless, both grooming and agonistic support are also directed to unrelated animals, raising the possibility of a role for reciprocal altruism in their evolution (for reviews, see Chapais 2001; Silk 2002). Also, because both grooming and agonistic support involve little cost for the actor (Dunbar and Sharman 1984; Chapais et al. 1991), they may be part of a service economy in which partner choice plays a preeminent role.
Primates have been shown to reciprocate both grooming and agonistic support, that is, to groom preferentially those individuals that groom them most and to support preferentially those individuals that support them most (de Waal and Luttrell 1989; Silk 1992; Payne et al. 2003; Schino et al. 2003). Reciprocation in kind (grooming for grooming and support for support) may not, however, tell the whole story of primate social complexity.
Prompted by the observation that primates often direct their grooming up the hierarchy (i.e., they groom preferentially higher ranking animals; for a recent meta-analysis, see Schino 2001), Seyfarth (1977) hypothesized that grooming may be exchanged with different benefits and that these benefits may be rank related. An obvious benefit that higher ranking animals are better able to return is agonistic support. Seyfarth thus hypothesized that monkeys compete for social access to high-ranking individuals and that these may return the benefits of received grooming in a different currency, namely, agonistic support. Later, research confirmed the tendency of primates to direct their grooming up the hierarchy (Schino 2001). However, evidence for an interchange of grooming with agonistic support remained elusive (for contrasting results, see, e.g., Seyfarth and Cheney 1984; Silk et al. 2004).
Meta-analysis, the modern statistical techniques used to quantitatively synthesize the results of a set of analyses, seems particularly well suited to address the problem of the relation between grooming and agonistic support. Meta-analysis is now widely used in clinical medicine and is becoming increasingly popular in ecology and evolution (Arnqvist and Wooster 1995; Egger et al. 2001). Its use to further our understanding of social behavior is in contrast still extremely limited. In this paper, I thus present a synthesis of the published data on the relation between grooming and agonistic support in primates using modern meta-analytical techniques in order to test the hypothesis that a positive relationship exists between these 2 variables. Specifically, I tested whether individuals that groomed more were also more likely to support each other during agonistic encounters.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Data collection and categorization
To locate studies, I relied on the PrimateLit database (freely available at http://primatelit.library.wisc.edu/). A search for studies that included the keywords "grooming" or "allogrooming" and "agonistic support" or "coalition" or "alliance" found 359 records. I checked those records and references therein. To be included in the meta-analysis, a study had to test the relation between grooming and agonistic support and to present a measure of effect size (generally an r value) or other information (generally a P value) that could be converted to an r value using standard meta-analytical techniques. Twenty-four studies satisfied the selection criteria. They reported 36 independent tests of the relation between grooming and agonistic support. A list of the studies included in the meta-analysis is presented in Table 1.
|
Most of the published tests included in the meta-analysis (34 out of 36) were correlational, in that they tested whether the within-group distribution of allogrooming was correlated with that of agonistic support. In other words, these studies tested whether those dyads that groomed preferentially also supported each other more frequently. Two studies investigated the temporal contingency between grooming and support, that is, they tested whether, within a given dyad, previous grooming increased the probability of later agonistic support.
As a measure of the quality of the different studies, I relied on the kind of statistical analysis used. Most of the correlational studies were based on some sort of matrix correlation test. Matrix correlations are statistical techniques that have been developed to avoid the problem of nonindependence of the different dyads in a social group (Hemelrijk 1990; de Vries 1993). A few of the correlational studies did not analyze their data using matrix correlations and are thus subject to pseudoreplication given that the different dyads in a groups cannot be considered as statistically independent. Those studies are likely to overestimate effect size. I considered as "good-quality" studies those that are based on matrix correlations and "poor-quality" studies those that (incorrectly) treated dyads as independent units of analysis. Of the 2 studies that investigated the temporal contingency between grooming and support, one was included in the good-quality group because all the dyads tested were formed by different individuals and could thus be correctly considered as independent. The other treated dyads formed by the same individual as independent and was thus included in the poor-quality group.
Seyfarth's model was initially applied to female primates. In fact, there is no a priori reason to assume that an interchange of grooming and agonistic support should be limited to females, as also male–male and male–female relationships may be serviced through the exchange of grooming. However, I also tested whether studies that limited their analysis to female–female interactions had an average effect size that differed from that of the other studies (that included males in their analyses or were limited to males).
A relation between grooming and agonistic support could theoretically arise as a by-product of the preferential association between kin that is typical of primate societies. In this case, grooming and agonistic support would not be maintained by reciprocal altruism but by kin selection. Some of the studies controlled for the influence of kinship, either statistically or by excluding dyads formed by kin. Studies that controlled kinship in fact only controlled for maternal kinship; no study controlled for both maternal and paternal kinship. Studies could report the results of analyses that controlled or did not control for the effect of kinship or both analyses. The "general" meta-analysis that I conducted initially included the largest possible number of studies and thus included both studies that controlled and studies that did not control for the effect of kinship (when both types of data were available, I used the analysis that controlled for kinship). I then compared effect size in studies that did or did not control for kinship. Finally, to evaluate the relation between grooming and agonistic support excluding the influence of kinship, I limited the meta-analysis to those studies that had controlled for kinship.
Data analysis
I conducted random-effect meta-analyses following Egger et al. (2001). As a measure of effect size (i.e., a standardized measure of the strength of a relationship between 2 variables), I used the Pearson's correlation coefficient r. The r values were either taken directly from the published study or obtained by transforming P values or other statistics according to Rosenberg et al. (2000). In a few cases, the authors of the original study presented raw data matrices of grooming and support. In those cases, I carried out Kr tests (Hemelrijk 1990) and then transformed the obtained P values into r values. In some cases, the authors of the original study had conducted multiple tests of the relation between grooming and support (e.g., testing separately aggressor support and victim support). In those cases, I averaged the different r values to obtain a single data point for each study. However, when the authors presented separate analyses for males and females, I considered them as 2 separate data points because they were based on different individuals. The unit of analyses were individual social groups; however, whenever possible, I also repeated analyses on the basis of species values following Gontard-Danek and Møller (1999). I did not apply any phylogenetic correction as no analytical tool has yet been developed to conduct a meta-analysis in a phylogenetic framework. Because, however, I simply calculated overall effects and did not investigate the factors modulating such effects, phylogenetic relations are unlikely to have affected my results.
Data entered into analyses were Fisher's Z transforms and their estimated variances obtained from r values and sample sizes according to Rosenberg et al. (2000). Results were then back transformed to r values for presentation.
Comparisons of different groups (e.g., "poor" and "good" studies) were based on Sharp (1998) and Thompson and Sharp (1999).
Meta-analyses are potentially affected by publication bias, that is, the tendency of nonsignificant results to remain unpublished (e.g., Møller and Jennions 2001). In the absence of publication bias, data points in a plot that relates effect size to sample size (i.e., a funnel plot) are expected to be distributed symmetrically around the average effect size and to show reduced variance as sample size increases. Various methods for testing and adjusting for publication bias have been developed. I relied on Egger et al. (1997) and on Duvall and Tweedie (2000a, 2000b) for testing and adjusting for publication bias. The former detects asymmetries in the funnel plot, whereas the latter identifies "missing" studies (again on the basis of funnel plot asymmetries), fills in the missing data, and recalculates the weighted average r using an iterative procedure. The adjusted weighted average r that is obtained is thus an estimate of the weighted average r that would be obtained in the absence of publication bias. Finally, I calculated fail-safe numbers following Rosenthal (1991). The fail-safe number is the number of hypothetical studies having an effect size equal to zero that are necessary to eliminate a significant overall effect. A fail-safe number greater than 5K + 10 (where K is the number of studies) is considered to indicate a robust result.
All analyses were run using Stata 9.1 (Sterne et al. 2001), except for fail-safe numbers that were calculated using MetaWin 2.0 (Rosenberg et al. 2000). Throughout the results, I present weighted average r values as measures of effect size and their 95% confidence intervals (95% CI).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
I first checked whether studies classified as having adopted a good or poor statistical analysis (as defined in the Methods) differed in their weighted average r. As expected, poor studies had a significantly larger effect size (z = 2.44, P = 0.015). I therefore decided, conservatively, to exclude poor studies from all further analyses. I also checked whether studies conducted in captivity differed from studies of wild populations. No significant difference emerged (z = 0.01, NS), and I therefore included both captive and wild studies in all further analyses.
A meta-analysis of the relation between grooming and agonistic support revealed a weighted average r significantly greater than zero (r = 0.166, 95% CI = 0.128–0.203, n = 31, z = 8.514, P < 0.001, fail-safe N = 808.3). The test by Egger et al. (1997) revealed a significant publication bias in this sample (intercept = 0.724, t = 2.16, P = 0.039). The trim and fill method of Duvall and Tweedie (2000a, 2000b) revealed 5 missing studies. Filling in those missing studies and recalculating the weighted average r confirmed that it was significantly greater than zero (r = 0.154, 95% CI = 0.108–0.198, n = 36, z = 6.499, P < 0.001). The funnel plot in Figure 1 shows the relation between sample size and effect size. As usual in meta-analyses, effect sizes were more variable when sample size was small and converged to a common value as sample size increased.
|
When the above analyses were repeated using species values, the results did not change (r = 0.193, 95% CI = 0.124–0.272, n = 12, z = 4.997, P < 0.001, fail-safe N = 113.8). Egger's test revealed no significant bias in this sample (intercept = 0.430, t = 0.35, NS). The trim and fill method did not reveal any missing study and thus left the estimate of weighted average r unchanged.
A comparison of studies that limited their analysis to female–female interactions with those that included males in their analyses, or were limited to males, revealed no significant difference (z = 0.36, NS).
A comparison of studies that controlled or did not control for the effect of kinship revealed that the former had a slightly smaller effect size, but the difference was not significant (z = 0.64, NS). When meta-analysis was limited to those studies that had controlled for kinship, weighted average r was still significantly greater than zero (r = 0.157, 95% CI = 0.087–0.224, n = 14, z = 4.403, P < 0.001, fail-safe N = 100.7). Egger's test revealed no significant bias in this sample (intercept = 0.854, t = 1.21, NS). The trim and fill method did not reveal any missing study and left the estimate of weighted average r unchanged.
When the above analyses were repeated using species values, the results did not change (r = 0.156, 95% CI = 0.086–0.225, n = 8, z = 4.316, P < 0.001, fail-safe N = 57.7). Egger's test did not reveal a significant bias in this sample (intercept = 1.213, t = 1.80, NS). The trim and fill method, however, identified one missing study. Filling in that missing study and recalculating the weighted average r confirmed that it was significantly greater than zero (r = 0.154, 95% CI = 0.061–0.245, n = 9, z = 3.221, P = 0.001).
Finally, I limited the meta-analysis to those studies that controlled for kinship and limited analysis to female–female interactions. Weighted average r was again significantly greater than zero (r = 0.151, 95% CI = 0.074–0.226, n = 11, z = 3.806, P < 0.001, fail-safe N = 66.4). Egger's test revealed no significant bias in this sample (intercept = 1.157, t = 1.30, NS). The trim and fill method did not reveal any missing study and left the estimate of weighted average r unchanged. Species data were insufficient to carry out a meta-analysis in this subsample.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The results of this study provide evidence of a significant positive relation between grooming and agonistic support in primates on the basis of a meta-analysis of 36 independent tests carried out on 14 different species. Previous reviews of this subject emphasized the inconsistency of the available knowledge (Henzi and Barrett 1999
; Silk et al. 2004). As evident from Figure 1, such inconsistency seems to be due to statistical sampling error, so that a meta-analysis was able to clearly identify a significant relation between grooming and agonistic support.
Overall, the relationship between grooming and support was rather weak (r = 0.157 controlling for kinship). Although current meta-analytical techniques allow testing and adjusting for publication bias, the power of these methods is reduced when smaller subsamples are analyzed. I cannot therefore exclude that my current estimates of effect sizes are slightly overestimated. Rather, weak effect sizes are also generally reported in ecological/evolutionary studies (Møller and Jennions 2002). Because agonistic support occurs at relatively low rates, obtaining adequate sample sizes to ensure statistical analyses have sufficient power is often problematic. These problems are not unique to agonistic support (e.g., deception and group hunting) and highlight the usefulness of meta-analysis in tackling biological phenomena characterized by small effect sizes and relative rarity.
The literature on reciprocal altruism is characterized by conflicting results, unreplicated findings, and contrasting interpretations (e.g., Lazarus and Metcalfe 1990; Milinski 1990; Noë 1992). The relation between grooming and agonistic support in primates is, in this context, a prime example. Initially conceived as a hypothetical assumption underlying a theoretical model of primate social structure (Seyfarth 1977), there has subsequently been a tendency to accept it uncritically as a fact. In more recent years, renewed criticism questioned the possibility that primates interchange different altruistic acts (Henzi and Barrett 1999). By quantitatively summarizing available information, meta-analysis seemed the ideal tool to be applied to such a confused field. Indeed, its results were clear-cut. A significant interchange of grooming and agonistic support was evidenced, and this effect was not due to the confounding influence of kinship.
The results reported here do not provide information on the proximate mechanism underlying the interchange of grooming and agonistic support. Symmetry based, attitudinal, and calculated reciprocity are 3 possible mechanisms that have been proposed to underlie the evolution of reciprocal altruism (Brosnan and de Waal 2002). Careful experimental and/or temporal analyses of reciprocal exchanges are needed to fully understand this subject. Also, because most of the data included into the meta-analysis were correlational, any inference about causal relations must be taken with caution.
Regardless of the proximate mechanisms involved, my results show that individuals that groom frequently also support each other frequently. In terms of ultimate function, grooming and agonistic support may be associated as part of a system of low-cost reciprocal altruism. When witnessing an agonistic interaction between 2 group mates, an animal has to choose which of the contestants to support, and its choice appears to be determined at least partially by the history of grooming received by each of the group mates. Such a system of low-cost reciprocal altruism would thus essentially be based on comparative partner choice. This interpretation is reminiscent of the "by-product reciprocity" of Sachs et al. (2004), but it differs in the more central role played by partner choice: subjects choose whom to assign the by-product benefits of their selfish behaviors in relation to the benefits received in the past.
Interestingly, up to now, the widespread observation that primates tend to direct their grooming up the hierarchy (Schino 2001) posed somewhat of a challenge to animal behaviorists. Although it suggested that grooming might be exchanged with rank-related benefits, little hard evidence of such interchange existed. Thus, the results presented here and those of a previous meta-analysis of grooming distribution (Schino 2001) nicely complement each other. The next step in our understanding of the evolution of primate reciprocal altruism will need an exploration of the interspecific and ecological variability that characterizes social primates. In particular, it remains to be fully understood why agonistic alliances are common in some species and rare in others and how the services primates exchange vary according to differences in ecological conditions (Barrett et al. 2002; Pazol and Cords 2005).
In conclusion, this meta-analysis shows that the inconsistent results of previous attempts to quantify the relationships between grooming and agonistic support were likely due to statistical sampling error and that a significant relation emerges when modern methods of systematic review are applied to the available literature. Meta-analysis represents a powerful tool for behavioral biologists investigating phenomena that, despite their biological significance, are characterized by small effect size and low frequency of occurrence.
|
ACKNOWLEDGEMENTS |
|---|
I thank Bonaventura Majolo, Jeroen Stevens, and Raffaella Ventura for kindly providing unpublished information and Robert Seyfarth for useful comments.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Arnqvist G and Wooster DC. (1995) Meta-analysis: synthesizing research findings in ecology and evolution. Trends Ecol Evol 10:236–240.[CrossRef]
Barrett L, Gaynor D, Henzi SP. (2002) A dynamic interaction between aggression and grooming reciprocity among female chacma baboons. Anim Behav 63:1047–1053.[CrossRef]
Bercovitch FB. (1988) Coalitions cooperation and reproductive tactics among adult male baboons. Anim Behav 36:1198–1209.[CrossRef]
Brosnan SF and de Waal FBM. (2002) A proximate perspective on reciprocal altruism. Hum Nat 13:129–152.
Chapais B. (2001) Primate nepotism: what is the explanatory value of kin selection? Int J Primatol 22:203–229.
Chapais B, Girard M, Primi G. (1991) Non-kin alliances and the stability of matrilineal dominance relations in Japanese macaques. Anim Behav 41:481–491.
Clutton-Brock TH. (2002) Breeding together: kin selection and mutualism in cooperative societies. Science 296:69–72.
Cooper MA, Bernstein IS, Hemelrijk CK. (2005) Reconciliation and relationship quality in Assamese macaques (Macaca assamensis). Am J Primatol 65:269–282.[CrossRef][Web of Science][Medline]
de Vries H. (1993) The rowwise correlation between two proximity matrices and the partial rowwise correlation. Psychometrika 58:53–69.[CrossRef]
de Waal FBM. (1997) The chimpanzee's service economy: food for grooming. Evol Hum Behav 18:375–386.[CrossRef]
de Waal FBM and Luttrell LM. (1986) The similarity principle underlying social bonding among female rhesus monkeys. Folia Primatol 46:215–234.
de Waal FBM and Luttrell LM. (1989) Mechanisms of social reciprocity in three primate species: symmetrical relationship characteristics or cognition? Ethol Sociobiol 9:101–118.
Dugatkin LA. (1988) Do guppies play tit for tat during predator inspection visits? Behav Ecol Sociobiol 23:395–399.[CrossRef]
Dunbar RIM. (1980) Determinants and evolutionary consequences of dominance among female gelada baboons. Behav Ecol Sociobiol 7:253–265.[CrossRef]
Dunbar RIM and Sharman M. (1984) Is social grooming altruistic? Z Tierpsychol 64:163–173.
Duvall S and Tweedie R. (2000a) A non-parametric "trim and fill" method of assessing publication bias in meta-analysis. J Am Stat Assoc 95:89–98.[CrossRef][Web of Science]
Duvall S and Tweedie R. (2000b) Trim and fill: a simple funnel plot based method of testing and adjusting for publication bias in meta-analysis. Biometrics 56:455–463.[CrossRef][Web of Science][Medline]
Egger M, Smith GD, Altman DG. (2001) Systematic reviews in health care: meta-analysis in context(BMJ Publishing Group, London).
Egger M, Smith GD, Schneider M, Minder C. (1997) Bias in meta-analysis detected by a simple graphical test. Br Med J 315:629–634.
Gontard-Danek MC and Møller AP. (1999) The strength of sexual selection: a meta-analysis of bird studies. Behav Ecol 10:476–486.
Hamilton WD. (1964) The genetical evolution of social behaviour. J Theor Biol 7:1–51.[CrossRef][Web of Science][Medline]
Hemelrijk CK. (1990) A matrix partial correlation test used in investigations of reciprocity and other social interaction patterns at group level. J Theor Biol 143:405–420.
Hemelrijk CK. (1994) Support for being groomed in long-tailed macaques Macaca fascicularis. Anim Behav 48:479–481.[CrossRef]
Hemelrijk CK and Ek A. (1991) Reciprocity and interchange of grooming and ‘support’ in captive chimpanzees. Anim Behav 41:923–935.[CrossRef]
Henzi SP and Barrett L. (1999) The value of grooming to female primates. Primates 40:47–59.
Jones CB. (1979) Grooming in the mantled howler monkey, Alouatta palliata Gray. Primates 20:289–292.
Jones CB. (2006) Exploring the utility of van Schaik's model of male coalitions for female mantled howler monkeys (Alouatta palliata). Lab Primate Newslett 45:4–6.
Kapsalis E and Berman CM. (1996) Models of affiliative relationships among free-ranging rhesus monkeys (Macaca mulatta). II. Testing predictions for three hypothesized organizing principles. Behaviour 133:1235–1263.
Koyama NF, Caws C, Aureli F. Interchange of grooming and agonistic support in chimpanzees. Int J Primatol Forthcoming.
Lazarus J and Metcalfe NB. (1990) Tit-for-tat cooperation in sticklebacks: a critique of Milinski. Anim Behav 39:987–988.[CrossRef]
Manson J, Rose LM, Perry S, Gros-Louis J. (1999) Dynamics of female-female relationships in wild Cebus capucinus: data from two Costa Rican sites. Int J Primatol 20:679–706.[CrossRef]
Matheson MD and Bernstein IS. (2000) Grooming, social bonding and agonistic aiding in rhesus monkeys. Am J Primatol 51:177–186.[CrossRef][Web of Science][Medline]
Milinski M. (1987) Tit for tat in sticklebacks and the evolution of cooperation. Nature 352:433–135.
Milinski M. (1990) No alternative to tit-for-tat cooperation in sticklebacks. Anim Behav 39:989–991.[CrossRef]
Møller AP and Jennions MD. (2001) Testing and adjusting for publication bias. Trends Ecol Evol 16:580–586.[CrossRef]
Møller AP and Jennions MD. (2002) How much variance can be explained by ecologists and evolutionary biologists? Oecologia 132:492–500.[CrossRef]
Nishida T and Hosaka K. (1996) Coalition strategies among adult male chimpanzees of the Mahale Mountains Tanzania. In McGrew WT, Marchant LF, Nishida T (Eds.). Great ape societies(Cambridge University Press, Cambridge) pp. 114–134.
Noë R. (1992) Alliance formation among male baboons: shopping for profitable partners. In Harcourt AH and de Waal FBM (Eds.). Coalitions and alliances in humans and other animals(Oxford University Press, Oxford) pp. 285–321.
Noë R and Hammerstein P. (1995) Biological markets. Trends Ecol Evol 10:336–340.[CrossRef]
O'Brien T. (1993) Allogrooming behaviour among adult female wedge-capped capuchin monkeys. Anim Behav 46:499–510.[CrossRef]
Packer C. (1977) Reciprocal altruism in Papio anubis. Nature 265:441–443.[CrossRef]
Payne HFP, Lawes MJ, Henzi SP. (2003) Competition and the exchange of grooming among female samango monkeys (Cercopithecus mitis erythrarchus). Behaviour 140:453–471.[CrossRef]
Pazol K and Cords M. (2005) Seasonal variation in feeding behaviour competition and female social relationships in a forest dwelling guenon the blue monkeys (Cercopithecus mitis stuhlmanni) in the Kakamega Forest, Kenya. Behav Ecol Sociobiol 58:566–577.[CrossRef]
Pereira ME and McGlynn CA. (1997) Special relations instead of female dominance for redfronted lemurs, Eulemur fulvus fulvus. Am J Primatol 43:239–258.[CrossRef][Web of Science][Medline]
Polizzi di Sorrentino E. 2005. Relazioni fra grooming, aggressioni e coalizioni nel macaco del Giappone (Macaca fuscata). Tesi di Laurea, Università La Sapienza, Rome.
Perry S. (1998) Male-male social relationships in wild white-faced capichins, Cebus capucinus. Behaviour 135:139–172.
Rosenberg MS, Adams DC, Gurevitch J. (2000) Meta-Win: statistical software for meta-analysis. Version 2.0(Sinauer Associates, Sunderland (MA)).
Rosenthal R. (1991) Meta-analytic procedures for social research(Sage, Beverly Hills (CA)).
Sachs JL, Mueller UG, Wilcox TP, Bull JB. (2004) The evolution of cooperation. Q Rev Biol 79:135–160.[CrossRef][Medline]
Schino G. (2001) Grooming competition and social rank among female primates: a meta-analysis. Anim Behav 62:265–271.[CrossRef]
Schino G, Ventura R, Troisi A. (2003) Grooming among female Japanese macaques: distinguishing between reciprocation and interchange. Behav Ecol 14:887–891.
Seyfarth RM. (1976) Social relationships among adult female baboons. Anim Behav 24:917–938.[CrossRef]
Seyfarth RM. (1977) A model of social grooming among adult female monkeys. J Theor Biol 65:671–698.[CrossRef][Web of Science][Medline]
Seyfarth RM. (1980) The distribution of grooming and related behaviours among adult female vervet monkeys. Anim Behav 28:798–813.[CrossRef]
Seyfarth RM and Cheney DL. (1984) Grooming, alliances and reciprocal altruism in vervet monkeys. Nature 308:541–543.[CrossRef][Medline]
Sharp SJ. (1998) Meta-analysis regression. Stata Technical Bull 42:16–22.
Silk JB. (1982) Altruism among female Macaca radiata: explanations and analysis of patterns of grooming and coalition formation. Behaviour 79:162–188.
Silk JB. (1992) The patterning of intervention among male bonnet macaques: reciprocity, revenge, and loyalty. Curr Anthropol 33:318–325.[CrossRef]
Silk JB. (2002) Kin selection in primate groups. Int J Primatol 23:849–875.[CrossRef]
Silk JB, Alberts SC, Altmann J. (2004) Patterns of coalition formation by adult female baboons in Amboseli, Kenya. Anim Behav 67:573–582.[CrossRef]
Sterne JAC, Bradburn MJ, Egger M. (2001) Meta-analysis in Stata. In Egger M, Smith GD, Altman DG (Eds.). Systematic reviews in health care: meta-analysis in context(BMJ Publishing Group, London) pp. 347–369.
Stevens JMG, Vervaecke H, de Vries H, van Elsacker L. (2005) The influence of the steepness of dominance hierarchies on reciprocity and interchange in captive groups of bonobos (Pan paniscus). Behaviour 142:941–960.[CrossRef]
Thompson SG and Sharp SJ. (1999) Explaining heterogeneity in meta-analysis: a comparison of methods. Stat Med 18:2693–2708.[CrossRef][Web of Science][Medline]
Trivers RL. (1971) The evolution of reciprocal altruism. Q Rev Biol 46:35–57.[CrossRef]
Ventura R, Majolo B, Koyama NF, Hardie S, Schino G. Reciprocation and interchange in wild Japanese macaques: grooming, co-feeding, and agonistic support. Am J Primatol Forthcoming.
Vervaecke H, de Vries H, van Elsacker L. (2000) The pivotal role of rank in grooming and support behaviour in a captive group of bonobos (Pan paniscus). Behaviour 137:1463–1485.[CrossRef]
Watts DP. (2002) Reciprocity and interchange in the social relationships of wild male chimpanzees. Behaviour 139:343–370.[CrossRef]
Wilkinson G. (1984) Reciprocal food sharing in the vampire bat. Nature 308:181–184.[CrossRef]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. R Madden and T. H Clutton-Brock Manipulating grooming by decreasing ectoparasite load causes unpredicted changes in antagonism Proc R Soc B, April 7, 2009; 276(1660): 1263 - 1268. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. N Radford Duration and outcome of intergroup conflict influences intragroup affiliative behaviour Proc R Soc B, December 22, 2008; 275(1653): 2787 - 2791. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Schino and F. Aureli Grooming reciprocation among female primates: a meta-analysis Biol Lett, February 23, 2008; 4(1): 9 - 11. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||