Flexible information-seeking in chimpanzees

Abstract

Humans can flexibly use metacognition to monitor their own knowledge and strategically acquire new information when needed. While humans can deploy these skills across a variety of contexts, most evidence for metacognition in animals has focused on simple situations, such as seeking out information about the location of food. Here, we examine the flexibility, breadth, and limits of this skill in chimpanzees. We tested semi-free-ranging chimpanzees on a novel task where they could seek information by standing up to peer into different containers. In Study 1, we tested n = 47 chimpanzees to assess if chimpanzees would spontaneously engage in information-seeking without prior experience, as well as to characterize individual variation in this propensity. We found that many chimpanzees engaged in information-seeking with minimal experience, and that younger chimpanzees and females were more likely to do so. In two subsequent studies, we then further tested chimpanzees who initially showed robust information-seeking on new variations of this task, to disentangle the cognitive processing shaping their behaviors. In Study 2, we examined how a subset of n = 12 chimpanzees applied these skills to seek information about the location versus the identity of rewards, and found that chimpanzees were equally adept at seeking out location and identity information. In Study 3, we examined whether a subset of n = 6 chimpanzees could apply these skills to make more efficacious decisions when faced with uncertainty about reward payoffs. Chimpanzees were able to use information-seeking to resolve risk and choose more optimally when faced with uncertain payoffs, although they often also engaged in information-seeking when it was not strictly necessary. These results identify core features of flexible metacognition that chimpanzees share with humans, as well as constraints that may represent key evolutionary shifts in human cognition.

Keywords: Comparative cognition; Comparative development; Decision-making; Metacognition.

Figure 1:. Basic setup for studies.

A chimpanzee stands up and looks inside containers in the task. Chimpanzees could not see into the containers from their typical position (seated in front of the table), but they could see if they stood or climbed up. When the tabletop was pushed forward, the ends of the strings were within reach so they could choose a container by pulling its attached string.

Figure 2:. Overview of studies and trial types.

All individuals first completed Study 1; individuals who met criteria in this study were eligible to proceed to subsequent studies (they completed either Study 2 or Study 3). Sessions typically comprised an initial introductory phase that introduced key aspects of the task or confirmed basic comprehension of the setup (such as pulling the container using strings, or preferences for different kinds of foods), and then presented the key trials for the study in a subsequent test phase.

Figure 3:. Spontanous information-seeking (Study 1).

(a) Overall, 72% of individuals looked within three minutes on their first experience with the novel problem on the long look trials; panel shows breakdown by sex and age cohort and sex. (b) Individual variation in propensity to engage in information-seeking. Younger chimpanzees and females showed more looking responses across the four long look trials each lasting up to 3 minutes; ribbon indicates 95% CI from GLMM model estimates of trial-by-trial data accounting for trial number as well as age, sex, and their interaction; scatter plot indicates individuals’ mean proportion of looks. (c) Choices in the short look trials by looking response and age cohort. Chimpanzees of all ages selected the correct option at high rates after they looked on these trials (in a 10s interval), but chose at chance if they did not. Dashed line indicates chance; error bars indicate SE. (d) Looking responses that occurred within 10s across all trial phases, broken down by age cohort. Chimpanzees showed low rates of baseline looking responses in the three initial introductory phases where such responses were not necessary, but showed increases in looking responses in the subsequent long look and short look trials; error bars indicate SE.

Figure 4:. Information-seeking about reward location versus identity (Study 2).

(a) Chimpanzees looked more often after hidden baiting compared to visible baiting, and did so at similar rates for information about the location of a reward and information about the identity of available rewards. (b) Across both conditions, chimpanzees selected the correct option at high rates when baiting was visible regardless of whether they looked. After hidden baiting, they also chose at high rates after looking, but chose at chance if they did not look. Error bars indicate SE, and dashed line indicates chance.

Figure 5:. Looking and choice patterns in response to risky outcomes in Study 3.

On risky choice trials, chimpanzees showed (a) a similar propensity to look across bating conditions, but (b) they were faster to look after hidden compared to visible baiting of risky outcomes. (c) On risk visibility exposure trials, which were identical to risky choice trials except that only the risky option was available, chimpanzees produced more looks following hidden baiting. (d) Chimpanzees selected the correct option on risky choice trials (that is, the option that provided the optimal reward on that trial, which depended on the risky outcome) at high rates when baiting was visible regardless of whether they looked, and when they looked after hidden baiting, but chose at chance if they did not see the baiting and did not produce a look. Error bars indicate SE, and dashed line indicates chance.