Animal personalities, i.e. consistent differences in behavior across time and/or context, have received increased attention of behavioral biologists over the last years. Recent research shows that personalities represent traits on which natural and sexual selection work and which can have substantial fitness consequences. The aim of this study is to establish the personality structure of crested macaque (Macaca nigra) males as foundation for future studies on its adaptive value. We collected behavioral data through focal animal sampling and additionally conducted two sets of playback experiments. Results of a factor analysis on the behavioral data revealed a four factor structure with components we labeled Anxiety, Sociability, Connectedness and Aggressiveness. Results from the experiments revealed an additional and independent Boldness factor but the absence of Neophilia. Overall, this structure resembles other macaque and animal species with the exception of Connectedness, which might be a consequence of the species' tolerant social style. Our results thus not only form the basis for future studies on the adaptive value of personality in crested macaques but also contribute an important data point for investigating the evolution of personality structure from a comparative perspective by refining, for example, which personality factors characterized the last common ancestor of hominids and macaques.
Citation: Neumann C, Agil M, Widdig A, Engelhardt A (2013) Personality of Wild Male Crested Macaques (Macaca nigra). PLoS ONE 8(8): e69383. doi:10.1371/journal.pone.0069383
Editor: Allan V. Kalueff, Tulane University Medical School, United States of America
Received: March 27, 2013; Accepted: June 9, 2013; Published: August 5, 2013
Copyright: © 2013 Neumann et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This project was funded by the German Research Foundation (http://www.dfg.de), within the Emmy-Noether Program (grant No. EN 719/2 to AE). Additionally, CN received a stipend of the Saxon State Ministry for Higher Education, Research and the Arts (http://sachsen.de/en/1218.htm). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: Two authors (AE and AW) currently serve on the Editorial Board of PlosOne. This does not alter the authors' adherence to all the PLOS ONE policies on sharing data and materials.
In recent years, the phenomenon that individuals in many, if not all, animal species differ from each other consistently in their behavior has received increasing attention from biologists and psychologists , . If such individual differences are stable over time and/or across contexts they are referred to as animal personality or temperament . Animal personalities are identified either by humans familiar with the study subjects via rating items in questionnaires (e.g. –) or through observational/experimental data (e.g. –). Both approaches usually subject primary data to factor analysis or related statistical methods, which allow identification of underlying dimensions that are then referred to as personality factors. Generally, both approaches lead to similar results when describing the personality structure (the combination of all present personality factors) of a given animal species (e.g., , but see , reviewed in ). Since personality can have profound effects on reproductive fitness, it is an important feature of animal biology that selective processes can work on –.
Personalities of non-human primates (from here on: primates) are reasonably well studied (reviewed in ) due to the phylogenetic proximity of primates to humans and the complexity of the social systems exhibited by most primate species . This growing body of data allows us to investigate the biological roots of human personality , ,  and recently, a first formal attempt was made to describe the evolution of personality within the catarrhine primates (, see also ). Weiss and colleagues  hypothesize that the personality structure of humans is the evolutionary consequence of a series of changes along our phylogenetic history, with the human personality structure resembling more other hominid personality structures than those of more distantly related species such as rhesus macaques (Macaca mulatta). They further suggest that degree of sociality (i.e., differences in social organization and social relationships) has been the major selective pressure leading to the observed inter-specific differences in personality. This latter proposition however raises the question as to how personality structures differ in closely related species in which sociality may not be confounded by overall social organization (as is likely to be the case, for example, within the hominids).
In an alternative scenario, it has been suggested that the evolution of different types of sociality in different species is the result of differences in personality , . For example, species in which individuals prefer close spatial proximity with conspecifics (an exemplified personality factor) may face increased feeding competition as compared to species in which individuals prefer greater spatial distance to conspecifics. This might then lead to more intense or more frequent aggressive behavior, which in turn may lead to the evolution of specific conflict management strategies not found in species that show lower levels of proximity . This example illustrates how differences in personality could lead to changes in patterns of social behavior and shape social styles as sets of correlated behavioral patterns .
Among primates, the macaques (Macaca) represent a particularly well suited genus to look into the evolution of personality structures and its link to social behavior. On the one hand all of the approximately 20 species share some fundamental features of social organization such as living in temporally relatively stable multi-male/multi-female groups and male dispersal around sexual maturity . On the other hand, macaques differ markedly in social style (, see also –), for example in regard to patterns of aggression and the degree of dominance asymmetries, conflict management strategies, and degree of kin bias in social interactions (e.g. , –). This particular diversity of social traits has led to the classification of macaque species into four grades of social styles ranging from so called despotic to tolerant species  and therefore represents an interesting model to study the link between sociality and personality.
Several macaque species have been studied in the context of personality. Our current knowledge about macaque personalities nevertheless comes foremost from studies on a single species, the despotic rhesus macaque (e.g. , , , ). In this species, at least three personality factors have been determined consistently across different studies: sociability (also labeled affiliation or not being solitary), aggressiveness (also labeled hostility), and fearfulness (also labeled excitability). These three factors seem to have deep phylogenetic roots beyond primates given that they appear not only in macaques , , but also in many other non-primate taxa (reviewed in ). Incidentally, they are thought to correspond to three dimensions of the dominating model of human personality – the Five Factor Model, i.e. Extraversion, Agreeableness, and Neuroticism , , . Despite these possible similarities, there are also differences in personality structure between different macaque species and deviations from the human model. For instance, a distinct and separate personality factor Dominance , , ,  has been described for hominids, but not for humans, and similar variation can be observed in the macaques: whereas in despotic rhesus macaques Dominance has been described as a distinct personality factor , it seems to be absent in the more tolerant Barbary macaques (M. sylvanus, ). Collectively, the available data suggest that personality structures of macaque species share some similarities but simultaneously show differences that are possibly linked to differences in species-specific social styles.
As illustrated within the macaques, this interplay of personalities and social styles is still poorly understood and certainly more data in all these domains are needed to form the basis for broader inter-species comparisons. An extensive body of data on macaque personality, social behavior and ecology already exists, that allows us to investigate this interaction within this interesting genus. Still, species at the tolerant end of the spectrum of social styles are underrepresented in all these respects and to the best of our knowledge, studies assessing personality structure in the most tolerant macaque species are completely missing.
The overall aim of this study is therefore to describe the personality structure of crested macaque males (M. nigra) as a foundation for future studies on its adaptive value. Following the macaque-typical pattern, crested macaques live in permanent multi-male/multi-female groups from which males disperse around reaching adulthood. Crested macaques have been classified into the very tolerant end of the macaque social style spectrum (c.f. ), based on the observation that, for example, social networks are diverse and that aggressive conflicts are frequently reconciled, and are of relatively low intensity with frequent occurrence of counter-aggression , , , , . We therefore expect crested macaque personality to be more similar to personalities of tolerant species as compared to more despotic species. In particular we expect it to reflect the species-typical tolerant social style, which should manifest itself in the absence of a Dominance factor and the emphasis of factors in the social positive domain , . In describing the crested macaque personality structure, we will enhance our knowledge on personality structures in tolerant primates as well as contribute to the clarification of the evolutionary history of personality within the macaques and primates including humans in general.
All research was conducted non-invasively on a wild population of crested macaques and in accordance with the Animal Behaviour Society's guidelines for the treatment of animals in behavioural research and teaching. In addition, we adhered to all relevant regulations of Indonesia and Germany. Permission to conduct the study in the Tangkoko-Batuangus Nature Reserve in Indonesia was granted by the Indonesian State Ministry of Research and Technology (RISTEK, permit 1189/FRP/SM/VI/2008), the Directorate General of Forest Protection and Nature Conservation (PHKA, permit SI.154/Set-3/2008) in Jakarta and the Department for the Conservation of Natural Resources (BKSDA, permit 58/SIMAKSI/BKSDA-SU/2009) in Manado. Since the study was non-invasive and approved by the local authorities in Indonesia, our institutions did not require approval by an ethics committee. Macaca nigra is classified as critically endangered  and our study did not affect the animals' welfare.
Study subjects and site
Between March 2009 and May 2011, we studied 37 males of two wild, non-provisioned, groups of crested macaques, groups R1 and PB, living in the Tangkoko Nature Reserve, Sulawesi, Indonesia , . The two groups comprised up to 85 individuals each, with 7–18 adult males present and are subject of research intermittently since the 1990's. All animals were completely habituated to human observers and adults were individually recognizable based on facial features and body markings, e.g., scars or broken limbs.
To assess personality, we used a combination of behavioral observations and experiments. First, we used focal animal and scan sampling  of 37 adult males (mean = 66.1 h, range = 0.6–130.0 h per male), and collected data on a range of specific behaviors and identities of other adults in spatial proximity of focal subjects. The selection of behavioral variables to include (Table 1) was designed to cover a broad field of social behavior and was based on a published account of the behavioral repertoire of crested macaques  and supplemented by variables suggested by a recent study on chimpanzee personality (, see also ). Focal protocols lasted 60 minutes during which continuous data on social and aggressive behavior and interaction partner identities were recorded. While following focal animals, these observational data were entered in handheld computers (Psion Workabout Pro G2) in real-time using spread-sheet software (PTab Spreadsheet v.3.0; Z4Soft). Additionally, we conducted scans at intervals of one, five, and thirty minutes to record general focal animal activity, identity of adult individuals in proximity, and whether the focal animal was outside vs. inside of the group, respectively (see Table 1). Data were collected by four observers and inter-observer reliability of the observed behaviors ranged between 0.75 and 1.00 as assessed by Pearson correlation coefficient and Cohen's kappa .
Second, we used playback experiments on 18 of our focal males, to assess two possible personality factors that are hard to quantify by passive observations alone: boldness and neophilia , . Boldness was assayed with the presentation of a dog bark bout (audio S1). Both crested macaque groups from time to time meet dogs from the nearby village in the forest and give alarm calls upon their sighting [Engelhardt et al., pers. obs.]. Neophilia was measured as the reaction to a donkey bray (unknown to test animals, audio S2). We conducted a total of 57 (n = 17 males) and 43 (n = 16 males) experimental trials in the dog and donkey condition, respectively. Males were presented with each of the stimuli repeatedly (up to 3 times), two consecutive trials of the same condition being separated by at least three weeks, and each male participating only once per day in an experiment. Stimuli were presented from a concealed DavidActive speaker (Visonik, Germany) connected to a Marantz PMD660 Flash-Disc recorder, placed 10–20 meters from the subject. The speaker was operated by an assistant hiding behind natural obstacles, for instance a tree trunk or buttress root, in such a way that neither the assistant nor the speaker was visible to the subject. Playbacks were only carried out when the subject sat calmly on the ground facing the direction in which the group was generally travelling. The speaker was placed in a 90° (±45°) angle relative to the subject's body orientation (see  for more details and an illustration of the experimental setup). The response to the playback stimulus was filmed with a digital video camera (Sony DCR-HC 90E) operated by the experimenter standing at a distance of about five meters from the subject (see video S1). In each experimental condition, we used the same recording as stimulus during all trials. We can therefore not rule out that subjects habituated to the repeated presentations of the same stimulus. Analytical methods, however, allow accounting for this possible bias in results (see methods on adjusted repeatabilities below).
We divided the overall data collection period into blocks of two months and calculated frequencies of behaviors, number of individuals in proximity and indices describing the diversity of individuals in proximity and grooming partners in each of these time blocks separately for each male (Table 1). There are two reasons for creating these time blocks. First, we wanted to assess temporal stability of our behavioral variables by calculating repeatabilities and repeated measurements are needed for each individual to obtain this measure , . Second, group composition, particularly with respect to adult males, changed frequently . Such an approach would lead to difficulties in the calculation of diversity indices, because in order to make these indices comparable they need to be standardized by accounting for maximum number of potential interaction partners .
An individual data point was included if the cumulative observation time for a given male and time block was at least six hours. If necessary, the raw behavioral variables were transformed (log, square root, arcsine) to achieve symmetric distributions. Variables were then standardized to a mean of zero and a standard deviation of one. Subsequently, all variables were tested for repeatability , . We considered variables to be significant if the 95% confidence interval around their repeatability estimate did not include zero. Only significantly repeatable variables, i.e. variables that showed temporal stability, were subsequently subjected to factor analysis (see below). Repeatability estimates we present are well in the range of other studies on primates and non-primates (e.g., , , –). We tested the behavioral variables for group differences by means of Mann-Whitney U tests. One variable (number of female neighbors) differed significantly between the two study groups. Recalculating repeatability controlling for this group difference lowered the repeatability estimate of this variable, but did not change its statistical significance.
Since factor analysis requires independent data, and our data structure consisted of repeated measurements of individuals, we averaged values of the two-month time blocks to obtain single values for each male, thereby avoiding pseudo-replication. This procedure resulted in a data set comprising 30 males, with at least one two-month time block during which a male was observed for at least six hours.
The Kaiser-Meyer-Olkin measure of sampling adequacy (KMO = 0.60) indicated that our data were suitable for factor analysis , even though overall sample size and case-variable ratio were small. We performed our analysis using the correlation matrix and a minimum residual solution. We decided to extract four factors based on visual inspection of a scree plot and eigenvalues . We used an oblique (type “oblimin”) instead of an orthogonal rotation to allow for factors to be correlated . We chose this approach because there is no a priori reason to assume that personality factors are independent of each other . To gauge the relative importance of the behavioral variables to the extracted factors, we used factor loadings . For interpretation, we considered variables that loaded saliently (absolute value ≥0.40) to contribute to a given factor . If a variable loaded saliently on more than one factor we interpreted this variable as contributing to the factor on which it loaded with the highest absolute value (e.g., ).
Male responses to the playback stimuli from all experimental trials were coded from videotapes frame-by-frame by two coders, one being blind to study design and experimental condition. We discarded 19 experimental trials that were judged non-valid by both coders, due to technical problems, or subject distraction. Response variable was the time subjects oriented themselves towards the speaker in the first 10 seconds after the start of the stimulus presentation (hereafter: orientation duration) , . We considered such a response to occur when the subject oriented its head towards or approached the speaker within an angle of about 22.5°. We considered stronger responses, i.e. longer orientation durations, to indicate bolder and more neophilic males. Both raters expressed high agreement in the orientation duration these responses had (Pearson correlation: r = 0.92, N = 81, p<0.001). Average duration values from both coders were used in subsequent analyses. We calculated adjusted repeatabilities (Radj, c.f. ) controlling for trial number within male subject and experimental condition to account for possible habituation effects. Since this algorithm does not permit confidence interval borders to be smaller than zero, we used p-values to determine statistical significance .
Finally, we computed correlations between the factors as well as between the factors and the responses to the two experiments. For this we extracted regression scores from the factor analysis for those males that were also subjects in the playback experiments, i.e. N = 17 (dog condition) and N = 16 (donkey condition). For the experimental data we used average male orientation durations (within each condition) for this calculation. We calculated Pearson correlation coefficients and since we had no specific hypotheses regarding possible relationships between personality factors, corrected the resulting p-values for multiple testing . All analyses were conducted in R 2.15.0  with the packages psych  and rptR .
The majority of the 22 behavioral variables were moderately repeatable, indicating that their expression was stable over time (Table 2). Five of the behavioral variables we considered were not repeatable, from which four reflected behavior towards other adult males: diversity of male neighbors (in close proximity) and the rates with which other males were approached, threatened and aggressed. In addition, rates of affiliation directed at individuals other than adult males, were not repeatable.
The factor analysis based on the remaining 17 repeatable variables explained 62% of the total variance. All variables loaded on at least one of the four factors with an absolute value of >0.4 with six variables loading on two factors. The solution of loadings of variables onto the extracted factors after oblimin rotation is presented in Table 3.
The first factor we extracted explained 21% of the variance with loadings of anxiety related behaviors (self-grooming, self-scratching). Additionally, males scoring higher on this factor were less active, gave less dominance displays (loud calls) and approached females more rarely. We labeled this factor Anxiety (Table 3).
The second factor explained 17% of variance and included variables that reflect diversity of male and female neighbors (diversity of close female neighbors, diversity of far female neighbors, diversity of far male neighbors) and the diversity of female grooming partners (female groomees). Additionally, this factor was related to smaller frequencies of yawning and spending more time in the core of the group. We labeled this factor Connectedness (Table 3).
The third factor accounted for 13% of variance with variables reflecting general social behavior, i.e. proportion of time spent grooming, diversity of grooming partners in general, and number of female neighbors. Interestingly, positive (affiliative) behavior directed at adult males loaded negatively on this factor. We named this factor Sociability (Table 3).
The fourth factor explained 11% of variance and reflected threat and overt aggression directed at individuals other than adult males. We labeled this factor Aggressiveness (Table 3).
Males reacted to the dog stimulus in all of the 39 trials. In the donkey condition, males reacted in 36 trials, while no reaction of males was visible on the video tapes in 6 trials. We found significant repeatability in response to the dog condition (Radj = 0.55, 95% CI: 0.20–0.63, p = 0.002). We therefore considered the responses to this playback reflecting Boldness. In contrast, we did not find responses to the donkey condition to be repeatable (Radj = 0.05, 95% CI: 0.00–0.27, p = 0.099) and therefore consider Neophilia to be absent.
Relationships between the factors
After controlling for multiple testing, we did not find any significant correlation between any pair of the four observationally assessed factors and the boldness factor (Pearson correlation coefficients, mean = −0.05, range = −0.39–0.37, p value range: 0.047–0.940, ten pairwise comparisons, Table 4). In addition, we found no significant relationship between any of these five factors and orientation duration in response to the donkey playback (Pearson correlation coefficients, mean = −0.02, range = −0.39–0.32, p value range: 0.131–0.877, five pairwise comparisons, Table 4).
Our results on observational and experimental data suggest that crested macaque personality comprises five distinct and unrelated factors: Anxiety, Connectedness, Sociability, Aggressiveness and Boldness (summarized in Table 5). This structure is generally similar to the structures observed in other macaque species (e.g., , , ). In addition, the structure our data suggest is characterized by the presence of two distinct factors that reflect socio-positive behavior (Connectedness and Sociability). Most notably, we identified a factor, Connectedness, which, to our knowledge, has not been described in studies of primate personalities before, and which covers aspects of social network diversity, a feature that might be of particular importance in tolerant as opposed to despotic primate species.
Our study is focused on adult males because ultimately we are interested in the fitness consequences of personality among males. Given this, we cannot exclude the possibility that our description of the personality structure of crested macaques is incomplete. However, to our knowledge no empirical study so far has shown sex-specific differences in personality structure within a species.
Anxiety, Aggressiveness and Sociability in our study matched factors that have been described in previous studies of macaque personalities , ,  and are widespread among other primate and non-primate species (reviewed in , ). Anxiety is commonly used to describe general unease and distress . That the factor we name Anxiety reflects the individual degree of unease and distress in our study species is evident through the loading of self-directed behavior onto this factor, which is a behavioral manifestation of physiological stress levels . In our study, males scoring high on Anxiety approached females less frequent than less anxious males. Anxiety might therefore reflect the reluctance or willingness of males to approach females, possibly mediated through general unease whilst in female proximity (see also ). In addition, our measure of general activity also loaded on the Anxiety factor. Similarly, in Barbary macaques, questionnaire items, i.e. adjectives to be rated by human observers, typically describing distress (e.g., tense, irritable, excitable) loaded on one single factor alongside items that describe general activity (active, lazy). Consequently, this factor has been labeled Activity/Excitability in this study . We suggest, however, that Anxiety might in fact be the more fitting label for crested and possibly also Barbary macaques since it captures more of an intrinsic feature as compared to Activity which may well be constrained by external factors, such as the environment. For example, Barbary macaques are more active (i.e. they rest less) when their home-range is covered with snow . Given that Barbary macaques are also classified as relatively tolerant, this combination of anxiety and activity represented in only one personality factor might constitute a general feature of the more tolerant macaque species. In contrast, in despotic rhesus macaques both factors are distinct (, see also  for this pattern in chimpanzees, Pan troglodytes).
In crested macaque males, the personality factor Aggressiveness covers threat behavior and overt aggression directed at females, sub-adults, juveniles and infants. Interestingly, aggressive behavior directed at adult male group members was not part of this factor. The reason for this was that rates of aggression towards adult males were not stable over time and hence not included into our factor analysis. As previously mentioned, we observed frequent changes in composition as well as in dominance relationships among adult males in our groups, whereas among adult females, group composition and dominance hierarchy remained much more stable over the course of our study , . It could be argued that temporal stability or instability in behavior is the consequence of a stable or instable social environment, and consequently, stable behavioral patterns may not reflect an intrinsic property (i.e. personality), but rather are the result of external constraints , . Therefore, the lack of repeatability in our study of aggression towards other males may have been a consequence of these dynamics. Indeed, having a stereotypic aggression rate towards other males regardless of the dynamics among adult males might be mal-adaptive in the sense that this might lead to costs imposed by overly frequent and possibly injuring aggression depending on how many males co-reside in the group.
Aggressiveness has also been found in other primate (including macaques) and non-primate species, but it is frequently labeled Confidence or Dominance in studies based on observers' ratings such as aggressive, bullying, dominant, submissive and confident (, , , but see ). It often correlates with behaviorally assessed (agonistic) dominance ranks , , , , . In studies of animal behavior, dominance is however considered a dynamic individual property that changes over time due to external events (e.g., challenges by other individuals or migration) and which, in the strictest sense, is the property of a single individual within a dyad where the other individual is subordinate at a given time ). Given the possible confusion with the term dominance as used in behavioral biology (, ; see Capitanio  for an illustrative example how such confusion between personality Dominance and behavioral dominance might easily arise) we suggest that the label of the Dominance facet of primate personality should be reevaluated. For rhesus macaques, for example, based on the items that describe it (see above) the label Dominance could, in alignment with our study, be replaced by Aggressiveness. Such more consistent labeling of personality factors would facilitate inter-specific comparisons.
The final personality factor that crested macaques share with other macaques is Sociability. This factor covers behavior that seems essential for building and maintaining social relationships with female group members, particularly the amount of grooming, the diversity of grooming partners and the number of female neighbors. Sociability as defined in our study, matches the Friendliness and Sociability components of other macaque species' personality structures , , , , in which it is associated with rating items like sociable, gentle and friendly , .
In addition to the three above mentioned personality factors that appear to generally occur in macaques, we found a further personality factor in crested macaques, Connectedness. Interestingly, this factor in addition to Sociability also covers behavior in the socio-positive domain. It describes the diversity of females which are groomed by a specific male, the diversity of adult individuals in proximity to him and the time males spend in the core of their social group. Connectedness has no obvious homologue in other macaque species. However, given the overall socio-positive notion of this factor we speculate that in rhesus and Barbary macaques it may be part of the Friendliness factor. Consequently, whereas Friendliness (or Sociability) constitutes a single factor in some species, we find that in crested macaques, socio-positive behavior is reflected in two distinct personality factors, Sociability and Connectedness.
Finding such an additional socio-positive dimension might be a consequence of the complex social network that individuals of tolerant macaque species, such as crested macaques, have as compared to individuals of more despotic species , . This wide network connecting also not related individuals shows that relationships between group members are far less constrained than in despotic species where relationships are maintained predominantly within fixed kin networks (e.g., , ). Two possible ways in which scoring high on such a personality factor might thus be biologically adaptive (fitness increase) are through a more diverse network of allies that could provide support in agonistic conflicts or through increased attractiveness to the opposite sex (e.g., –). Further studies will be necessary to determine whether Connectedness indeed has fitness benefits and whether it is a general feature of socially tolerant primates such as crested macaques.
The final personality factor we identified in crested macaque males is Boldness as evidenced by repeatable reactions towards the auditory presentation of a threating stimulus. Boldness has rarely been studied in primates in general . Such a factor might however evolve under the selective pressure of predation, particularly in species in which predators are mobbed upon detection as is the case for crested macaques on sight of reticulated pythons (Python reticulatus), their presumed primary predator . Scoring high in Boldness could under these circumstances not only help to roust the predator, but may also be used as a potentially costly signal of social status or as a means of attractiveness to potential mating partners. In this sense, we cannot rule out the possibility that the composition of the audience present during the playback trials influenced males' reactions to the stimulus. Intuitively, crested macaque Boldness appears to be similar to the Confidence factor found in other macaques. This notion is however rather speculative and based on rating items used in questionnaire studies, such as fearful or timid , . At the same time, since we also consider our Aggressiveness factor as equivalent to rhesus and Barbary Confidence, we would have to expect a salient positive correlation between crested macaque Boldness and Aggressiveness, which was absent. For now, we hypothesize that Boldness makes up a unique factor in crested macaques, but cannot rule out that this is due to the different approaches with which we determined Boldness and Aggressiveness. Given the prevalence of Boldness in non-primate animals in general ,  and its hypothetical absence in primates (, but see ), future studies of primate personality should incorporate either items like bold into their questionnaires or try to assess Boldness experimentally . This should clarify in how far Boldness constitutes an independent factor of primate personality, or is just a facet of some other factor, such as Aggressiveness or Confidence.
In agreement with most studies of macaques, we found no evidence for the existence of a distinct Neophilia dimension (but see Openness with the items curious and inventive in rhesus macaques, ). Neither did we observe any significant correlation between the response to our novelty experiments and any of the other five personality factors. This absence is somewhat surprising, since a Neophilia/Exploration factor (or the human equivalent Openness) is present in many animal species including primates (, , , see also , ). However, the actual correlation coefficients of the playback response with Aggressiveness and Boldness were modest, but not negligible (−0.39 and 0.32), suggesting that the absence of significant relationships in our study might potentially have been a problem of statistical power or inappropriate stimulus choice . In line with this possible overlap with other factors, studies in other macaque species found rating items that describe Neophilia such as curious or exploratory load on diverse factors such as Sociability , , Friendliness  or Activity/Excitability , . An alternative explanation for the absence of Neophilia may be related to the modality in which we presented the stimulus. Macaques, as many primates, are vision-dominant, suggesting that an auditory stimulus may be less likely to elicit consistent responses in our experimental setup. More studies are certainly needed to confirm or reject the absence of Neophilia in macaques.
Based on our results, some tentative comparisons can also be made between macaque and human personality in order to investigate the evolutionary roots of the current human personality structure as posited in the Five Factor Model , . The combination of the two socio-positive factors we found in crested macaques (Connectedness and Sociability) appear to match human Extraversion, whereas Aggressiveness may reflect the human Agreeableness vs. antagonism axis . Anxiety and, to a lesser degree, Boldness fit the descriptors of human Neuroticism . In contrast, human Conscientiousness and Openness appear not to be reflected in crested macaque personality as distinct factors. Although crested macaque personality factors do not match perfectly with those of humans, our results shed important light onto the evolution of personality structures within the primate taxon.
Our attempt to directly link crested macaque with human personality structure is however a simplification of the hypothetical evolutionary scenario proposed to explain the evolution of human personality structure from a macaque-like ancestor . Here we will use two examples to illustrate how incorporating data from additional species may help to formulate alternative hypotheses to explain the evolutionary history of personality structures. In their personality phylogeny using rhesus macaques as an outgroup to hominids, Weiss and colleagues  suggest that Anxiety and Activity are two distinct ancestral factors. Our data and recent work on Barbary macaques, however, allow an alternative scenario. In both Barbary and crested macaques, a single factor describes a combination of Anxiety and Activity (labeled Anxiety in crested macaques (see above) and Activity/Excitability in Barbary macaques ). Given that Barbary macaques represent the sister taxon to all other extant macaque species , , it may thus be that such a broader, singular Anxiety/Activity factor is the actual ancestral state. In addition, the presence of a single Sociability factor in Barbary and rhesus macaques may also represent the ancestral macaque state, whereas the occurrence of the two distinct factors we identified in crested macaques (Sociability and Connectedness) most likely is a derived feature. We therefore speculate that a hypothetical ancestor to macaques (and hominids) was characterized by a single Anxiety/Activity factor and a single Sociability factor, similar to present-day Barbary macaques .
Based on this suggested proto-macaque, we can formulate alternative hypotheses of how personality factors evolved in higher lineages. For example, according to a recently proposed scenario, the hominid Extraversion factor evolved from a combination of a “pure” Activity factor and a Sociability factor . Our results, however, indicate the possibility that such a single Activity factor was not the ancestral state of catarrhine primates. Hence, in order to evolve Extraversion in the hominid lineage (c.f. ), a separation of our suggested ancestral Anxiety/Activity factor into two distinct factors needs to occur, after the phylogenetic split of hominids and macaques. With the currently available data, both scenarios are equally parsimonious and only future studies that contribute more data will help to fully understand the evolution of primate personality structures.
Whereas a thorough treatment of the possible evolution of primate personalities is well beyond the scope of this paper, we find overall support for the hypothesis of Weiss and colleagues  in that crested macaque personality structure resembles more other macaque personality structures than hominid personalities. The comparative approach is thus surely a promising one and the more data points we acquire the clearer the overall picture of personality evolution will become. In addition, more and more data are being generated concerning neurobiological and endocrinological mechanisms underlying personality variation (e.g., –). By integrating these approaches with ethological data and with the recent progress in behavior genetics (e.g., , ), new avenues will open to study the evolutionary paths and selective pressures leading to the striking variation in personality structures we observe across the animal kingdom.
Audio recording of the dog bark used as stimulus in the playback experiments.
Audio recording of the donkey bray used as stimulus in the playback experiments.
Example of an experimental trial in the dog condition. The speaker is placed left, from the viewer's perspective. Towards the end of the clip, an alarm call is audible given by an individual not visible in the frame.
We thank Julie Duboscq, Cédric Girard-Buttoz, Jérôme Micheletta, Keith Hodges, and Nadja Richter for insightful discussions. The editor and four anonymous reviewers provided helpful criticism on the manuscript. We gratefully acknowledge the permission of the Indonesian State Ministry of Research and Technology (RISTEK), the Directorate General of Forest Protection and Nature Conservation (PHKA) in Jakarta and the Department for the Conservation of Natural Resources (BKSDA) in Manado, particularly Pak Untung and Pak Yunus, to conduct this research in the Tangkoko-Batuangus Nature Reserve. We thank all the members of the Macaca Nigra Project for their support in the field and Friederike Schröder for coding the playback videos.
Conceived and designed the experiments: CN AW AE. Performed the experiments: CN. Analyzed the data: CN. Contributed reagents/materials/analysis tools: MA AW AE. Wrote the paper: CN AW AE. Field site maintenance and logistics: MA.
- 1. Gosling SD (2001) From mice to men: what can we learn about personality from animal research? Psychol Bull 127: 45–86 doi:–10.1037/0033–2909.127.1.45.
- 2. Gosling SD, John OP (1999) Personality dimensions in nonhuman animals a cross-species review. Curr Dir Psychol Sci 8: 69–75 doi:–10.1111/1467–8721.00017.
- 3. Réale D, Reader SM, Sol D, McDougall PT, Dingemanse NJ (2007) Integrating animal temperament within ecology and evolution. Biol Rev 82: 291–318 doi:–10.1111/j.1469–185X.2007.00010.x.
- 4. Weiss A, Adams MJ, Widdig A, Gerald MS (2011) Rhesus macaques (Macaca mulatta) as living fossils of hominoid personality and subjective well-being. J Comp Psychol 125: 72–83 doi:10.1037/a0021187.
- 5. Konečná M, Weiss A, Lhota S, Wallner B (2012) Personality in Barbary macaques (Macaca sylvanus): temporal stability and social rank. J Res Pers 46: 581–590 doi:10.1016/j.jrp.2012.06.004.
- 6. Stevenson-Hinde J, Zunz M (1978) Subjective assessment of individual rhesus monkeys. Primates 19: 473–482 doi:10.1007/BF02373309.
- 7. Seyfarth RM, Silk JB, Cheney DL (2012) Variation in personality and fitness in wild female baboons. P Natl Acad Sci USA 109: 16980–16985 doi:10.1073/pnas.1210780109.
- 8. Dammhahn M (2012) Are personality differences in a small iteroparous mammal maintained by a life-history trade-off? Proc R Soc B 279: 2645–2651 doi:10.1098/rspb.2012.0212.
- 9. Koski SE (2011) Social personality traits in chimpanzees: temporal stability and structure of behaviourally assessed personality traits in three captive populations. Behav Ecol Sociobiol 65: 2161–2174 doi:–––10.1007/s00265–011–1224–0.
- 10. Konečná M, Lhota S, Weiss A, Urbánek T, Adamová T, et al. (2008) Personality in free-ranging Hanuman langur (Semnopithecus entellus) males: subjective ratings and recorded behavior. J Comp Psychol 122: 379–389 doi:10.1037/a0012625.
- 11. Uher J, Asendorpf JB, Call J (2008) Personality in the behaviour of great apes: temporal stability, cross-situational consistency and coherence in response. Anim Behav 75: 99–112 doi:10.1016/j.anbehav.2007.04.018.
- 12. Koski SE (2011) How to measure animal personality and why does it matter? Integrating the psychological and biological approaches to animal personality. In: Inoue-Murayama M, Kawamura S, Weiss A, editors. From Genes to Animal Behavior. Tokyo: Springer. 115–136.
- 13. Dingemanse NJ, Both C, Drent PJ, Tinbergen JM (2004) Fitness consequences of avian personalities in a fluctuating environment. Proc R Soc B 271: 847–852 doi:10.1098/rspb.2004.2680.
- 14. Smith BR, Blumstein DT (2008) Fitness consequences of personality: a meta-analysis. Behav Ecol 19: 448–455 doi:10.1093/beheco/arm144.
- 15. Schuett W, Tregenza T, Dall SRX (2010) Sexual selection and animal personality. Biol Rev 85: 217–246 doi:–10.1111/j.1469–185X.2009.00101.x.
- 16. Freeman HD, Gosling SD (2010) Personality in nonhuman primates: a review and evaluation of past research. Am J Primatol 72: 653–671 doi:10.1002/ajp.20833.
- 17. Mitani JC, Call J, Kappeler PM, Palombit RA, Silk JB, editors (2012) The Evolution of Primate Societies. University Of Chicago Press. 744 p.
- 18. Uher J (2008) Comparative personality research: methodological approaches. Eur J Pers 22: 427–455 doi:10.1002/per.680.
- 19. King JE, Figueredo AJ (1997) The five-factor model plus dominance in chimpanzee personality. J Res Pers 31: 257–271 doi:10.1006/jrpe.1997.2179.
- 20. Capitanio JP (2004) Personality factors between and within species. In: Thierry B, Singh M, Kaumanns W, editors. Macaque Societies. Cambridge: Cambridge University Press. 13–37.
- 21. Thierry B, Iwaniuk AN, Pellis SM (2000) The influence of phylogeny on the social behaviour of macaques (Primates: Cercopithecidae, genus Macaca). Ethology 106: 713–728 doi:–10.1046/j.1439–0310.2000.00583.x.
- 22. Thierry B (2011) The macaques: a double-layered social organization. In: Campbell CJ, Fuentes A, MacKinnon KC, Bearder SK, Stumpf RM, editors. Primates in Perspective. Oxford: Oxford University Press. 229–241.
- 23. Thierry B (2007) Unity in diversity: lessons from macaque societies. Evol Anthropol 16: 224–238 doi:10.1002/evan.20147.
- 24. Thierry B (1990) Feedback loop between kinship and dominance: the macaque model. J Theor Biol 145: 511–522 doi:––10.1016/S0022–5193(05)80485–0.
- 25. De Waal FBM, Luttrell LM (1989) Toward a comparative socioecology of the genus Macaca: different dominance styles in rhesus and stumptail monkeys. Am J Primatol 19: 83–109 doi:10.1002/ajp.1350190203.
- 26. Thierry B (2000) Covariation of conflict management patterns across macaque species. In: Aureli F, de Waal FBM, editors. Natural Conflict Resolution. Berkely: University of California Press. 106–128.
- 27. Duboscq J, Micheletta J, Agil M, Hodges K, Thierry B, et al. (2013) Social tolerance in wild female crested macaques (Macaca nigra) in Tangkoko-Batuangus Nature Reserve, Sulawesi, Indonesia. Am J Primatol 75: 361–375 doi:10.1002/ajp.22114.
- 28. Berman CM, Thierry B (2010) Variation in kin bias: species differences and time constraints in macaques. Behaviour 147: 1863–1887 doi:10.1163/000579510X539691.
- 29. Bolig R, Price CS, O'Neill PL, Suomi SJ (1992) Subjective assessment of reactivity level and personality traits of rhesus monkeys. Int J Primatol 13: 287–306 doi:10.1007/BF02547817.
- 30. Capitanio JP (1999) Personality dimensions in adult male rhesus macaques: prediction of behaviors across time and situation. Am J Primatol 47: 299–320 doi:–;10.1002/(SICI)1098–2345(1999)47:4<299::AID-AJP3>3.0.CO;2-P.
- 31. Sussman AF, Ha JC, Bentson KL, Crockett CM (2013) Temperament in rhesus, long-tailed, and pigtailed macaques varies by species and sex. Am J Primatol 75: 303–313 doi:10.1002/ajp.22104.
- 32. Digman JM (1990) Personality structure: emergence of the five-factor model. Annu Rev Psychol 41: 417–440 doi:10.1146/annurev.ps.41.020190.002221.
- 33. John OP, Srivastava S (1999) The Big Five trait taxonomy: history, measurement, and theoretical perspectives. In: Pervin LA, John OP, editors. Handbook of Personality: Theory and Research. New York: Guilford Press. 102–138.
- 34. Weiss A, Gartner MC, Gold KC, Stoinski TS (2013) Extraversion predicts longer survival in gorillas: an 18-year longitudinal study. Proc R Soc B 280: 20122231 doi:10.1098/rspb.2012.2231.
- 35. Weiss A, King JE, Perkins L (2006) Personality and subjective well-being in orangutans (Pongo pygmaeus and Pongo abelii). J Pers Soc Psychol 90: 501–511 doi:–10.1037/0022–3522.214.171.1241.
- 36. Petit O, Abegg C, Thierry B (1997) A comparative study of aggression and conciliation in three cercopithecine monkeys (Macaca fuscata, Macaca nigra, Papio papio). Behaviour 134: 415–432 doi:10.1163/156853997X00610.
- 37. Sueur C, Petit O, De Marco A, Jacobs AT, Watanabe K, et al. (2011) A comparative network analysis of social style in macaques. Anim Behav 82: 845–852 doi:10.1016/j.anbehav.2011.07.020.
- 38. Supriatna J, Andayani N (2008) Macaca nigra. IUCN 2012 IUCN Red List of Threatened Species Version 20122 <www.iucnredlist.org>. Accessed 2013 Mar 27.
- 39. Neumann C, Assahad G, Hammerschmidt K, Perwitasari-Farajallah D, Engelhardt A (2010) Loud calls in male crested macaques, Macaca nigra: a signal of dominance in a tolerant species. Anim Behav 79: 187–193 doi:10.1016/j.anbehav.2009.10.026.
- 40. Altmann J (1974) Observational study of behavior: sampling methods. Behaviour 49: 227–266 doi:10.1163/156853974X00534.
- 41. Thierry B, Bynum EL, Baker SC, Kinnaird MF, Matsumura S, et al. (2000) The social repertoire of Sulawesi macaques. Primate Research 16: 203–226.
- 42. Martin PR, Bateson P (1993) Measuring Behaviour: An Introductory Guide. 2nd ed. Cambridge: Cambridge University Press.
- 43. Carter AJ, Marshall HH, Heinsohn R, Cowlishaw G (2012) How not to measure boldness: novel object and antipredator responses are not the same in wild baboons. Anim Behav 84: 603–609 doi:10.1016/j.anbehav.2012.06.015.
- 44. Micheletta J, Waller BM, Panggur MR, Neumann C, Duboscq J, et al. (2012) Social bonds affect anti-predator behaviour in a tolerant species of macaque, Macaca nigra. Proc R Soc B 279: 4042–4050 doi:10.1098/rspb.2012.1470.
- 45. Nakagawa S, Schielzeth H (2010) Repeatability for Gaussian and non-Gaussian data: a practical guide for biologists. Biol Rev 85: 935–956 doi:–×10.1111/j.1469–185×.2010.00141.x.
- 46. Lessells CM, Boag PT (1987) Unrepeatable repeatabilities: a common mistake. Auk 104: 116–121 doi:10.2307/4087240.
- 47. Neumann C, Duboscq J, Dubuc C, Ginting A, Irwan AM, et al. (2011) Assessing dominance hierarchies: validation and advantages of progressive evaluation with Elo-rating. Anim Behav 82: 911–921 doi:10.1016/j.anbehav.2011.07.016.
- 48. Cheney DL (1992) Intragroup cohesion and intergroup hostility: the relation between grooming distributions and intergroup competition among female primates. Behav Ecol 3: 334–345 doi:10.1093/beheco/3.4.334.
- 49. Schuett W, Dall SRX (2009) Sex differences, social context and personality in zebra finches, Taeniopygia guttata. Anim Behav 77: 1041–1050 doi:10.1016/j.anbehav.2008.12.024.
- 50. David M, Auclair Y, Cézilly F (2012) Assessing short- and long-term repeatability and stability of personality in captive zebra finches using longitudinal data. Ethology 118: 932–942 doi:–10.1111/j.1439–0310.2012.02085.x.
- 51. Dammhahn M, Almeling L (2012) Is risk taking during foraging a personality trait? A field test for cross-context consistency in boldness. Anim Behav 84: 1131–1139 doi:10.1016/j.anbehav.2012.08.014.
- 52. Field A, Miles J, Field Z (2012) Discovering Statistics Using R. 1st ed. London: Sage Publications. 992 p.
- 53. Budaev SV (2010) Using principal components and factor analysis in animal behaviour research: caveats and guidelines. Ethology 116: 472–480 doi:–10.1111/j.1439–0310.2010.01758.x.
- 54. Maciej P, Patzelt A, Ndao I, Hammerschmidt K, Fischer J (2013) Social monitoring in a multilevel society: a playback study with male Guinea baboons. Behav Ecol Sociobiol 67: 61–68 doi:–––10.1007/s00265–012–1425–1.
- 55. Benjamini Y, Hochberg Y (1995) Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc B Met 57: 289–300 doi:10.2307/2346101.
- 56. R Core Team (2012) R: A Language and Environment for Statistical Computing. Vienna, Austria. Available: http://www.R-project.org/. Accessed 2012 Sep 10.
- 57. Revelle W (2012) psych: Procedures for Psychological, Psychometric, and Personality Research. Available: http://personality-project.org/r. Accessed 2012 Sep 10.
- 58. Schielzeth H, Nakagawa S (2011) rptR: Repeatability for Gaussian and non-Gaussian data. Available: http://R-Forge.R-project.org/projects/rptr/. Accessed 2012 Sep 10.
- 59. Maestripieri D, Schino G, Aureli F, Troisi A (1992) A modest proposal: displacement activities as an indicator of emotions in primates. Anim Behav 44: 967–979 doi:––10.1016/S0003–3472(05)80592–5.
- 60. Evers E, de Vries H, Spruijt BM, Sterck EHM (2011) Better safe than sorry – socio-spatial group structure emerges from individual variation in fleeing, avoidance or velocity in an agent-based model. PLoS ONE 6: e26189 doi:10.1371/journal.pone.0026189.
- 61. Majolo B, McFarland R, Young C, Qarro M (forthcoming) The effect of climatic factors on the activity budgets of Barbary macaques (Macaca sylvanus). Int J Primatol. Available: http://link.springer.com/article/10.1007/s10764-013-9678-8. Accessed 28 April 2013.
- 62. Dingemanse NJ, Kazem AJN, Réale D, Wright J (2010) Behavioural reaction norms: animal personality meets individual plasticity. Trends Ecol Evol 25: 81–89 doi:10.1016/j.tree.2009.07.013.
- 63. Mondragón-Ceballos R, Santillán-Doherty AM (1994) The relationship between personality and age, sex and rank in captive stumptail macaques. In: Roeder J-J, Thierry B, Anderson JR, Herrenschmidt N, editors. Current Primatology, vol. II: Social Development, Learning and Behavior. Strasbourg: Universite Loius Pasteur. 241–250.
- 64. Caine NG, Earle H, Reite M (1983) Personality traits of adolescent pig-tailed monkeys (Macaca nemestrina): an analysis of social rank and early separation experience. Am J Primatol 4: 253–260 doi:10.1002/ajp.1350040304.
- 65. Drews C (1993) The concept and definition of dominance in animal behaviour. Behaviour 125: 283–313 doi:10.1163/156853993X00290.
- 66. Micheletta J, Waller BM (2012) Friendship affects gaze following in a tolerant species of macaque, Macaca nigra. Anim Behav 83: 459–467 doi:10.1016/j.anbehav.2011.11.018.
- 67. McDonald DB (2007) Predicting fate from early connectivity in a social network. P Natl Acad Sci USA 104: 10910–10914 doi:10.1073/pnas.0701159104.
- 68. Wilson ADM, Krause S, Dingemanse NJ, Krause J (2013) Network position: a key component in the characterization of social personality types. Behav Ecol Sociobiol 67: 163–173 doi:––10.1007/s00265–012–1428-y.
- 69. Kulik L, Muniz L, Mundry R, Widdig A (2012) Patterns of interventions and the effect of coalitions and sociality on male fitness. Mol Ecol 21: 699–714 doi:–10.1111/j.1365–294X.2011.05250.x.
- 70. Sih A, Hanser SF, McHugh KA (2009) Social network theory: new insights and issues for behavioral ecologists. Behav Ecol Sociobiol 63: 975–988 doi:–––10.1007/s00265–009–0725–6.
- 71. McGuire MT, Raleigh MJ, Pollack DB (1994) Personality features in vervet monkeys: the effects of sex, age, social status, and group composition. Am J Primatol 33: 1–13 doi:10.1002/ajp.1350330102.
- 72. Drea CM (1998) Social context affects how rhesus monkeys explore their environment. Am J Primatol 44: 205–214 doi:–;#10.1002/(SICI)1098–2345(1998)44:3<205::AID-AJP3>3.0.CO;2-#.
- 73. Bergman TJ, Kitchen DM (2009) Comparing responses to novel objects in wild baboons (Papio ursinus) and geladas (Theropithecus gelada). Anim Cogn 12: 63–73 doi:–––10.1007/s10071–008–0171–2.
- 74. Li J, Han K, Xing J, Kim H-S, Rogers J, et al. (2009) Phylogeny of the macaques (Cercopithecidae: Macaca) based on Alu elements. Gene 448: 242–249 doi:10.1016/j.gene.2009.05.013.
- 75. Morales JC, Melnick DJ (1998) Phylogenetic relationships of the macaques (Cercopithecidae: Macaca), as revealed by high resolution restriction site mapping of mitochondrial ribosomal genes. J Hum Evol 34: 1–23 doi:10.1006/jhev.1997.0171.
- 76. Anestis SF (2011) Primate personality and behavioral endocrinology. In: Weiss A, King JE, Murray L, editors. Personality and Temperament in Nonhuman Primates. New York: Springer. 169–192. Available: http://link.springer.com/chapter/10.1007/978-1-4614-0176-6_7.
- 77. Young LJ, Wang Z (2004) The neurobiology of pair bonding. Nat Neurosci 7: 1048–1054 doi:10.1038/nn1327.
- 78. Yokoyama C, Onoe H (2011) Molecular brain imaging of personality traits in nonhuman primates: a study of the common marmoset. In: Inoue-Murayama M, Kawamura S, Weiss A, editors. From Genes to Animal Behavior. Tokyo: Springer. 389–406. Available: http://link.springer.com/chapter/10.1007/978-4-431-53892-9_19.
- 79. Higley JD, Suomi SJ, Chaffin AC (2011) Impulsivity and aggression as personality traits in nonhuman primates. In: Weiss A, King JE, Murray L, editors. Personality and Temperament in Nonhuman Primates. New York: Springer. 257–283. Available: http://link.springer.com/chapter/10.1007/978-1-4614-0176-6_10.
- 80. Adams MJ (2011) Evolutionary genetics of personality in nonhuman primates. In: Inoue-Murayama M, Kawamura S, Weiss A, editors. From Genes to Animal Behavior. Tokyo: Springer. 137–164. Available: http://link.springer.com/chapter/10.1007/978-4-431-53892-9_6.