Chimpanzees (Pan troglodytes) flexibly adjust their behaviour in order to maximize payoffs, not to conform to majorities

Abstract

Chimpanzees have been shown to be adept learners, both individually and socially. Yet, sometimes their conservative nature seems to hamper the flexible adoption of superior alternatives, even to the extent that they persist in using entirely ineffective strategies. In this study, we investigated chimpanzees' behavioural flexibility in two different conditions under which social animals have been predicted to abandon personal preferences and adopt alternative strategies: i) under influence of majority demonstrations (i.e. conformity), and ii) in the presence of superior reward contingencies (i.e. maximizing payoffs). Unlike previous nonhuman primate studies, this study disentangled the concept of conformity from the tendency to maintain one's first-learned strategy. Studying captive (n=16) and semi-wild (n=12) chimpanzees in two complementary exchange paradigms, we found that chimpanzees did not abandon their behaviour in order to match the majority, but instead remained faithful to their first-learned strategy (Study 1a and 1b). However, the chimpanzees' fidelity to their first-learned strategy was overridden by an experimental upgrade of the profitability of the alternative strategy (Study 2). We interpret our observations in terms of chimpanzees' relative weighing of behavioural options as a function of situation-specific trade-offs. More specifically, contrary to previous findings, chimpanzees in our study abandoned their familiar behaviour to maximize payoffs, but not to conform to a majority.

Figure 1. Depicted are the tokens that the subgroups in Study 1a were trained on: brown, plastic sticks for the majority (a), and white, plastic cups for the minority (b).

Figure 2. Schematic overview of the experimental setup in Study 1a, Leipzig Zoo (a) and Study 1b and 2, Chimfunshi Wildlife Orphanage Trust (b).

Figure 3. Mean (± s.e.m.) percentage by which the minority chimpanzees of Study 1 (n=5) observed majority strategy demonstrations, both in absolute frequencies (#) and in number of unique demonstrators (d), supplemented with the median percentages (with the lower and upper error hinge representing the first and third quartile, respectively) by which the minority chimpanzees chose to use the majority strategy per day (Exp. 1a: mean choices per individual per day = 39.9 token exchanges, range 38-103; Exp. 1b: mean choices per individual per day = 11.4 exchanges at the vending machines, range 7-23).

Figure 4. Median (with the lower and upper boxplot hinge representing the first and third quartile, respectively, and the crosses within the boxplots representing the means) preference for the trained strategy of the majority chimpanzees in Zambia throughout Study 2.

Data point at time point “1-10” represent the median preference of the majority chimpanzees for the trained strategy over the first 10 days (Study 1b). Data points at time points 11 to 20 refer to the median preferences for the trained strategy (i.e. the least productive) in the ‘superior reward contingency’ design (Study 2).