Information

Do we have evidence of heritable personality traits in chimpanzees?


I have read that chimpanizees have distinct personalities. Do we have evidence of heritable personality traits in them?


There has been some research in order to explore this direction, for instance, see this reference:

A. Weiss et al. - "The Heritability of Personality Factors in Chimpanzees", Behavior Genetics, Vol. 30, No. 3, 2000.,

in which the authors study five human-like factors - Surgency, Dependability, Emotional Stability, Agreeableness, and Openness.

Quoting verbatim their findings:

Only Dominance… was significantly heritable. Dependability showed evidence that it might be heritable, but the parameter estimates were not statistically significant… Two other personality traits, Agreeableness and Emotional Stability, showed positive heritability estimates but the parameter estimates were nonsignificant and small. Openness and Surgency showed no evidence of heritability.

In fact, the statistical correlation between dominance and heritability makes a strong sense. Here's what I make out of this - this was a correlation based study, or kind of an experiment, hence the stress on statistical significance. But the question it raises is - is there a causal relationship between the two, i.e. does there exist any proper mechanism for the inheritance of "dominance". Finding such a mechanism was clearly beyond the scope of this study, but I believe the requisite answer for your question is if one stumbles on to such mechanisms, and not statistical correlations, which are mere pointers.

But on the basis of the study at least, all that one can safely conclude is - statistical correlations suggest that inheritance of "dominance" appears plausible, but the exact mechanism of how this implements out in practice, is beyond the scope of this study.


Personality Traits in Rhesus Macaques (Macaca mulatta) Are Heritable but Do Not Predict Reproductive Output

There is growing evidence that behavioral tendencies, or “personalities,” in animals are an important aspect of their biology, yet their evolutionary basis is poorly understood. Specifically, how individual variation in personality arises and is subsequently maintained by selection remains unclear. To address this gap, studies of personality require explicit incorporation of genetic information. Here, we explored the genetic basis of personality in rhesus macaques by determining the heritability of personality components and by examining the fitness consequences of those components. We collected observational data for 108 adult females living in three social groups in a free-ranging population via focal animal sampling. We applied principal component analysis to nine spontaneously occurring behaviors and identified six putative personality components, which we named Meek, Bold, Aggressive, Passive, Loner, and Nervous. All components were repeatable and heritable, with heritability estimates ranging from 0.14 to 0.35. We found no evidence of an association with reproductive output, measured either by infant survival or by interbirth interval, for any of the personality components. This finding suggests either that personality does not have fitness-related consequences in this population or that selection has acted to reduce fitness-associated variation in personality.

This is a preview of subscription content, access via your institution.


Why chimpanzees are great for research

Psychologists have been doing a lot of research into psychopathy — what makes a psychopath, whether there are biological markers for the disorder, or whether it can be prevented or cured.

There are always limitations with the results from animal models being translated to humans, so I asked Latzman why he chose to work with chimpanzees at all.

"You can investigate behavioural or emotional processes that are not just similar but parallel to processes seen in humans, and then also consider neurological or biological processes that are also parallel between humans and chimpanzees," Latzman explained. "Chimpanzee models further allow us to largely remove socio-cultural factors from the equation resulting in accounts more closely rooted within biology."

In other words, they display behavioural issues in their purest form because they don't have to bother with the social pressures of humans. Chimpanzees are never taught they have to share, they don't have to be nice to each other, and they don't have to say sorry or wait for their turn.

In humans, any number of different social factors could affect the outcome of someone's personality. In chimpanzees, all these outside forces can be monitored.


Methods

Site and communities

Gombe National Park consists of a narrow, 35 km 2 strip of mountainous terrain along the eastern edge of Lake Tanganyika in Tanzania. It contains a complex mosaic of habitat transitions from riverine forest in the valleys to deciduous woodland and grassland on the ridges 37 . Gombe contains three communities of chimpanzees (from north to south): Mitumba, Kasekela and Kalande (Fig. 1). Goodall began habituating the Kasekela community to the presence of human researchers in 1960 (ref. 3), and demographic records have been continuously kept ever since. The chimpanzees were provisioned with bananas at an artificial feeding station from 1963 to 2000, with daily records made of their behavior 3,38 . Since the early 1970s, Tanzanian field assistants have conducted almost-daily dawn-to-dusk focal follows 39 of chimpanzees as they travel throughout their range. During these focal follows, the field assistants systematically record changes in party composition and location and make a narrative record of the behavior of the focal chimpanzee and events in the group 3,38 . In each focal follow, the field assistants focused attention on a single individual, observing all of its behavior: feeding, fighting, mating, parenting, grooming, tool use, hunting, and so forth. Focal targets varied daily, rotating through most of the chimpanzees in each community each month, providing the field assistants opportunities to get to know many chimpanzees in great detail. Between 1966 and 2014 the Kasekela community varied in size from 38 to 63 individuals. The Kahama community consisted of at least 15 individuals (and possibly some additional unhabituated females) that split from the Kasekela community in the early 1970s. Kasekela chimpanzees killed at least six individuals from Kahama in a series of violent attacks, and the Kahama community ceased to exist by 1977 (ref. 3).

The ranges of the Kaskela and Mitumba communities cover the years 2000 to 2009. Earlier ranges are described in Williams et al. 82 and Wilson et al. 83

Efforts to habituate the Mitumba community started in 1985. Field assistants regularly provisioned Mitumba chimpanzees at an artificial feeding area from 1992–2000, and collected demographic and behavioral data. From the mid-1990s, field assistants conducted focal follows using methods identical to those used for the Kasekela community. Almost all members of the Mitumba community were habituated by 1994 (ref. 40). Between 1994 and 2014 the community varied in size from 20 to 28 individuals.

Chimpanzees

We initially collected 494 personality ratings that described 141 eastern chimpanzees who lived or were living in Gombe National Park from ca. 1963 to the present. We excluded ratings of chimpanzees that were likely to be less informative or valid. The remaining data consisted of chimpanzees who (a) survived to be at least five years old, (b) had been observed for at least two years, (c) had been sighted on at least 100 days over the qualifying period, i.e., when that chimpanzee was a regular member of the community, and, because the first field assistants were hired in late 1969 and the early to mid-1970s, (d) had been alive and observed after 1970. The excluded chimpanzees included one chimpanzee from Mitumba who did not survive to be at least five years old (criterion a) and nine chimpanzees from Kasekela who had not been sighted on at least 100 days (criterion c). In addition, a later inspection of records after data had been collected revealed that three chimpanzees from Kasekela who had been rated may have been confused with other chimpanzees, and so we excluded ratings of these chimpanzees as well.

After excluding the 34 ratings of these 13 individuals, the data set consisted of 460 personality ratings of 128 chimpanzees. This data set made up our working sample and, unless noted, will be described throughout. This sample comprised 56 males and 72 females who either belonged to the Kasekela community (46 males and 58 females including 6 males and 6 females that fissioned into the Kahama community), the Mitumba community (10 males and 11 females), or who had been members of both communities (3 females). Sixty-three of these individuals (32 males and 31 females) were alive on 11 October 2010, the date we began to collect ratings.

Ratings

Questionnaire

Personality ratings were made on a modified form of the Hominoid Personality Questionnaire (HPQ) 41 . A full history of the development of the HPQ is available elsewhere 42 a brief history follows. The original HPQ consists of 54 items and was an expanded version of the 48-item Orangutan Personality Questionnaire 43 , which was, in turn, an expanded version of the 43-item Chimpanzee Personality Questionnaire 44 . Of the 43 Chimpanzee Personality Questionnaire items, 41 were sampled from a taxonomy of the five human personality dimensions 45 . The two additional items, ‘clumsy’ and ‘autistic,’ were devised by King and Figueredo, the authors of the Chimpanzee Personality Questionnaire 44 . Of the additional items in the full HPQ, five were chosen to increase the number of traits related to Openness, three were chosen to increase the number of traits related to Neuroticism, and three were chosen to increase the number of items related to Conscientiousness.

As noted in papers describing the development of the HPQ 41,43,44 , only items considered applicable to chimpanzees and other nonhuman primates were sampled. Each HPQ item comprises a trait adjective paired with one to three sentences that clarified the trait adjective’s meaning in the context of chimpanzee behavior. For example, the clarifying sentence for the item ‘helpful’ was ‘Subject is willing to assist, accommodate, or cooperate with other chimpanzees.’ The HPQ also includes printed instructions that asked raters to rate each item on a scale ranging from 1 (‘Displays either total absence or negligible amounts of the trait’) to 7 (‘Displays extremely large amounts of the trait’), to base judgments on their understanding of typical chimpanzee behavior, and not to discuss their ratings with others.

Two important things to note about the HPQ are: (a) although the items included representatives of traits associated with the well-established five human personality dimensions—Neuroticism, Extraversion, Openness to Experience, Agreeableness, and Conscientiousness—the questions were adapted for use in nonhuman primates and (b) traits across the five human dimensions were chosen not to try and impose a human personality structure onto chimpanzee or primates, but to ensure that a wide range of traits related to behavior, affect, and cognition have been sampled 41,43,44 . Therefore, although these items may be similar in humans and other primates, they are not identical, and ratings of primates on this questionnaire do not necessarily result in dimensions corresponding to human personality dimensions. For example, a study of rhesus macaques found six personality dimensions—Confidence, Openness, Dominance, Friendliness, Activity, and Anxiety 10 —three of which resembled those found in earlier studies of this species, even though those earlier studies used different instruments to assess personality 46 .

The HPQ and its predecessors have predominated in research on the personalities of chimpanzees and the other great apes 18,41,43,44,47,48 . The Chimpanzee Personality Questionnaire was used to assess the personalities of 100 zoo-housed chimpanzees, and ratings revealed the presence of five personality dimensions—Extraversion, Conscientiousness, Agreeableness, Neuroticism, and Openness—similar to those found in humans, and an additional chimpanzee-specific dimension labeled ‘Dominance’ 44 . The interrater reliabilities of these dimensions were on par with interrater reliabilities derived in studies of humans including self- and spouse-ratings and studies in which individuals were rated by peers 44 . Later studies of chimpanzees using the Chimpanzee Personality Questionnaire or the HPQ demonstrated that these ratings are repeatable 49 , associated with behavior 50 , and linked to genetic variation 51–53 , and neurobiology 54 . Moreover, although the HPQ originated in zoo research and in an English-speaking country, the HPQ and related instruments have been used successfully for measuring chimpanzee personality in other contexts and/or in other languages. Studies of chimpanzees in a naturalistic sanctuary in the Republic of Congo using the English-language version and a French translation of the Chimpanzee Personality Questionnaire 55 Yerkes National Primate Research Center using the English language version of the Chimpanzee Personality Questionnaire 47 and zoos, a sanctuary, and two research institutes in Japan using a Japanese translation of the HPQ 41 all revealed personality dimensions similar to those identified in King and Figueredo’s original study of zoo chimpanzees. The interrater reliabilities observed in these studies were also consistent with those found in King and Figueredo’s original study.

For the present study, we made three modifications to the HPQ. The first modification was to change references to zoo-specific environmental features in clarifying sentences, e.g., ‘the enclosure,’ to neutral phrases, e.g., ‘the environment’. The second modification was to translate the HPQ items into Swahili, the national language of Tanzania (most raters were native speakers of the local language, Kiha, but all were fluent in Swahili, the working language at Gombe). The translation was done by Munira Massoud, a Tanzanian student at the University of Minnesota, fluent in both Swahili and English. The translations were then cross-checked by specialists in chimpanzee research who were competent in both English and Swahili (SK and MLW). In addition, we noticed that two items (‘innovative’ and ‘inventive’) had the same adjective label in Swahili (‘mbunifu’), although the behavioral descriptors differed. With the help a co-author (DAC) who was fluent in English and Swahili, we changed the adjective label to ‘mtengenezaji’. Because the correlation between the items ‘innovative’ and ‘inventive’ was higher in the Mitumba sample (r=0.53, P=0.014), where they had different adjective labels, than in the Kaskela sample (r=0.22, P=0.022), there is no evidence that the responses on these items by raters of the Kasekela chimpanzees were biased because these items had the same label. The third modification was to reduce the length of the questionnaire to 24 items. Although reducing the number of items risked reducing the reliability of scales 56 , we considered this risk acceptable, because we considered it more important to obtain ratings on as many chimpanzees as possible, and we were concerned that including too many items would make the raters’ response burdens excessive. In fact, collecting data on this sample using the 24-item version of the questionnaire took nearly two months.

When choosing items to include in the shorter version of the Swahili-language HPQ, we sought to meet four criteria. The first was that for each of the six dimensions identified in previous studies 41,44 we should, if possible, have four items, and wherever possible, have an even number of items that were positively and negatively associated with those dimensions. The second was to maximize the item overlap with studies that have used earlier versions of the questionnaire, e.g., the Chimpanzee Personality Questionnaire. To these ends we decided that each dimension should be represented by at most 1 of the 11 items that were in the HPQ, but that were not in the Chimpanzee Personality Questionnaire. We met these first two criteria for Dominance, Extraversion, Conscientiousness, Agreeableness, and Neuroticism. However, for Openness, we had to select two Openness items from the HPQ, because only two of the Chimpanzee Personality Questionnaire items were related to Openness. Our third criterion was that the items that we chose had displayed high interrater reliabilities in previous studies that used the Chimpanzee Personality Questionnaire or HPQ to rate chimpanzees. The fourth criterion was to choose items that were strongly related to the dimensions, i.e., had high absolute loadings on these dimensions in previous studies.

The full length English language version of the Hominoid Personality Questionnaire for chimpanzees is provided in Supplementary File 1, the full-length Swahili version in Supplementary File 2, and the brief, 24-item Swahili version in Supplementary File 3.

Raters

This study benefits from the experience of an extraordinary group of individuals, the Tanzanian field assistants at Gombe. These men spent up to 35 years observing chimpanzees in the wild, which is more time than anyone else in history with the possible exception of local field assistants working at other long-term studies such as Mahale 57 . Chimpanzees can live well into their 50s in the wild, but at Gombe, most die before they are 40 (ref. 58). The most experienced field assistants thus observed many chimpanzees for their entire lives, from birth to death. These field assistants have limited formal education, with little or no understanding of English, the language in which most of the technical scientific literature is written. Nonetheless, nobody knows these individual chimpanzees better than they do.

Eighteen of these field assistants who had extensive daily experience in observing and tracking the chimpanzees and recording their behavior 3,38 rated the chimpanzees. Each field assistant rated between 21 and 43 chimpanzees (mean subjects per rater=27.4, s.d.=5.6). One field assistant accidentally rated one chimpanzee a second time three days later in this case, the correlation between the ratings across the 24 items was 0.66. Each chimpanzee was rated by two to nine field assistants (mean raters per subject=3.5, s.d.=1.3).

Of these 18 field assistants, 10 were still working with the chimpanzee community that they rated, two had traded work stations (Q moved to Kasekela and O moved to Mitumba), and so were asked to rate the chimpanzees from the community where they had worked longest (Mitumba for Q, and Kasekela for O), four were retired, and two were no longer employed. The 12 field assistants who were still employed had been working for a mean of 16.7 years at the time they rated the chimpanzees and knew the chimpanzees they rated for 1 to 25 years (mean=10.7, s.d.=5.3). The six field assistants who had retired or who were no longer employed had worked for a mean of 28.3 years at the time they rated the chimpanzees and had known the chimpanzees they rated for less than one year to 35 years (mean=11.4, s.d.=7.7).

The starting dates of employment ranged from 1969 to 2003 (median=1987.5). The eight field assistants who were no longer working with the chimpanzee community that they rated ended their work with that community from one to five years before doing the ratings (median=4.5). This meant that in many cases, ratings of the chimpanzees by different raters reflect the same chimpanzee over different parts of that chimpanzee’s life. Therefore, instead of providing a single age for each chimpanzee in this dataset, we determined, for each chimpanzee-rater pairing, the median age of the chimpanzee. For further details on which chimpanzees were rated by which field assistants and the employment dates of field assistants for the Kasekela and Mitumba communities, see Supplementary Tables 1 and 2, respectively.

Procedure

Ratings were conducted when field assistants were not watching or interacting with chimpanzees. Most ratings of the Kasekela chimpanzees were conducted at the Jane Goodall Institute’s Education Centre in Kigoma, Tanzania. Up to two field assistants completed questionnaires at any given time while sitting at separate desks on opposite sides of a room. A few field assistants completed some of their questionnaires at home. Most ratings of the Mitumba chimpanzees were conducted in Gombe National Park, either in a room with two desks or at the field assistant’s home. During ratings of the Kasekela chimpanzees and the Mitumba chimpanzees, an interpreter was on hand to answer questions about the questionnaire or about a specific item or set of items. Most of the field assistants completed the questionnaires on their own. However, in a few cases, mostly confined to some of the older, retired field assistants, the interpreter read each question to the field assistant and recorded his answer.

We covered the field assistants’ travel expenses, meals, and, when necessary, their accommodation costs. In addition, field assistants who were still employed were paid 2,000 Tanzanian shillings (

$1.26 in 2011) for each chimpanzee they rated and retired field assistants and field assistants who were no longer employed were paid 4,000 Tanzanian shillings (

$2.52 in 2011) for each chimpanzee they rated. Thus, for each chimpanzee rated, field assistants who were still employed earned

3.6 times the median net hourly wage in 2012 in Tanzania (550 Tanzanian shillings) 59 and field assistants who were retired or no longer working earned

7.3 times the median hourly wage.

Missing data

Of the 11,040 possible responses (24 items for 460 ratings), 32 were missing. The missing responses came from ratings of 23 chimpanzees by 11 field assistants. In 21 cases (the rating of one chimpanzee by one rater) one item was not completed, in three cases two items were not completed, and in one case, five items were not completed. For the purposes of our analyses, as in similar studies of wild nonhuman primates 4,10,18 , we substituted missing item data with the mean for that item.

Data aggregation

For the purposes of analyses other than obtaining interrater reliabilities, including data reduction and examining correlations between these ratings and the earlier ratings on the EPI 7 , we calculated the mean personality rating across raters for each personality variable for each chimpanzee.

Factor scores

Based on the results of previous studies 41,44 and using items that we found to be reliable (see the ‘Interrater reliabilities of items’ section in Technical Validation), we created six personality dimension variables to represent each of the six chimpanzee personality dimensions. We created these scores for the raw ratings (the non-aggregated data) and for the ratings that had been averaged across raters (the aggregated data). In both the non-aggregated and aggregated data, missing ratings were substituted with the mean of that rating.


Behaviour, Management and Conservation

Primates are the most studied animals in zoos (Melfi, 2009 ). The largest focus of this research in British and Irish zoos is on social behaviour, while in American zoos the focus is on reproduction and endocrinology (Melfi, 2005 ). In Europe, the United States and Japan there have been a large number of studies on primate cognition (Hopper, 2017 ). However, very little research is carried out on personality, and even less on the link between personality and behaviour (Stoinski et al., 2004 ). Moreover, methodological studies, such as validating personality-trait ratings with behavioural observations, are often the focus of this work. The validity of one type of assessment versus another has been covered elsewhere, with trait ratings shown to be just as, if not more, reliable than behavioural observations (e.g. Vazire et al., 2007 ) therefore, links between the two have been found and tend to be evident. For example, gorilla Dominance is related to received and given displacement (negative and positive, respectively) (Kuhar et al., 2006 Schaefer & Steklis, 2014 ) and Extraversion/Agreeableness is related to affiliative, grooming, playing, approach and touching behaviours (Schaefer & Steklis, 2014 ). Similarly, Chimpanzee Extraversion is related to affiliative behaviour (Pederson et al., 2005 Massen & Koski, 2014 ). A relationship has also been shown between Chimpanzee Dominance and agonistic behaviours (a positive relationship) and submissive behaviours (a negative relationship) (Freeman et al., 2013 ). In the same study a relationship was also found between Extraversion and contact aggression. This type of research could have a direct impact on management decisions. For example, gorillas housed in solitary situations rated lower on the Understanding factor (Kuhar et al., 2006 ). Although this result needs to be validated, if it were the case, and Understanding was predictive of a need for solitary housing, planning for housing provision could be made easier, both within single zoos and across zoos. This could lead to a reduction of aggressive encounters, because gorillas rated lower for Understanding were also more likely to engage in non-contact aggression (Kuhar et al., 2006 ). It has been suggested that gorilla groups should be formed when the apes are young for the best outcomes (Stoinski et al., 2004 ), so it is vital to understand any age-related differences in personality, but this is not always possible in zoos (see Huskisson & Chism, 2018 ). Knowing that silverback gorillas rated highly on the personality dimension Understanding are more accepting of new members into their troop (Stoinski et al., 2004 ) could be a vital piece of information that facilitates successful groupings. Longitudinal studies are particularly important for assessing any links between personality and age. For example, in Chimpanzees, Extraversion and Openness decline with age, while Agreeableness and Conscientiousness increase (King, J. E., et al., 2008 ). In orangutans, Agreeableness declines with age, and while Extraversion also decreases as orangutans age, it does so differently from Chimpanzees (Weiss & King, 2015 ). Zoos are often in the position of having to move individuals within groups of animals, which can pose risks to both the individuals involved and their caretakers (Powell, 2010 ). Personality is another tool zoos can use to decrease risk and increase success when making decisions about forming new groups of great apes.

Personality can also affect other areas of animal management for example, personality studies have shown that captive breeding can be facilitated in a number of ways. In felids, providing secluded enclosures and ample hiding spaces for animals that rate higher on tense or fearful personality factors can improve breeding success, as was found with Cheetahs Acinonyx jubatus (Wielebnowski, 1999 ). Similarly, making species-specific changes in enclosures and husbandry to align with personality differences has been shown to benefit breeding in Giant pandas Ailuropoda melanoleuca (Powell & Svoke, 2008 ). Using measurements such as faecal corticosteroids and behavioural observations can establish parameters for well-being that facilitate improved reproductive success, as has been found in Clouded leopards Neofelis nebulosa (Wielebnowski et al., 2002 ).

Primate behaviour in zoos in response to exposure to humans, typically unknown visitors rather than familiar keepers, may also be impacted by personality. For example, solitary, irritable and aggressive Diana monkeys Cercopithecus diana diana demonstrate increased abnormal behaviour during peak visitor times, while active, playful and excitable individuals show species-typical behaviour, including play (Barlow et al., 2007 ). For zoos, this information could be vital not only to animal welfare but also for visitor satisfaction, as visitors tend to prefer naturalistic exhibits that promote activity (Ryan & Saward, 2004 Fernandez et al., 2009 ). Gorillas show more relaxed behaviours during low visitor density, and higher rates of aggression, auto-grooming and stereotypies during high visitor density [Wells, 2005 see Ross et al. ( 2007 ) and Wells ( 2007 ) for discussion of these results]. Changes in group structure, such as the birth of an infant, may also have an effect on how gorillas react to visitors (e.g. Collins & Marples, 2016 ). Visitor density is not a stressor for all individuals (Hosey, 2000 ), and crowd size may or may not have an effect on behaviour in various species in zoos (Ross et al., 2007 Bonnie et al., 2016 Jones et al., 2016 Martin, R. A., & Melfi, 2016 Polgár et al., 2017 ). Using available knowledge about individual personality to schedule exposure to visitors could increase welfare and decrease unwanted behaviours in primates in zoos.

Enrichment can be used to challenge animals mentally and physically (Shepherdson, 1998 ), and can be conceived of on both the species and individual level (Mellen & Sevenich MacPhee, 2001 ). While little direct research has been carried out on personality and welfare, one study with Chimpanzees found that welfare measures correlate with subjective well-being measures, and that higher Extraversion and lower Neuroticism were related to higher ratings on the combined welfare/subjective well-being measure (Robinson et al., 2017 ). Understanding an individual's personality can help keepers to create a proactive (rather than reactive) environment for zoo animals (Mellen & Sevenich MacPhee, 2001 ).

Great ape cognition and its relationship to personality is a growing field of study. Chimpanzees rated higher on Openness are more willing to participate in research studies (Herrelko et al., 2012 ). Furthermore, those rated higher on Neuroticism performed more self-directed behaviours during the study (Herrelko et al., 2012 ). This has implications for animal management, where animals rated higher on Openness could be chosen first for training, which may provide the more reluctant animals with positive role models.

While personality has the potential to predict behaviour, it is possibly also influenced by environment. Different types of rearing, including group size, group composition and experience of other infants within the group, may all have a different effect upon individuals. For instance, Chimpanzees that are reared in larger groups (≥ 7 individuals) are rated more highly on positive personality traits (e.g. playful) than those reared in smaller groups (≤ 3 individuals), but also are rated as more Irritable and Excitable/Slow (Murray, 1998 ). However, a later study by Martin, J. E. ( 2005 ) found no such associations, so more work on this subject is needed.

Finally, zoos are one of the largest supporters of conservation, both financially and with expertise (Barongi et al., 2015 ). Personality can affect strategies for survival (Watters & Meehan, 2007 ), as well as reintroduction efforts (Bremner-Harrison et al., 2004 ). Although relatively few zoo animals are reintroduced to the wild, some projects have been initiated and personality should be a consideration (e.g. Cocks & Bullo, 2008 King, T., & Courage, 2008 ). Some of these projects have brought back species from near extinction (e.g. Black-footed ferret Mustela nigripes, California condor Gymnogyps californianus, Golden lion tamarin Leontopithecus rosalia, the Karner Blue butterfly Lycaeides melissa samuelis, Oregon spotted frog Rana pretiosa, Palila Loxioides bailleui, Red wolf Canis rufus and Wyoming toad Anaxyrus baxteri) (see examples in Dobson & Lyles, 2000 Walters et al., 2010 ). Studies carried out in zoos may inform work that is carried out in the wild and vice versa. For example, one study found a similar personality structure in zoo and sanctuary Chimpanzees (King, J. E., et al., 2005 ), indicating that captivity may not affect the way that personality traits cluster together. In that study, the Chimpanzees at the sanctuary (wild born and rescued from a variety of non-normal situations) were being conditioned for reintroduction into the wild.

Studies carried out at zoos could inform managers of how animals should be grouped for release, how they may react to release and in what stages release should occur. By using zoological records, such as information on pedigree, researchers can take this one step further to look at the genetics of natural selection and adaptation in captivity (Pelletier et al., 2009 ). This could help zoos to limit contemporary evolution in captivity, which may affect future reintroductions to the wild, as well as such health outcomes as breeding success (Pelletier et al., 2009 ).

The results of research into personality and behaviour of wild animals in zoos is promising, and should be encouraged. If personality allows for the prediction of behaviour, then zoo-animal management can be improved once such knowledge is available. The types of studies described above could be invaluable in animal management, as personality assessment would take some of the guesswork out of management decisions, which has the potential to increase psychological and physical welfare for animals in zoos.

These examples reveal some possibilities of the implications for animal management and, although zoos already play a role in learning about animal behaviour, there is still a wealth of work that could be done. For example, studies could be carried out in zoos to investigate how a better understanding of individual personality could make it possible to predict behaviour in primates. Research on human personality and behaviour is vast and varied, and can help zoos to begin to understand and predict how the animals in their care may react individually to different situations, as mentioned above. However, adaptations to study design are vital. Researchers focusing on humans often have access to larger sample sizes, allowing for generalizations about populations however, researchers in zoos frequently do not have large numbers of animals to study. For example, for great apes in a single zoo there are usually only a small number of individuals and they are often housed together. This leads to numerous studies of single zoos, which are valuable but not definitive or generalizable, narrowing their usefulness. To counteract these limitations, many researchers recommend multi-zoo studies (Swaisgood & Shepherdson, 2005 Whitham & Wielebnowski, 2013 ). However, these too come with challenges. Including more than one zoo in a research project requires much time and effort to establish and maintain professional relationships, as well as ensuring the data collected from each institution is of the same quality and detail. Depending on the type of study, this could be relatively easy. For example, a pivotal study of personality and longevity in lowland gorillas included 43 zoos but the data came from one database, all of the zoos involved were from countries speaking the same language and one of the authors worked at one of the participating zoos (Weiss et al., 2013 ). However, things become more difficult if data need to be collected by a researcher present at a zoo (e.g. behavioural data) or if the zoos involved are not all from countries speaking the same language (e.g. involving translations of test materials). Data collection from more than one zoo could then be hindered by monetary and time constraints, as well as lack of staff or a lack of interest. Anecdotally, experience demonstrates that personality studies still seem to face these problems in zoos, despite growing literature showing their importance. However, the Association of Zoos and Aquariums (AZA) has increased its focus on research, recently creating a Research and Technology Committee. AZA expects its members to conduct or facilitate research, and provides some grant funding for such projects. In addition, new technological projects, such as ZooMonitor, a Web application designed for easier and more reliable data collection, is available to zoos at little to no cost (https://zoomonitor.org: Lincoln Park Zoo, Chicago, IL, USA). This type of technology could facilitate multi-zoo studies with only minimal training required.


Killer whales share personality traits with humans, chimpanzees

Killer whales display personality traits similar to those of humans and chimpanzees, such as playfulness, cheerfulness and affection, according to new research published by the American Psychological Association.

Researchers in Spain analyzed the personality traits of 24 captive killer whales (Orcinus orca) at SeaWorld Orlando, SeaWorld San Diego and the Loro Parque zoo in Tenerife, Spain, which operates its killer whale program in partnership with SeaWorld. Six of the killer whales were caught in the wild while the remainder were born in captivity.

Trainers and other staff who worked closely with the killer whales completed surveys ranking each animal on a list of 38 personality traits, including playfulness, independence, stubbornness, bravery, sensitivity and protectiveness. Those traits were analyzed and compared with previous studies of the same personality traits for chimpanzees and humans. The research was published online in the Journal of Comparative Psychology.

"This is the first study to examine the personality traits of killer whales and how they relate to us and other primates," said lead researcher Yulán Úbeda, a doctoral student in psychology at the University of Girona in Spain. "These similar personality traits may have developed because they were necessary to form complex social interactions in tightly knit groups that we see in killer whales, humans and other primates."

The study used a common personality measure called the five-factor model, which assesses five personality dimensions, including extraversion, agreeableness, conscientiousness, dominance and carefulness. The model, which was developed in the 1930s, describes personality traits using a combination of single adjectives or descriptive phrases.

Researchers found that the personality traits of killer whales were similar to those of both humans and chimpanzees but more similar to chimpanzees. Killer whales were similar to chimpanzees and humans for the extraversion factor (e.g., playful, gregarious and sociable). Killer whales and chimpanzees also shared a combination of personality traits for conscientiousness (e.g., constant, stubborn and protective) and agreeableness (e.g., patient, peaceable and not bullying), along with some personality traits relating to dominance.

The findings may suggest some evolutionary convergence where the personality traits of killer whales and primates are similar because of the advanced cognitive abilities required for complex social interactions. Killer whales can live into their 90s in tightly knit pods that hunt together and share their food, displaying advanced communication skills and cooperation. Those traits received international news coverage when a 20-year-old killer whale, known as J-35 or Tahlequah, began pushing her dead newborn calf last July off the coast of Vancouver Island. With the help of other members of her pod, she kept the dead calf afloat for 17 days while swimming hundreds of miles, an exhausting effort that interfered with the pod's ability to hunt.

There is some debate among researchers about whether the mother whale's actions displayed grief over her loss. Despite their name, killer whales actually are the largest member of the dolphin family. Both killer whales and dolphins have been known to keep dead newborns afloat, but not for such a lengthy period. Grief is a complex emotion that hasn't been observed in most animals, although some research points to mourning by some species that live in tight-knit groups, including chimpanzees, elephants and giraffes.

The current study didn't analyze any aspects of grief in killer whales, and it's difficult to know why killer whales carry dead offspring since killer whales can't communicate with humans, Úbeda said.

"Some previous studies suggest that the mother's contact with the lifeless body could be important for the mother to make a psychological adjustment to the death of her offspring," she said. "In any case, those behaviors show how complex these animals are."

Previous research has shown that captivity can alter the personality of killer whales, increasing neuroticism and aggression. Physical changes, such as dorsal fin collapse, also have been observed. SeaWorld, which was involved in this study, has faced scrutiny over the cramped living conditions for its captive killer whales.

The current study didn't analyze the effects of captivity and was conducted with captive killer whales because it would be difficult to assess the personality traits of killer whales in the wild, Úbeda said. The personality traits that were observed in the small sample size of 24 captive killer whales could differ from the personality traits of killer whales in the wild, she said.


How Neanderthal are you… and can you blame your DNA? Personality quiz

Y ou’ve heard of your IQ (Intelligence Quotient) and your EQ (Emotional Intelligence Quotient) but what about your NQ (Neanderthal Quotient)? A new study from the State University of New York analysed 200 Homo sapiens to see which personality traits they shared with our distant cousins Homo neanderthalensis. So, what about you? How often (never/occasionally/often) do you…

(a) Fantasise about sex with someone other than your partner? (b) Avoid talking to people you don’t know very well? (c) Feel so nervous that nothing could calm you down? (d) Show a lack of imaginativeness in new situations?

If you mostly answered never/occasionally, then you have a low NQ. If you mostly answered often, you have a relatively high NQ. At a glance, this looks crude: we Homo sapiens are monogamous, sociable, calm and imaginative, while Neanderthals were promiscuous and brutish. But when the researchers took DNA samples from the participants and looked for correlations between their personalities and their genetic overlap with Neanderthal DNA, the correlations were small but statistically significant. Since generally the evidence for heritable personality traits is strong, it is not so far-fetched to imagine a future in which our personalities are measured not with questionnaires, but by sequencing our DNA.


Discussion

Zebrafish offspring activity appears to be affected by interactions between relatively stable, putatively genetic and plastic, epigenetic factors of paternal behaviour where both, behaviour associated with paternal boldness and social status influenced offspring activity. In particular, males starting off as dominant and being subordinate in the second round showed the strongest effects on ejaculate traits as well as offspring activity, as their sperm was consistently slower than the sperm of all other males, but their offspring maintained a higher activity at the age of ten days pf after the second round. Males that maintained a dominant status during both experimental rounds showed the strongest shift in offspring activity when compared across the two rounds, whereas their sperm traits were similar to those of males being subordinate across both rounds or switching from subordinate to dominant. Overall, our results suggest that there is an interaction between parental genetic and non-genetic factors that determine offspring behaviour.

Interestingly, we found no significant relationship between paternal boldness and dominance rank, even in males that maintained their social status throughout the experiment. This finding is somewhat surprising given the fact that these relationships have been documented in zebrafish before [23], but also in other species including the rainbowfish Melanotaenia duboulayi [44] and the zebrafinch Taeniopygia guttata [45]. Our results may indicate that social status and boldness are not necessarily as strongly coupled as previously assumed. Differences between studies may arise due to variation in experimental design. In fact, the previous study in zebrafish showing that boldness could be predicted based on their social status was assessed in individuals of both sexes across three different assays without testing for repeatability [23]. Furthermore, the variables analysed differed somewhat between this previous study and included distance moved, but no variable on freezing behaviour, which we found to be an important indicator of anxiety and shyness. These discrepancies across different studies confirm that we still have very little understanding of the heritability and consistency of behavioural patterns. Nevertheless, our results support the theoretical predictions that personality traits are context-dependent and are consistent with findings in a range of other taxa [19, 20, 25]. A study in the domestic fowl Gallus gallus domesticus showed that current social status contributes critically to both variation and stability in behavioural responses [22]. Similarly, social context was found to influence behaviour in mink Neovison vison, despite generally stable behavioural responses in repeated trials [46]. Furthermore, in the African cichlid Oreochromis mossambicus, behavioural responses were inconsistent over time and largely depended on the social context [47].

The environmental context, both social and ecological, can play a major role for the expression of behavioural phenotypes [20]. The social context has recently been highlighted in affecting the expression of personality in group living animals, and plays fundamental roles for individual behaviour as well as in group dynamics (reviewed in [48]). In fact, even short-term social circumstances and experiences can have profound effects on behavioural performance. Bystander effects, where changes in behaviour are the result of watching the performance of conspecifics, influence the level of boldness in rainbow trout Onchorhyncus mykiss [49]. Similarly, ecological conditions may play an important role in the manifestation of personality. The introduction of predation pressure can induce a correlation between boldness and aggression, which is absent under low predation pressure in threespined stickleback, Gasterosteus aculeatus [50]. The correlation between proportion of extra-pair paternity and response to novel objects also varies according to operational sex ratio in a large-scale study on captive zebra finches Taeniopygia guttata [24]. These findings emphasize the context-dependence of behavioural traits and the importance of taking the context into account when performing behavioural studies.

Social circumstances are known to not only influence behavioural traits but also affect male physiology and as a result ejaculate traits such as sperm number [51], sperm velocity ([12],e.g. [52, 53]) and sperm morphometry [54, 55]. These traits are generally assumed to increase a male’s success during sperm competition and result in higher reproductive success. However, the fitness advantage may not only be due to increased fertilisation success, but also due to increased fitness in offspring of better competitors [56, 57]. In fact, we found a significant link between sperm swimming velocity and offspring activity. In our study, the decline of sperm velocity over time was markedly different in males that switched from dominant to subordinate status (DomSub males), but not in all other males. In addition, DomSub males that were classified as particularly bold in the novel object trial had faster swimming sperm. This pattern was directly reflected in offspring activity: offspring sired by DomSub males experienced no drop in activity levels between day seven and day ten, whereas offspring sired by all other males did show a drop in activity levels. It is possible that offspring activity is associated with personality and could potentially be predictive of future social status. This could be an interesting area of future research.

The evidence for direct links between offspring traits and sperm-mediated epigenetic effects is mounting rapidly. In particular, the long-term transgenerational effects of maternal exposure to toxins such as vinclozolin on offspring traits in rats Rattus norvegicus have been linked to epigenetically induced changes in sperm methylation patterns across multiple subsequent generations (for example [58, 59]). However, so far only few studies provided evidence for the effects of short-term exposure to less toxic treatments, which influence not only the male ejaculate but also offspring performance. In Drosophila melanogaster for example, a two-day short-term sugar treatment in fathers affected offspring metabolic state and obesity, and was connected with chromatin-state alterations in sperm [60]. In the red flour beetle Tribolium castaneum, transgenerational immune priming was not only linked to paternal transmission, but also indicated transfer via sperm [61]. Our recent study in zebrafish demonstrated the link between the social environment, specifically the level of male-male competition for access to females, sperm velocity, and offspring hatching speed and survival [12].


Testing chimps in Tanzania over decades suggests personality types are stable

Credit: CC0 Public Domain

A team of researchers affiliated with institutions in the U.S., the U.K. and Tanzania has found evidence that suggests personality traits in chimpanzees are relatively stable over long periods of time. In their paper published in the journal Scientific Data, the group describes the history of the testing, the types of tests given and what they revealed.

Jane Goodall is among the most famous researchers in the world today, still nearly a half-century after her groundbreaking work studying chimpanzees in their natural environment in Gombe National Park, Tanzania. Back then, many have noted, the consensus among scientists was that personality traits in animals were invented by amateurs succumbing to anthropomorphism. Goodall was one of the first to suggest very strongly that this was not the case and that animals, particularly chimpanzees, had unique personalities every bit as real as humans.

To offer proof, she and her colleagues took and administered tests to one another that served to classify the personality traits they found in the chimps they were studying. Each of the researchers was asked to rate each of the chimps under observation regarding their personalities for such things as how aggressive or trustful they found them. Since that time, a lot of research has been conducted regarding animal personalities and the consensus has changed—now, it is believed that most animals have unique personalities. In this new effort, the researchers have revisited the histories of the same chimps that were tested in the 1970s and have tested other chimps to learn more about the stability of personality traits as the chimps live their lives over many years.

The researchers used different tests, but note the categories in the new tests could be correlated with traits on the original tests. To learn about trait stability in the original chimps (most of which have died), the researchers asked people who had studied them over the years to take the tests. In addition, they also asked other workers studying other wild chimps to take the tests at different points in time. The researchers then compared the personality scores for all of the chimps across multiple time periods. They report that they found remarkable consistency. While there were some caveats and some minor changes due to accumulation of wisdom as the chimps aged, their basic personalities, like those of humans, remained intact.

Abstract
Researchers increasingly view animal personality traits as products of natural selection. We present data that describe the personalities of 128 eastern chimpanzees (Pan troglodytes schweinfurthii) currently living in or who lived their lives in the Kasekela and Mitumba communities of Gombe National Park, Tanzania. We obtained ratings on 24 items from an established, reliable, well-validated questionnaire used to study personality in captive chimpanzee populations. Ratings were made by former and present Tanzanian field assistants who followed individual chimpanzees for years and collected detailed behavioral observations. Interrater reliabilities across items ranged from acceptable to good, but the personality dimensions they formed were not as interpretable as those from captive samples. However, the personality dimensions corresponded to ratings of 24 Kasekela chimpanzees on a different questionnaire in 1973 that assessed some similar traits. These correlations established the repeatability and construct validity of the present ratings, indicating that the present data can facilitate historical and prospective studies that will lead to better understanding of the evolution of personality in chimpanzees and other primates.


New Study of 17,000 Human Traits — and 100% of Them Are Heritable

From the “Discussion” section of the study:

WE HAVE conducted a meta-analysis of virtually all twin studies published in the past 50 years, on a wide range of traits and reporting on more than 14 million twin pairs across 39 different countries. Our results provide compelling evidence that all human traits are heritable: not one trait had a weighted heritability estimate of zero.

The relative influences of genes and environment are not randomly distributed across all traits but cluster in functional domains. In general, we showed that reported estimates of variance components from model-fitting can underestimate the true trait heritability, when compared with heritability based on twin correlations. Roughly two-thirds of traits show a pattern of monozygotic and dizygotic twin correlations that is consistent with a simple model whereby trait resemblance is solely due to additive genetic variation. This implies that, for the majority of complex traits, causal genetic variants can be detected using a simple additive genetic model.

Meta-Analysis of the Heritability of Human Traits Based on Fifty Years of Twin Studies

Despite a century of research on complex traits in humans, the relative importance and specific nature of the influences of genes and environment on human traits remain controversial. We report a meta-analysis of twin correlations and reported variance components for 17,804 traits from 2,748 publications including 14,558,903 partly dependent twin pairs, virtually all published twin studies of complex traits. Estimates of heritability cluster strongly within functional domains, and across all traits the reported heritability is 49%. For a majority (69%) of traits, the observed twin correlations are consistent with a simple and parsimonious model where twin resemblance is solely due to additive genetic variation. The data are inconsistent with substantial influences from shared environment or non-additive genetic variation. This study provides the most comprehensive analysis of the causes of individual differences in human traits thus far and will guide future gene-mapping efforts. All the results can be visualized using the MaTCH webtool.