SEARCH

SEARCH BY CITATION

Keywords:

  • gorilla;
  • personality;
  • well-being;
  • five-factor model;
  • captive management;
  • conscientiousness

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

The purpose of this study was to determine the personality structure of eight male gorillas (five silverbacks and three blackbacks) housed at the Sedgwick County Zoo in Wichita, Kansas and to determine if personality predicts behavior and subjective well-being in male gorillas living in bachelor groups. We used the Hominoid Personality Questionnaire which contains 54 descriptive adjectives with representative items from the human five-factor model. Rates of 12 behaviors that are broadly defined as agonistic or affiliative were independently recorded and calculated. Principal components analysis yielded three reliable personality factors: Dominance, Extraversion/Agreeableness, and Conscientiousness. These results are the first potential quantitative evidence for a Conscientiousness factor in a hominoid other than chimpanzees and humans. This suggests that Conscientiousness originated with the common ancestor of male gorillas and humans around 10 million years ago. These results indicate that humans can reliably assess the personality and subjective well-being of captive male gorillas living in bachelor groups with robust levels of inter-rater reliability and validity. Furthermore, personality can accurately predict behavior (r = 0.79; n = 13) and subjective well-being (r = 0.83; n = 5) in gorillas and provide convergent and discriminant validity for the personality factors. The results advocate for the use of personality questionnaires in the captive management of bachelor gorillas over long-term multi-institutional behavioral studies. Am. J. Primatol. 76:879–889, 2014. © 2014 Wiley Periodicals, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Personality assessment can be a valuable tool to aid in the captive management of gorillas, specifically with the formation and maintenance of bachelor groups, because it is a quick, low-cost alternative to protracted, multi-institutional behavioral studies [Kuhar et al., 2006]. Bachelor group formation is of concern because of the male-biased birth sex ratio in the 1980s [Faust et al., 2000], intense male–male competition for females, and the space constraints of captivity. Therefore, usually only one adult breeding male can be in any group with females. This leaves a “surplus” of male gorillas that must be housed in bachelor gorilla groups [Stoinski et al., 2004]. Personality factors that are predictive of high affiliative behavior rates and low aggressive behavior rates may be useful in predicting which males are better suited to living in all-male groups [Kuhar et al., 2006]. For example, silverbacks that scored high on Understanding (Agreeableness) were more accepting of blackbacks maturing into silverbacks within their bachelor group than those who scored low on Understanding [Kuhar et al., 2006]. Personality also has important health and life history implications for both humans [Chapman et al., 2011] and non-human primates [Capitanio, 2011]. Personality impacts longevity in captive gorillas with more extraverted gorillas living longer [Weiss et al., 2013], and must have played a role in the evolution of hominins [King et al., 1999] and other great apes.

Recent studies have shown that gorilla, chimpanzee, and orangutan personality shares many similarities to the human five-factor model, along with some individual species differences [Gold & Maple, 1994; King & Figueredo, 1997; Weiss et al., 2006]. The human five-factor model consists of Extraversion, Agreeableness, Neuroticism, Conscientiousness, and Openness [Goldberg, 1990]. The most significant difference in personality structure between humans, chimpanzees, gorillas, and orangutans is that humans and chimpanzees have been shown to have a Conscientiousness factor [King & Figueredo, 1997], while orangutans [Weiss et al., 2006] and gorillas [Eckardt et al., 2012; Gold, 1993] have not. The present study is the first to demonstrate the existence of a conscientiousness factor in captive gorillas.

Humans can reliably assess personality and subjective well-being in chimpanzees and orangutans with high levels of reliability and validity [Freeman & Gosling, 2010; King & Landau, 2003; Weiss et al., 2006]. Subjective well-being is an important expression of personality and has been shown to correlate with personality factors in humans [Diener et al., 2000; McCrae & Costa, 1991], chimpanzees [King & Landau, 2003], and orangutans [Weiss et al., 2006]. Therefore, it can also be used as a measure of the external validity of the personality factors [Weiss et al., 2006]. In addition, subjective well-being has been shown to have a strong positive effect on longevity of zoo-housed orangutans [Weiss et al., 2011a].

Personality assessment as an alternative to behavioral studies is only useful if personality factors accurately predict behavior [Kuhar et al., 2006]. Previous studies of captive gorillas have shown significant yet weak correlations between personality factors and behavior [Gold, 1993; Kuhar et al., 2006]. Specifically, Kuhar et al. [2006] found that Extraversion was significantly positively related to affiliative behavior (initiate and receive) and Dominance was significantly positively related to displacement (initiate) and significantly negatively related to displacement (receive). Two recent studies that used trait adjectives instead of personality factors to correlate with behavior also found significant relationships [Eckardt et al., 2012; Murray, 2011].

The primary purpose of this study is to determine the personality structure of male gorillas in order to provide insight for the management of captive gorillas and specifically for the formation of bachelor groups. Knowledge of gorilla personality structure can also provide insight into the evolution of personality in hominoids. The secondary purpose of this study is to determine if personality predicts behavior and subjective well-being in gorillas. Both, if demonstrated, provide construct validity to the personality dimensions. If personality accurately predicts behavior in captive gorillas as it does in chimpanzees [Pederson et al., 2005], then zoos could use personality trait assessment as a quick, low-cost alternative to long-term multi-institutional behavioral studies in order to identify those males most suited to living in all-male groups [Gold, 1993; Gold & Maple, 1994; Kuhar et al., 2006].

Hypotheses and Predictions

Hypothesis 1: Personality predicts behavior in captive male gorillas living in bachelor groups. Predictions 1–4 are for convergent validity. Predictions 5 and 6 will provide discriminant validity. We used [Kuhar et al., 2006] as a model for our predicted significant and non-significant correlations.

Prediction 1: Dominance will be significantly positively correlated with displacement rates (initiated) and significantly negatively correlated with displacement rates (received) [Kuhar et al., 2006].

Prediction 2: Extraversion/Agreeableness will be significantly positively correlated with the affiliative (initiated) and affiliative (received) behavior categories [Kuhar et al., 2006].

Prediction 3: Extraversion/Agreeableness will be significantly positively correlated with play rates [Vazire et al., 2007].

Prediction 4: Extraversion/Agreeableness will be significantly positively correlated with approach rates (initiated) [Pederson et al., 2005].

Prediction 5: Extraversion/Agreeableness will not be significantly correlated with the displacement (initiated) or the displacement (received) rates [Kuhar et al., 2006].

Prediction 6: Dominance will not be significantly correlated with the affiliative (initiated) or affiliative (received) behavior categories [Kuhar et al., 2006].

Hypothesis 2: Personality predicts subjective well-being in captive male gorillas living in bachelor groups [King & Landau, 2003; Weiss et al., 2006]. Predictions 1 and 2 will provide convergent validity.

Prediction 1: There will be a significant positive relationship between Extraversion/Agreeableness and the subjective well-being summation measure [King & Landau, 2003; Weiss et al., 2006].

Prediction 2: There will be a significant positive relationship between Dominance and an ability to achieve goals [King & Landau, 2003; Weiss et al., 2006].

METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

This study was noninvasive. The research complied with regulations and guidelines prescribed by the IACUC committee of the Sedgwick County Zoo, the IRB of Rutgers University, and the American Society of Primatologists' Principles for the Ethical Treatment of Non-Human Primates.

Study Site and Subjects

This research was conducted at the Downing Gorilla Forest exhibit at the Sedgwick County Zoo in Wichita, Kansas from August 2007 until March 2008. Subjects were eight male western lowland gorillas (Gorilla gorilla gorilla) housed at the Sedgwick County Zoo in Wichita, Kansas in three social groups. There were five males classified as silverbacks (≥13 years old) and three males classified as blackbacks (between 8 and 12 years old) [Harcourt & Stewart, 2007]. Two groups consisted of two silverbacks each and the third group contained one silverback with three blackbacks. All of the subjects were born in captivity, with four parent-reared and four hand-reared [Schaefer, 2011].

Behavioral Observation Methods

Behavioral observations consisted of 50-min focal animal follows on all eight gorillas in which all-occurrence data were collected [Altmann, 1974]. There was a total of 199 50-min focal animal follows with an average of around 25 focal follows for each gorilla, and at least an additional 400 ad libitum behavioral observation hours. All behavior category definitions were taken from Stoinski et al. [2004] except for Touch, Approach, and Look (Watch) Visitors which were taken from Pederson et al. [2005], and Solo Play and Groom which were taken from Schaller [1963]. Behavior rates for the correlations were calculated using only the 199 50-min focal animal follows. For duration behaviors (e.g., Play), 5 sec must have elapsed between bouts for the behavior to be scored more than once and the behavior must have occurred for at least 5 continuous seconds to be recorded [Stoinski et al., 2004].

Raters

Two male and six female zoo employees who regularly worked with the gorillas filled out the personality questionnaires. They had known the gorillas for a period ranging from 2 months to 3.8 years, with an average of 2.5 years. In addition, Schaefer rated the gorillas and had known them for a period of 7.5 months [Schaefer, 2011].

Personality Questionnaire

The Hominoid Personality Questionnaire (HPQ) contains 54 items, 43 of which comprised the original chimpanzee study by King and Figueredo [1997]. Each item is an adjective followed by one to three sentences that define the adjective within the context of gorilla behavior. Of the original 43 items, 41 were adapted from Goldberg's Taxonomy of the Big Five [1990] and 2 (autistic and clumsy) were created by King and Figueredo [1997]. As much as possible, an equal number of terms were selected from each of the five factors used in the human study keeping in mind which factors seemed relevant to the particular species being studied [King, personal communication]. Over the course of two subsequent studies [Weiss et al., 2006, 2009], 11 items were added to the original HPQ [Weiss et al., 2006, 2009] that related to Neuroticism (anxious, vulnerable, and cool), Openness (curious, conventional, individualistic, innovative, and unperceptive), and Conscientiousness (thoughtless, quitting, and distractible) factors because too few of those had been included in the original chimpanzee study [Weiss et al., 2009]. Each item was rated on a seven-point Likert scale on which 1 indicates displays either total absence or negligible amounts of the trait and 7 indicates displays extremely large amounts of the trait. The HPQ instructs raters to base ratings on overall impressions and not estimated frequencies of particular behaviors. The HPQ also instructs raters to not discuss their ratings with other raters. There were missing data: one gorilla was not rated on four items by one rater. A second gorilla was not rated on five items by the same rater. A third gorilla was not rated on four items by a different rater. In all cases, the mean score for that particular gorilla on that trait derived from the other eight raters was used to substitute for the missing data [Weiss et al., 2006].

Subjective Well-Being Questionnaire

The subjective well-being questionnaire contained four items that have been used previously to assess chimpanzee subjective well-being [King & Landau, 2003] and orangutan subjective well-being [Weiss et al., 2006]. The first item asked raters to assess the balance of positive versus negative moods in the target gorilla; the second item asked raters to indicate how pleasurable and satisfying social interactions were for the target gorilla; the third item asked raters to indicate how successful the gorilla was at achieving its own set of goals; and the fourth item asked how happy the rater would be if he or she were the target gorilla for a week. The raters were asked to rate each gorilla on a seven-point Likert scale. There were no missing data from the subjective well-being questionnaires.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Inter-Rater Reliability and Internal Consistency

Inter-rater reliability was estimated using two of Shrout and Fleiss's [1979] intra-class correlation coefficients (ICCs): ICC (3, 1), which indicate the reliability of ratings for a typical, single rater, and ICC (3, k), which indicates the reliability for all the raters (k) averaged together. All ICC's were calculated with SPSS software. Inter-rater reliabilities of the 54 personality items were calculated from data on eight gorillas that had been rated by nine raters per gorilla. One item (unperceptive) had negative reliability values and was eliminated from additional statistical analyses. The reliabilities of individual ratings, ICC (3, 1) ranged from 0.03 (disorganized) to 0.73 (dominant), with a mean reliability of 0.30. The reliabilities of mean ratings, ICC (3, k) ranged from 0.28 (thoughtless) to 0.96 (dominant) with a mean reliability of 0.71 (Table I). The reliabilities of individual ratings, ICC (3, 1) for the three personality factors ranged from 0.54 (conscientiousness) to 0.74 (dominance) with a mean reliability of 0.64. The reliabilities of mean ratings, ICC (3, k) ranged from 0.91 (conscientiousness) to 0.96 (dominance) with a mean reliability of 0.94 (Table II). For subjective well-being, individual rating reliabilities, ICC (3, 1), ranged from 0.27 (Goals) to 0.45 (BeGorilla and Subjective Well-Being Summation), with a mean reliability of 0.38. Reliabilities of mean ratings, ICC (3, k), ranged from 0.77 (Moods and Goals) to 0.88 (BeGorilla and Subjective Well-Being Summation), with a mean reliability of 0.83 (Table II). The Cronbach's alpha for the three personality factors ranged from 0.88 (conscientiousness) to 0.95 (dominance and extraversion/agreeableness) with a mean alpha of 0.93. The Cronbach's alpha for the subjective well-being summation measure was 0.90 (Table II).

Table I. Inter-Rater Reliabilities of 54 Personality Traits
TraitICC (3, k)ICC (3, 1)TraitICC (3, k)ICC (3, 1)
Note
  1. Inter-rater reliabilities of traits are listed from highest to lowest. If two or more traits have the same ICC (3, k) value, they are then listed alphabetically.

Dominant0.960.73Stable0.780.29
Dependent0.950.69Vulnerable0.780.29
Persistent0.940.62Gentle0.770.28
Submissive0.940.62Helpful0.770.27
Affectionate0.930.60Unemotional0.770.27
Active0.910.53Depressed0.740.24
Sociable0.910.54Stingy0.740.24
Cautious0.880.46Sympathetic0.740.24
Excitable0.880.44Anxious0.710.21
Timid0.880.45Conventional0.700.20
Solitary0.870.43Erratic0.690.20
Friendly0.860.40Decisive0.680.19
Impulsive0.860.41Reckless0.670.18
Independent0.860.41Individualistic0.650.17
Playful0.860.40Autistic0.630.16
Bullying0.850.39Jealous0.600.14
Curious0.850.38Imitative0.570.13
Irritable0.830.35Protective0.550.12
Aggressive0.820.34Quitting0.460.09
Lazy0.820.34Manipulative0.440.08
Defiant0.800.30Sensitive0.410.07
Inventive0.800.30Predictable0.400.07
Innovative0.790.29Intelligent0.390.07
Cool0.780.28Clumsy0.290.04
Distractible0.780.28Thoughtless0.280.04
Fearful0.780.28Disorganized0.220.03
Inquisitive0.780.28Unperceptive−0.20−0.02
Table II. Inter-Rater Reliabilities and Internal Consistency of Personality Factors and Subjective Well-Being Measures
ItemsICC (3, k)ICC (3, 1)Cronbach's alpha
Note
  1. Extra/Agree, extraversion/agreeableness; Moods, the extent to which the gorilla is in a positive mood; SocInt, the extent to which social interactions are satisfying to the gorilla; Goals, the extent to which the gorilla is successful in achieving its goals; BeGorilla, the extent to which the rater would like to be a particular gorilla; SWBsum, the summation of the four subjective well-being scores.

Dominance0.960.740.95
Extra/Agree0.940.650.95
Conscientiousness0.910.540.88
Moods0.770.28NA
SocInt0.870.43NA
Goals0.770.43NA
BeGorilla0.880.45NA
SWBsum0.880.450.90

Principal Components Analysis

We conducted a principal components analysis of the 53 reliable mean personality ratings using SPSS software. Six components had an eigenvalue greater than 1.00. An examination of the scree plot suggested three principal components. Parallel analysis or a Monte Carlo simulation [Horn, 1965; O'Connor, 2000] using SAS software indicated that only the first three eigenvalues (16.55, 14.56, and 10.24) were greater than expected by chance at the 95% confidence level. An oblique (promax) rotation revealed that the absolute inter-factor correlations ranged from 0.01 to 0.38 with a mean of 0.17. We therefore retained the orthogonal (varimax) rotation [Konečná et al., 2012; Weiss et al., 2011b].

Rotating the three components based on the eight mean ratings to the three components based on the 72 individual ratings (nine ratings for all eight gorillas) via targeted orthogonal Procrustes rotation [McCrae et al., 1996] using Orthosim 2.1 software indicated that the structures replicated: congruence coefficients for all three components exceeded 0.96 and the congruence for the entire structure was 0.97. The item congruencies were all greater than 0.90, with the exception of two (conventional: 0.82; irritable: 0.86). Since the number of observations that made up the latter structure was larger, comparable structures provide evidence that the structure based on mean ratings for eight subjects was stable [Konečná et al., 2012].

Additionally, we reduced our data using regularized exploratory factor analysis (REFA), a method developed specifically for small sample sizes [Jung, 2013; Jung & Lee, 2011; Jung & Takane, 2008] using MATLAB software. The factor structures extracted by REFA, with an anti-image regularization scheme and varimax rotation [Jung & Lee, 2011; Jung, personal communication] and by PCA with varimax rotation were highly comparable. The REFA solution had slightly lower loadings of factors, relative to PCA, as predicted [Jung, 2013; Konečná et al., 2012]. Targeted orthogonal Procrustes rotation [McCrae et al., 1996] confirmed this similarity-there was an overall solution congruence of 1.00 when rounded to two decimal places. The results of only the PCA factor solution are shown (Table III) because the two solutions were so similar, and PCA has been the precedent for non-human primate personality research [King & Figueredo, 1997; Konečná et al., 2008, 2012; Weiss et al., 2006, 2011b]. Because of our small sample size we only retained those items that had communalities ≥0.60. Based on this cut-off value, seven items (distractible, intelligent, imitative, thoughtless, clumsy, disorganized, and autistic) were removed from the final factor solution. The quality of factor solutions will increase if communalities of items are high (≥0.60) and if factors are highly over-determined (having a high number of variables per factor) regardless of sample size [MacCallum et al., 1999, 2001; Preacher & MacCallum, 2002]. Previous recommendations for the minimum sample size necessary for a factor analysis are not valid or useful because even very small sample sizes well below 50 can yield stable factor structures [de Winter et al., 2009]. The final factor solution contained 46 items and accounted for 89% of the total variance. Absolute factor loadings ≥0.40 were considered salient. There was some factorial complexity: 17 items had two salient loadings and 3 items had three salient loadings. In these cases, the item was assigned to the factor that had the highest loading (Table III).

Table III. Final Factor Structure of 46 Mean Adjectival Ratings
ItemsDOMEXTRACONSCIENItemsDOMEXTRACONSCIEN
Note
  1. DOM, dominance; EXTRA, extraversion/agreeableness; CONSCIEN, conscientiousness; D, dominance; E, extraversion/agreeableness; C, conscientiousness; salient items are in boldface. Final factor structure from PCA contains 46 adjectives.

FearfulD−0.980.08−0.08AffectionateE−0.310.840.36
DecisiveD0.970.000.11SympatheticE−0.370.820.39
VulnerableD−0.960.010.07UnemotionalE−0.12−0.750.54
SubmissiveD−0.950.020.13InquisitiveE0.580.72−0.31
DominantD0.94−0.13−0.03CuriousE0.510.70−0.39
TimidD−0.94−0.060.26ProtectiveE0.170.680.43
PersistentD0.920.23−0.08ManipulativeE0.560.66−0.16
IndependentD0.90−0.25−0.23IndividualisticE0.520.55−0.28
StingyD0.88−0.07−0.31ErraticC0.01−0.18−0.93
BullyingD0.87−0.19−0.31ImpulsiveC0.270.14−0.90
AnxiousD−0.82−0.07−0.21ExcitableC0.140.34−0.89
DependentD−0.810.550.08StableC0.46−0.020.88
CautiousD−0.790.210.45PredictableC0.13−0.550.82
InventiveD0.780.37−0.08DefiantC0.470.22−0.81
AggressiveD0.75−0.36−0.45IrritableC0.12−0.48−0.78
InnovativeD0.730.560.03GentleC−0.420.450.78
QuittingD−0.63−0.49−0.26CoolC0.48−0.290.76
JealousD0.600.51−0.58RecklessC0.420.49−0.66
SociableE−0.080.960.10SensitiveC−0.500.390.60
ActiveE−0.150.94−0.26    
PlayfulE−0.050.92−0.13    
LazyE0.03−0.900.18    
SolitaryE−0.19−0.900.11    
HelpfulE−0.330.880.30    
DepressedE−0.03−0.880.33    
FriendlyE−0.120.870.38    
ConventionalE−0.23−0.860.40    

Subjective Well-Being

We conducted a principal components analysis of the mean ratings of the four subjective well-being items. Only the first factor had an eigenvalue >1.00 and this factor accounted for 85% of the total variance. All four items had loadings >0.40. In addition, the four subjective well-being items were significantly correlated with a mean correlation of 0.80, and therefore each gorilla's subjective well-being score was defined as the summation of the mean ratings for all four items (SWBsum) [King & Landau, 2003; Weiss et al., 2006].

Personality, Behavior, and Subjective Well-Being Correlations

All predictions were accepted. The mean significant correlation between personality and behavior was 0.79 (n = 13) (Table IV) and between personality and subjective well-being was 0.83 (n = 5) (Table V). Dominance was significantly positively correlated with the displace (initiated) category and significantly negatively correlated with the displace (received) category. Dominance was not significantly correlated with the affiliative (initiated) or the affiliative (received) categories. Extraversion/Agreeableness was significantly positively correlated with the affiliative (initiated and received), play, and approach (initiated) categories. Although not predicted, Extraversion/Agreeableness was also significantly positively correlated with the groom (initiated), touch (initiated and received), and approach (received) categories and significantly negatively correlated with the agonistic (received) and non-contact aggression (received) categories. Extraversion/Agreeableness was not significantly correlated with displace (initiated) or displace (received) categories. No predictions were made regarding the Conscientiousness factor, but it was significantly negatively correlated with the non-contact aggression (visitors) behavior category and the watch (visitors) and public orientation categories. Extraversion/Agreeableness in gorillas was significantly positively related to all subjective well-being items except for the ability to achieve goals. Dominance was significantly positively related only to the ability to achieve goals.

Table IV. Correlations Between Personality Factors and Behavior
BehaviorDominanceExtraversion/AgreeablenessConscientiousness
Note
  • Pearson's correlations; I = initiated; R = received; a = predicted and provides convergent validity (one-tailed); b = predicted and provides discriminant validity (one-tailed); c = not predicted but provides convergent validity (two-tailed); all other correlations were two-tailed. Agonistic = displace, non-contact aggression, and contact aggression; Affiliative = groom, touch, sex, approach, and play; Public orientation = non-contact aggression (visitors), watch (visitors), and approach (visitors).

  • *

    P < 0.05.

  • **

    P < 0.01.

Agonistic (I)0.65−0.260.02
Agonistic (R)−0.21−0.71* (c)0.42
Displace (I)0.78* (a)0.19 (b)−0.34
Displace (R)−0.70* (a)−0.34 (b)0.16
Non-contact aggression (I)0.35−0.450.21
Non-contact aggression (R)0.13−0.72* (c)0.44
Contact aggression (I)0.630.56−0.01
Contact aggression (R)−0.060.26−0.04
Affiliative (I)−0.33 (b)0.85** (a)0.25
Affiliative (R)−0.28 (b)0.82** (a)0.32
Approach (I)−0.120.89** (a)0.13
Approach (R)−0.110.78* (c)0.37
Groom (I)0.180.89** (c)−0.01
Groom (R)0.280.340.32
Touch (I)−0.190.81* (c)0.19
Touch (R)−0.070.73* (c)0.33
Sex (I)−0.690.120.42
Sex (R)−0.270.580.01
Play−0.420.81** (a)0.27
Solo play−0.200.78* (c)−0.13
Public orientation0.010.02−0.80*
Approach (visitors)0.410.47−0.53
Non-contact aggression (visitors)0.020.09−0.67* (c)
Watch (visitors)0.01−0.01−0.83*
Table V. Correlations Between Personality and Subjective Well-Being
ItemsDominanceExtraversion/AgreeablenessConscientiousness
Note
  • Pearson's correlations; a = predicted and provides convergent validity (one-tailed). All other correlations were two-tailed. Moods = the extent to which the gorilla is in a positive mood; SocInt = the extent to which social interactions are satisfying to the gorilla; Goals = the extent to which the gorilla is successful in achieving its goals; BeGorilla = the extent to which the rater would like to be a particular gorilla; SWBsum = the summation of the four subjective well-being scores.

  • *

    P < 0.05.

  • **

    P < 0.01.

Moods0.520.76* (a)0.03
SocInt0.200.93** (a)−0.04
Goals0.88** (a)0.34−0.06
BeGorilla0.590.79* (a)−0.22
SWBsum0.580.78* (a)−0.08

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

This study demonstrates that personality and subjective well-being in captive gorillas can be assessed by human raters with high levels of reliability and validity as it has been the case in other great apes [King & Figueredo, 1997; Weiss et al., 2006]. A principal components analysis showed that personality in captive male gorillas living in bachelor groups may be comprised of three components: Dominance, Extraversion/Agreeableness, and Conscientiousness. This is the first study to quantitatively demonstrate the existence of a Conscientiousness factor in gorillas, which is therefore a more ancient trait than previously thought. It had previously been shown to exist only in chimpanzees and humans.

The first factor, Dominance, was similar to the gorilla [Gold, 1993], chimpanzee [King & Figueredo, 1997], and orangutan [Weiss et al., 2006] Dominance factors (Table VI) and indicates the importance of this aspect of personality in these hominoids [Kuhar et al., 2006; Weiss et al., 2006]. In an earlier study of captive gorillas, Dominance was also the inverse of the gorilla Fearful factor [Gold, 1993] (Table VI). A Dominance factor is not present in humans but some of the traits that comprise the Dominance factor in the great apes are contained in the Extraversion factor in humans. Dominance may be more important in non-human animals than in humans because of the multiple social roles that individual humans assume in society [Gosling & John, 1999]. Dominance has been demonstrated in other non-human primate species [Weiss et al., 2011b] and many other animal species [Gosling & John, 1999].

Table VI. Correlations Between Factor Scores as Defined by Sedgwick County Zoo Gorillas', Captive Gorillas' [Gold, 1993], Captive Chimpanzees', and Captive Orangutans' Personality Structures
Sedgwick County Zoo gorillasDominanceExtraversion/AgreeablenessConscientiousness
Note
  • Pearson's correlations; adjectives within a factor with a (+) sign were added together, while adjectives within a factor with a (−) sign were subtracted. The total from this summation was then divided by the total number of adjectives within the factor to give the factor score [Kuhar et al., 2006]. We computed factor scores for the eight gorillas using the list of adjectives for each personality factor from the previous captive gorilla [Gold, 1993], chimpanzee [King & Figueredo, 1997], and orangutan [Weiss et al., 2006] studies and correlated those with the eight gorillas' factor scores from this study [Konečná et al., 2008, 2012; Weiss et al., 2011b]. Since the Gold [1993] study used a different instrument we identified equivalent adjectives from the instrument we used based on their respective definitions.

  • *

    P < 0.05.

  • **

    P < 0.01.

Captive gorillas [Gold, 1993]
Dominance0.94**−0.03−0.56
Extraversion0.070.99**−0.22
Fearful−0.98**0.030.23
Understanding−0.180.73*0.55
Captive chimpanzees
Dominance0.99**−0.03−0.32
Surgency−0.120.98**−0.02
Dependability−0.57−0.150.96**
Agreeableness−0.420.700.55
Emotionality0.090.45−0.89**
Openness0.73*0.57−0.33
Captive orangutans
Extraversion0.230.95**−0.35
Dominance0.94**0.05−0.61
Neuroticism−0.660.10−0.47
Agreeableness−0.280.87**0.33
Intellect0.91**−0.28−0.24

The second factor, Extraversion/Agreeableness resembles the Extraversion factor in captive gorillas [Gold, 1993], chimpanzees [King & Figueredo, 1997], and orangutans [Weiss et al., 2006] and the Agreeableness (Understanding) factor in captive gorillas and orangutans (Table VI). It includes both social aspects of extraversion (e.g., sociable, friendly) and active (e.g., playful, active) aspects of extraversion [King & Figueredo, 1997]. It is also similar to the Extraversion and Agreeableness factors found in humans [McCrae & Costa, 1991]. Extraversion and Agreeableness have been demonstrated in other non-human primate species [Konečná et al., 2012; Weiss et al., 2011b] and many other non-human animals [Gosling & John, 1999].

The third factor, Conscientiousness, resembles the Conscientiousness factor found in humans [Goldberg, 1990] and chimpanzees [King & Figueredo, 1997] but with some differences. It includes three facets (aggressive, unpredictable, and emotional (Neuroticism)) [King, personal communication], is similar to the Dependability factor in chimpanzees, and is the inverse of the Emotionality factor in chimpanzees. It may be that the most ancestral form of Conscientiousness with all three facets appeared in the ancestor of gorillas, chimpanzees, and humans while the third facet, emotionality (Neuroticism), became part of a distinct Neuroticism factor in the last common ancestor of only humans and chimpanzees. Lastly, the aggressive (irritable) facet of the Conscientiousness factor in the common ancestor of chimpanzees and humans became part of the low pole of Agreeableness in humans, while chimpanzees retained the more ancestral form of Conscientiousness [Gosling & John, 1999; Weiss et al., 2006]. Conscientiousness was not found in orangutans [Weiss et al., 2006], wild gorillas [Eckardt et al., 2012], or captive gorillas [Gold, 1993], so its presence in captive gorillas needs to be confirmed with future studies because of our small sample size.

A Conscientiousness factor may not have been found in the previous captive gorilla study [Gold, 1993] because a different instrument (Maddingley Questionnaire) was used that did not contain adjectives that would lead to a distinct Conscientiousness factor [King, personal communication]. The wild gorilla study [Eckardt et al., 2012] that used the HPQ may not have found a Conscientiousness factor because the captive environment, that is higher interaction with humans, may highlight certain personality traits in captive gorillas [Freeman et al., 2013; Stoinski et al., 2012] that are not as evident in wild gorillas. In chimpanzees, two factors (Neuroticism and Openness) did not replicate across different types of captive environments [Weiss et al., 2009]. The captive environment also promotes certain behaviors in gorillas, such as regurgitation and reingestion [Gould & Bres, 1986], that are not present in wild gorillas. Lastly, the Big Five may not be as universal among all human cultures as previously thought [McCrae et al., 1996]. Among the Tsimane forager-farmers of Bolivia, Extraversion, Agreeableness, and Conscientiousness replicated while Neuroticism and Openness did not [Gurven et al., 2013]. If the Big Five is not universal among Homo sapiens or Pan troglodytes, then it is not surprising that it is not universal among two different species (gorilla, beringei) within the Gorilla genus.

The best way to demonstrate validity of an instrument is to have strong significant correlations between personality dimensions and behaviors that are expected to be associated with those personality dimensions (convergent validity) and non-significant correlations between personality factors and behaviors that are not expected to be associated with those personality dimensions (discriminant validity) [Freeman & Gosling, 2010; Pederson et al., 2005]. Based on the results of our predictions, we accept Hypothesis 1: Personality predicts behavior in captive male gorillas living in bachelor groups. The mean significant correlation between personality and behavior in this study was 0.79 (n = 13) (Table IV), which is higher than the mean in the chimpanzee study (0.35) [Pederson et al., 2005], all nonhuman primate studies (0.25) [Freeman & Gosling, 2010], and studies on humans (range: 0.1–0.3) [Meyer et al., 2001].

Based on the results of our predictions we can accept Hypothesis 2: Personality predicts subjective well-being in captive male gorillas living in bachelor groups. Our finding of a significant relationship between Extraversion/Agreeableness and all subjective well-being items except the ability to achieve goals is consistent with the findings for humans [McCrae & Costa, 1991], chimpanzees [King & Landau, 2003], and orangutans [Weiss et al., 2006]. Our finding of a significant positive relationship between Dominance and only one subjective well-being item, an ability to achieve goals, is consistent with the findings for orangutans [Weiss et al., 2006]. However in chimpanzees, in addition to being positively related to ability to achieve goals, Dominance was also positively related to all of the other subjective well-being items. Since male gorillas form close bonds with female gorillas [Harcourt & Stewart, 2007], dominance may not be as important to their well-being as it is for chimpanzees where males and females do not form long-term bonds. Additionally, at least for wild mountain gorillas, food is more abundant and evenly spaced [Harcourt & Stewart, 2007] so that the aggressive qualities used for hunting in chimpanzees [Goodall, 1986] are not as important to gorillas' well-being. Furthermore, in contrast to male chimpanzees, male gorillas in bachelor groups do not experience potential lethal aggression from other males [Goodall, 1986; Harcourt & Stewart, 2007]. Similarly, the solitary nature of adult male orangutans [Galdikas, 1985] may explain why Dominance is not linked to all of the subjective well-being items in orangutans as it is in chimpanzees. Subjective well-being is an important expression of personality and has been shown to correlate with personality factors in humans [Diener et al., 2000; McCrae & Costa, 1991], chimpanzees [King & Landau, 2003], and orangutans [Weiss et al., 2006]. The significant correlations between the personality factors and subjective well-being in this study provide external validity to the personality factors [Weiss et al., 2006].

This study is the first to suggest that a Conscientiousness factor exists in extant gorillas, and is therefore phylogenetically older than previously thought. This is not surprising considering humans and gorillas shared a last common ancestor around 10 million years ago and share 98% of their DNA [Scally et al., 2012]. Personality has been postulated to have an impact on human life history traits [Nettle, 2006] and could have had a similar impact on the survival and reproductive success of the common ancestor of humans and gorillas. There is a fundamental difference in the social structure of gorillas and orangutans, particularly in the relative importance of the adult male to the daily life of the adult females and their young. Adult male and female orangutans do not associate except during the time of mating [Galdikas, 1985], whereas silverbacks are the focus of the gorilla group, with male–female relationships providing the cohesive nature of the groups [Harcourt & Stewart, 2007]. In addition, adult male gorillas are unusual among primates in showing a high amount of paternal care and interaction with the young of the group [Harcourt & Stewart, 2007]. The social structure of extant gorillas indicates that the suite of correlated traits that comprises the Conscientiousness factor could be relevant to individual gorillas' reproductive success. Perhaps, a correlated combination of low aggression, low emotionality and high predictability in male gorillas would be favored by female gorillas to ensure that males will interact gently with females and their offspring. Individual variation between adult male gorillas in the suite of correlated traits that make up the Conscientiousness factor would be necessary for the emergence of this factor in the common ancestor of gorillas and humans and the maintenance of this factor in the gorilla lineage [Dingemanse & Réale, 2005; Réale et al., 2010].

This study has significant implications for the captive management and captive well-being of gorillas. The management of bachelor groups is an important issue currently in the captive management of gorillas. Presently, there are 28 males that reside in all-male groups in ten zoos in North America, and with 80 males under the age of 12 [Wharton, 2003], this number is expected to increase in the near future [Faust et al., 2000; Stoinski et al., 2004]. The results of this study could help to identify particular males on the basis of personality who are more suited to living in all-male groups (e.g., those males who score high on Conscientiousness; high on Extraversion/Agreeableness [Kuhar et al., 2006] and low on Dominance) [Gold, 1993], or those who insure varying personality profiles among group members (e.g., one gorilla can rank high on Dominance, but certainly not all).

The main limitation of this study was the small sample size which might be responsible for the smaller number of factors identified in this study—three factors in gorillas in comparison to six for chimpanzees and five for orangutans. Additionally, our findings are based on only male gorillas living in bachelor groups. Lastly, there were no species-specific items relevant only to gorillas and not to chimpanzees and orangutans added to the HPQ used in this study. It is possible that some relevant items from a gorilla (emic) perspective were missed [Uher & Asendorpf, 2008].

Future research should focus on determining the personality structure of both female and male captive gorillas living in family groups and bachelor groups, wild populations of western lowland gorillas, and bonobos (both captive and wild) to complete the picture of the evolution of personality in hominoids. Phylogenetic analyses and correlations that connect brain anatomy and genetic differences between hominoids to differences in personality structure will further validate the importance of the evolution of personality as a character trait in the biological sense in our primate lineage [Weiss et al., 2006]. In addition, more data on the relationship between personality factors and life history variables such as longevity [Weiss et al., 2013] will enhance our understanding of personality evolution in hominoids.

ACKNOWLEDGMENTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

We would like to express our gratitude to the Sedgwick County Zoo, especially Dr. Sandy Wilson, Mark C. Reed, and Mike Quick, for their support and permission to conduct this research. We would like to thank the gorilla keepers of the Sedgwick County Zoo, especially Danielle Decker, Scott LaPlante, Nathan Sexton, Ashley Suttles, Micala Teetzen, Julie Fritz, Amanda Jacquot, and Kelly Rivera for completing the personality and subjective well-being questionnaires. We thank Dr. James King for generosity with his knowledge, statistical expertise, and questionnaires. We thank Dr. Sunho Jung for providing his MATLAB codes and assistance with the REFA. We thank Netzin Steklis for her support and help with locating a research site. We thank Dr. Robin Fox, Dr. Lee Cronk, Dr. Colin G. Beer, and two anonymous reviewers for helpful comments on an earlier version of this manuscript.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES
  • Altmann J. 1974. Observational study of behavior: sampling methods. Behaviour 49:227267.
  • Capitanio JP. 2011. Individual differences in emotionality: social temperament and health. American Journal of Primatology 73:507515.
  • Chapman BP, Roberts B, Duberstein P. 2011. Personality and longevity: knowns, unknowns, and implications for public health and personalized medicine. Journal of Aging Research 2011:124.
  • de Winter JCF, Dodou D, Wieringa PA. 2009. Exploratory factor analysis with small sample sizes. Multivariate Behavioral Research 44:147181.
  • Diener E, Suh EM, Lucas RE, Smith HL. 2000. Subjective well-being: three decades of progress. Psychological Bulletin 125:276302.
  • Dingemanse NJ, Réale D. 2005. Natural selection and animal personality. Behaviour 142:11651190.
  • Eckardt W, Steklis D, Steklis N, et al. 2012. Assessing gorilla personality using the five factor model: (Abstract #284). Proceedings of the XXIV Congress of the International Primatological Society. Abstract retrieved from https://www.asp.org/ips/ips2012/conferenceschedule.cfm?year=2012#284
  • Faust LJ, Thompson SD, Earnhardt JE, et al. 2000. Innovative techniques in population demography to improve captive management of western lowland gorillas. Paper presented at The apes: challenges for the 21st century. Naperville, IL.
  • Freeman HD, Gosling SD. 2010. Personality in nonhuman primate: a review and evaluation of past research. American Journal of Primatology 72:653671.
  • Freeman HD, Brosnan SF, Hopper LM, et al. 2013. Developing a comprehensive and comparative questionnaire for measuring personality in chimpanzees using a simultaneous top-down/bottom-up design. American Journal of Primatology 75:10421053.
  • Galdikas BMF. 1985. Adult male sociality and reproductive tactics among orangutans at Tanjung puting. Folia Primatologica 45:924.
  • Gold KC. 1993. Behavioral profiles of captive lowland gorillas: effects of age, sex, rearing history, and physical environment. [dissertation]. Atlanta, GA: Georgia Institute of Technology. 154 p. Available from: University Microfilms, Ann Arbor, MI; AAT 9415637.
  • Gold KC, Maple TL. 1994. Personality assessment in the gorilla and its utility as a management tool. Zoo Biology 13:509522.
  • Goldberg LR. 1990. An alternative description of personality: the big five structure. Journal of Personality and Social Psychology 59:12161229.
  • Goodall J. 1986. The chimpanzees of Gombe: patterns of behavior. Cambridge, MA: Harvard University Press. 673 p.
  • Gosling SD, John OP. 1999. Personality dimensions in non-human animals: a cross-species review. Current Directions in Psychological Science 8:6975.
    Direct Link:
  • Gould E, Bres M. 1986. Regurgitation and reingestion in captive lowland gorillas: description and intervention. Zoo Biology 5:241250.
  • Gurven M, von Rueden C, Massenkoff M, Kaplan H, Vie ML. 2013. How universal is the big five? Testing the five-factor model of personality variation among forager-farmers in the Bolivian Amazon. Journal of Personality and Social Psychology 104:354370.
  • Harcourt AH, Stewart KJ. 2007. Gorilla society: conflict, compromise, and cooperation between the sexes. Chicago: University of Chicago Press. 439 p.
  • Horn JL. 1965. A rationale and test for the number of factors in factor analysis. Psychometrica 30:179185.
  • Jung S. 2013. Exploratory factor analysis with small sample sizes: a comparison of three approaches. Behavioural Processes 97:9095.
  • Jung S, Lee S. 2011. Exploratory factor analysis for small samples. Behavior Research Methods 43:701709.
  • Jung S, Takane Y. 2008. Regularized exploratory factor analysis. In: Shigemasu K, Okada A, Imaizumi T, Hoshino T, editors. New trends in psychometrics. Tokyo: University Academic Press. p 141149.
  • King JE, Figueredo AJ. 1997. The five-factor model plus dominance in chimpanzee personality. Journal of Research in Personality 31:257271.
  • King JE, Landau VI. 2003. Can chimpanzee happiness be estimated by human raters? Journal of Research in Personality 37:115.
  • King JE, Rumbaugh DM, Savage-Rumbaugh ES. 1999. Perception of personality traits and Semantic learning in evolving hominids. In: Corballis MC, Lea SEG, editors. The descent of mind: psychological perspectives on hominid evolution. Oxford: Oxford University Press. p 98115.
  • Konečná M, Lhota S, Weiss A, et al. 2008. Personality in free-ranging hanuman langur (Semnopithecus entellus) Males: subjective ratings and recorded behavior. Journal of Comparative Psychology 122:379389.
  • Konečná M, Weiss A, Lhota S, Wallner B. 2012. Personality in Barbary macaques (Macaca sylvanus): temporal stability and social rank. Journal of Research in Personality 46:581590.
  • Kuhar CW, Stoinski TS, Lukas KE, Maple TL. 2006. Gorilla behavior index revisited: age, housing, and behavior. Applied Animal Behavior Science 96:315326.
  • MacCallum RC, Widaman KF, Zhang S, Hong S. 1999. Sample size in factor analysis. Psychological Methods 4:8499.
  • MacCallum RC, Widaman KF, Preacher KJ, Hong S. 2001. Sample size in factor analysis: the role of model error. Multivariate Behavioral Research 36:611637.
  • McCrae RR, Costa PT. 1991. Adding liebe und arbeit: the full five-factor model and well-being. Personality and Social Psychology Bulletin 17:227232.
  • McCrae RR, Zonderman AB, Costa PT, Bond MH, Paunonen SV. 1996. Evaluating replicability of factors in the revised NEO personality inventory: confirmatory factor analysis versus procrustes rotation. Personality Processes and Individual Differences 70:552566.
  • Meyer GJ, Finn SE, Eyde LD, et al. 2001. Psychological testing and psychological assessment: a review of evidence and issues. American Psychologist 56:138165.
  • Murray L. 2011. Predicting primate behavior from personality ratings. In: Weiss A, King JE, Murray L, editors. Personality and temperament in nonhuman primates. New York: Springer. p 129167.
  • Nettle D. 2006. The evolution of personality variation in humans and other animals. American Psychologist 61:622631.
  • O'Connor BP. 2000. SPSS and SAS programs for determining the number of components using parallel analysis and Velicer's MAP test. Behavior Research Methods, Instruments, and Computers 32:396401.
  • Pederson AK, King JE, Landau VI. 2005. Chimpanzee personality predicts behavior. Journal of Research in Personality 39:534549.
  • Preacher KJ, MacCallum RC. 2002. Exploratory factor analysis in behavior genetics research: factor recovery with small sample sizes. Behavior Genetics 32:153161.
  • Réale D, Dingermanse NJ, Kazwm AJN, Wright J. 2010. Evolutionary and ecological approaches to the study of personality. Philosophical Transactions of the Royal Society B: Biological Sciences 365:39373946.
  • Scally A, Dutheil JY, Hillier LW, et al. 2012. Insights into hominid evolution from the gorilla genome sequence. Nature 483:169175.
  • Schaefer SA. 2011. Personality, subjective well-being, and vocal communication in captive male western lowland gorillas [dissertation]. New Brunswick, NJ: Rutgers University. 109 p. Available from: University Microfilms, Ann Arbor, MI; AAT 3474899.
  • Schaller GB. 1963. The mountain gorilla: ecology and behavior. Chicago: University of Chicago Press.
  • Shrout PE, Fleiss JL. 1979. Intraclass correlations: uses in assessing rater reliability. Psychological Bulletin 86:420428.
  • Stoinski TS, Kuhar CW, Lukas CW, Maple TL. 2004. Factors influencing the formation and maintenance of all-male gorilla groups in captivity. Zoo Biology 23:189203.
  • Stoinski TS, Jaicks HF, Drayton LA. 2012. Visitor effects on the behavior of captive western lowland gorillas: the importance of individual differences in examining welfare. Zoo Biology 31:586599.
  • Uher J, Asendorpf JB. 2008. Personality assessment in the great apes: comparing ecologically valid behavior measures, behavior ratings, and adjective ratings. Journal of Research in Personality 43:821838.
  • Vazire S, Gosling SD, Dickey AS, et al. 2007. Measuring personality in nonhuman animals. In: Robins RW, Fraley RC, Krueger RF, editors. Handbook of research methods in personality psychology. New York: Guilford Press.
  • Weiss A, King JE, Perkins L. 2006. Personality and subjective well-being in orangutans. Journal of Personality and Social Psychology 90:501511.
  • Weiss A, Inoue-Murayama M, Hong KW, et al. 2009. Assessing chimpanzee personality and subjective well-being in Japan. American Journal of Primatology 71:283292.
  • Weiss A, Adams J, King JE. 2011a. Happy orangutans live longer lives. Biology Letters 7:872874.
  • Weiss A, Adams MJ, Widdig A, Gerald MS. 2011b. Rhesus macaques (Macaca mulatta) as living fossils of hominoid personality and subjective well-being. Journal of Comparative Psychology 125:7283.
  • Weiss A, Gartner MC, Gold KC, Stoinski TS. 2013. Extraversion predicts longer survival in gorillas: an 18-year longitudinal study. Proceedings of the Royal Society B: Biological Sciences 280:15.
  • Wharton D. 2003. North American regional studbook of the western lowland gorilla. New York: Wildlife Conservation Society.