Journal of ComparativePsychology Copyright 1998 by the AmericanPsychologicalAssociation,Inc. 1998, Vol. 112, No. 2, 107-118 0735-7036/98/$3.00 Personality Dimensions in Spotted Hyenas (Crocuta crocuta) Samuel D. Gosling University of California, Berkeley Personality ratings of 34 spotted hyenas (Crocuta crocuta) were made by 4 observers who knew the animals well. Analyses suggest that (a) hyena personality traits were rated with generally high reliability; (b) 5 broad dimensions (Assertiveness, Excitability, Human- Directed Agreeableness, Sociability, and Curiosity) captured about 75% of the total variance; (c) this dimensional stnlcture could not be explained in terms of dominance status, sex, age, or appearance; and (d) as expected, female hyenas were more assertive than male hyenas. Comparisons with previous research provide evidence for the cross-species generality of Excitability, Sociability, and especially Assertiveness. Discussion focuses on methodological issues in research on animal personality and on the potential contributions this research can make for understanding the biological and environmental bases of personality. In 1950, Beach lamented the high proportion of animal studies that were performed on the albino rat, and he urged researchers to diversify the range of species they studied. In research on animal personality and temperament, Beach's recommendation was apparently heeded. Over the past 20 years, there has been a steady increase in published findings. A reasonably wide variety of species has been studied: octopuses (Mather & Anderson, 1993), snakes (Herzog & Burghardt, 1988), fish (Francis, 1990; Wilson, Coleman, Clark, & Biederman, 1993), rats (Garcia-Sevilla, 1984), mice (Whitney, 1970), cats (Feaver, Mendl, & Bateson, 1986), dogs (Hart & Miller, 1985), wolves (MacDonald, 1983), pigs (Forkman, Furuhaug, & Jensen, 1995), goats (Lyons, Price, & Moberg, 1988), primates (Bolig, Price, O'Neill, & Suomi, 1992; Buirski, Kellerman, Plutchik, Weininger, & Buirski, 1973; Byrne & Suomi, 1995; Caine, Earle, & Reite, 1983; Gold & Maple, 1994; King & Figueredo, 1997; McGuire, Raleigh, & Pollack, 1994; Nash & Chamove, 1981; Watson & Ward, 1996), and many others (see Gosling, 1998, for a comprehensive review). These studies typically aim to identify the major dimensions underlying personality descriptions of the target species. For example, in a seminal study, Stevenson-Hinde, Stillwell- Barnes, and Zunz (1980b) used two observers to obtain personality ratings of 46 captive rhesus monkeys and found Preparation of this article was supported by a University of California graduate fellowship to Samuel D. Gosling, by National Institute of Mental Health (NIMH) Grant MH-39917 and by NIMH Grant MH-49255. I am grateful to Elizabeth M. Coscia, Jason E. Hawk, Kathleen A. Moorehouse, and Mary L. Weldele for performing the personal- ity ratings and to Christine M. Drea, Laurence G. Frank, and Stephen E. Glickman for serving as expert consultants. I thank Jennifer S. Beer, Ver6nica Benet-Martlnez, Jack Block, Robert Hogan, Oliver P. John, James E. King, Usha L. McFarling, Delroy L. Panlhus, and Richard W. Robins for their helpful comments on earlier drafts of this article. Correspondence concerning this article should be addressed to Samuel D. Gosling, Department of Psychology, 3210 Tolman Hall, University of California, Berkeley, California 94720-1650. Elec- tronic mail may be sent to samiam@uclink.berkeley.edu. three major dimensions of personality: Confidence, Excitabil- ity, and Sociability. The above list of animal studies is based on a rather broad definition of personality. In both the human and animal domains, the distinction between personality and tempera- ment is often blurred. Moreover, some of the studies mentioned above refer to little more than individual differ- ences in specific behaviors during a single testing situation. A more satisfactory definition of personality refers to an individual's distinctive pattern of behavior that is consistent across time and situations (Pervin & John, 1997). The present study aims to contribute to the corpus of animal personality research by providing an analysis of the personality attributes of spotted hyenas (Crocuta crocuta), a previously unstudied species with a particularly interesting social structure. My main goal was to examine the major dimensions underlying individual differences in hyena per- sonality. I also explored whether these dimensions could be artifactual, reflecting nonpersonality characteristics, such as sex, age, appearance, and dominance status. In addition, to test whether the personality dimensions identified in this research are unique to hyenas, I compared findings from this study with those from three previous studies of animal personality. Using Personality Traits to Describe Animals Personality traits can be thought of as aggregated sum- mary trends in behavior (Buss & Craik, 1983), capturing an individual's characteristic patterns of behavior that persist across time and situations. Moreover, personality traits summarize behavior at a psychologically meaningful level (Funder & Colvin, 1991). That is, the various specific behaviors that are subsumed by a trait form a functionally coherent category; the behaviors of "biting people" and "scratching people" are both subsumed under the same term (aggressive) because, for most purposes, they can be treated as equivalent. But, at what level of specificity should researchers assess individual differences within a species of nonhuman animals? Does it make sense to focus on specific behaviors (e.g., "bit another individual"), or should research- 107