Do chimpanzees seek explanations? Preliminary comparative investigations

Canadian Journal of Experimental Psychology, Jun 2001 by Daniel J Povinelli, Sarah Dunphy-Lelii

Abstract During the past decade, considerable effort has been devoted to understanding whether chimpanzees reason about unobservable variables as explanations for observable events. With respect to physical causality, these investigations have explored chimpanzees' understanding of gravity, force, mass, shape, and so on. With respect to social causality, this research has focused on the question of whether they reason about mental states such as emotions, desires, and beliefs. In the studies reported here, we explored whether the chimpanzee's natural motivation for object exploration is modulated by a cognitive system that seeks explanations for unexpected events. We confronted both chimpanzees and young children with simple tasks which occasionally could not be made to work. We coded their reactions to determine if they appeared to be searching for an apparent cause (or explanation) of the task failure. The results of these preliminary studies point to both similarities and differences in how young children and chimpanzees react to such circumstances.

Resume Depuis dix ans, des efforts considerables ont ste deployes en vue de comprendre dans quelle mesure les chimpanzee arrivaient a raisonner en utilisant des variables non-observables pour expliquer des evenements observables. En ce qui a trait a la causalite physique, les etudes en question ont explore la comprehension qu'ont les chimpanzee de la gravite, de la force, de la masse, de la forms, et ainsi de suite. En ce qui concerns la causalite sociale, la presents etude s'est centre sur la question de savoir si les chimpanzee raisonnent au sujet des state mentaux tels que les emotions, les desirs et les croyances. Dans lee etudes ici rapportees, nous avons examine si la propension naturelle des chimpanzes a explorer les objets etait modules par un systems cognitif recherchant des explications a des evenements inattendus. Nous avons confronts des chimpanzes et de jeunes enfants a des taches simples qui, a l'occasion, se revelaient irrealisables. Nous avons codifis leurs reactions afin de determiner s'ils semblaient chercher une cause apparente (ou une explication) a l'echec dans la tache. Les resultats de ces etudes preliminaires signalent a la fois des ressemblances et des differences dans la facon dont les enfants et les chimpanzes reagissent a ces situations. The human penchant for explanation permeates nearly every aspect of our social, emotional, and physical lives. Our natural language brims with terms of causal explanation, presumably expressing our species' fundamental curiosity about the causal relationships that underlie the physical and social events that cascade around us. Questions about "why?" and "how come?" punctuate our language from a very early age, and this quest for explanation remains firmly in place in adolescence and adulthood. By four or five years of age, children are able to produce explicit explanations for events (Crowley & Siegler, 1999; Wellman & Gelman, 1998), and an understanding of abstract causal relations appears to emerge even earlier (Bullock & Gelman, 1979; Gopnik & Sobel, 2000; Shultz, 1982; Shultz, Altmann, & Asselin, 1986a). The young child's drive for explanation has lead some to liken young children to scientists engaged in theory formation and hypothesistesting (e.g., Carey, 1985; Gopnik, 1988; Keil, 1987; Wellman, 1990). Indeed, Gopnik and Meltzoff (1997) have turned the analogy around, arguing that scientists are actually like young children: They cling to an environment in which child-like drive for explanation can continue to flourish.

The capacity for explanation may be a universal trait of the human species. Explanation appears in every culture and historical time period that has been studied (although what qualifies as an acceptable explanation for a given event may differ dramatically; for discussions of causal explanations in various cultures see Lewis, 1985; Morris, Nisbett, & Peng, 1985). Gopnik (2000) speculated that explanation is a fundamental drive of our species's psychological make-up. In short, she argued that causal explanations constitute intrinsically rewarding experiences much like the outcomes of other physiological drives like eating, drinking, and mating (see Gopnik & Meltzoff, 1997). On this view, the human mind can be thought of as being biologically adapted to seek explanations for events they witness (Go nik & Meltzoff, 1997).

Explanation is clearly an important part of human psychology, but is this psychological disposition unique to humans? Perhaps the most obvious place to turn for an answer to this question is other primate species: Do they exhibit behaviours which suggest that they seek explanations for events they witness? Certainly, nonhuman primates exhibit a drive to explore social and physical relationships. What is less clear is whether they are seeking empirical generalizations (the predictable regularities that exist in the world), or whether they are also learning about underlying causal relations. Recent investigations focusing on nonhuman primates' understanding of both social and physical causality have produced conflicting results. Some researchers argue that chimpanzees, at least, appear to appeal to mental states (i.e., emotions, desires, beliefs) and unobservable causal phenomena (i.e., gravity, force, mass) to account for, or explain, the events they observe (e.g., Boesch & Boesch-Achermann, 2000; Hare, Call, & Tomasello, 2000; Visalberghi, Fragaszy, & Savage-Rumbaugh, 1985). Other researchers have highlighted data which suggests that even in the case of chimpanzees, such sophisticated social behaviour is underwritten by the detection and understanding of the predictable regularities that exist in the world, not inferences about unobservable mental states or causal forces (e.g., Povinelli, 2000; Tomasello & Call, 1997). Because they are our nearest living relatives, the question of whether chimpanzees generate cognitive structures equivalent to human explanations is crucial to developing an understanding of the nature and timing of the evolution of this capacity.