Research reportIndividual differences in choice (in)flexibility but not impulsivity in the common marmoset: An automated, operant-behavior choice task

•We run a pilot for operant-choice behaviour in the common marmoset (Callithrix jacchus).•Marmosets had choice between a Small & Soon (SS) vs a Large & Late (LL) food reward.•Subjects were independently classified: “flexible” vs “non-flexible” and “maximizer” vs “non-maximizer”.•Initially high LL preference and low motor impulsivity predicted flexibility of choices and better food payoff.•Marmosets are suitable to investigate individual differences in flexibility by operant panels.

Individual differences in behavioural flexibility are a significant issue in human psychopathology as well as in its animal models. We aimed to investigate individual variations of operant-choice behaviour in the common marmoset (Callithrix jacchus), a small New World primate, using a new operant panel with two hand-poking holes.MethodsExperimental subjects (N = 16) were presented with a choice between a Small & Soon (SS) vs a Large & Late (LL) food reward. After extensive training (31 daily sessions with no delay, during which a basal, large-reward preference developed), the delay before release of LL was progressively increased (from 0 to 60 s, during 16 daily sessions; indifferent point at delay = 9 s). Subjects were classified as either “flexible” or “non-flexible”, respectively, based on a decrease (or not) in the preference for LL with increasing delays. Each subject was also classified as “maximizer” (or “non-maximizer”) based on capacity (or not) to maximize the food payoff as delay increased.ResultsUpon delays shorter than the indifferent point (<9 s), none of the subjects showed a shift from LL to SS, denoting a lack of delay-induced, cognitive impulsivity. Individual differences only emerged upon delays longer than the indifferent point (>9 s), when a preference shift could be interpreted as economically-driven. In general, a profile of few unrewarded hand-pokes in reaction to initial delays (i.e., a low motor impulsivity) and of clear-cut basal LL preference seemed to predict elevated flexibility of choices and better food payoff, which was typical of subjects classified as both “flexible & maximizer”.ConclusionThese results provide normative data on the marmosets, which can be used as a model for the investigation of 1) individual differences in behavioural flexibility, as well as 2) biological mechanisms rooted in our evolutionary history.