In 2022, charitable giving accounted for roughly 2% of the gross domestic product in the U.S., with 64% of donations being made by individuals. Despite the significant impact of this industry on the economy, little is known about the socio-affective constructs that mediate charitable giving. In this dissertation, I assess the influences of a range of motivational and cognitive states and traits on real-world charity donations. In Chapter 1, I demonstrate that individual differences in narcissism and approval-seeking predict decisions to make public vs. anonymous donations, and that these relationships are modulated by social information about the decisions made by other donors. In Chapter 2, I use a real-time interactive “advisor-decider” task, in which advice given by one participant results in an onerous workload for another participant, to show that self-conscious affect based on performance in one domain shapes decisions to engage in charitable giving in an unrelated domain. Advisors that performed at or worse than the norm, in terms of giving incorrect advice, made more frequent subsequent charity donations: Intriguingly, when advisors were given social information about their performance relative to the norm, this pattern was reversed, such that advisors that performed worse than the norm made less frequent donations. Finally, in Chapter 3, I explored whether a recently demonstrated preference for instrumental divergence – the degree to which voluntary actions differ with respect to their outcome distributions – depends on whether monetary decision outcomes are kept for oneself or donated to a charity of one’s choice. I found that the intrinsic motivation associated with greater instrumental control applies selectively to decisions that earn money for oneself – no such preference was observed for decisions that earned money for charities. Collectively, these studies serve to better characterize the social, cognitive, and motivational processes that mediate prosocial behavior.

An ability to quickly learn about relationships between actions and outcomes is essential for adaptive behavior. Such learning can be complicated when the action-outcome relationship depends on the latency with which the action is performed. Inferences about such latency- modified contingencies can greatly improve an agent’s performance by allowing for a timing of responses that optimizes the probability of an outcome given an action. Here we specify a Bayesian reinforcement learner that infers the functional and causal form of the relationship between response latencies and response contingent outcome probabilities. The performance of this Bayesian learner is contrasted with that of a model-free actor-critic algorithm, using behavioral data from three latency-modified schedules of reinforcement. Results suggest that a model-based characterization of latency-modified schedules provides a superior account of free operant behavior. Next the test of latency-modified schedules is extended to multi-action contexts where the latency of performing one action modifies the probability of an outcome given performance of a different action. Participants quickly discovered optimal rates of responding on both the modified and modifying actions, even when rewards contingent on performing the modifying action provided local incentives for deviating from the optimal rate. A final set of experiments tested the hypothesized explicit inference about latency- modified schedules. In summary, these models and experiments highlight the advantages of incorporating latency into a causal structure of actions and goals.

Consensus seeking – abandoning one’s own judgment to align with a group majority – is a fundamental feature of human social interaction. Notably, such striving for majority affiliation often occurs in the absence of any apparent economic or social gain, suggesting that achieving consensus might have intrinsic value. The current work assessed the expression and transfer of valence associated with social conformity, and the relation between conformity and exploratory behavior. In the first two studies, using a paradigm in which participants assumed the role of a juror evaluating a series of misdemeanor criminal cases, we found that contexts that had been repeatedly paired with consensus decisions were rated as more likable, and selected more frequently in a two-alternative forced choice test, than were contexts paired with dissent from a unanimous majority. The second of these studies ruled out inferences about the accuracy of the majority opinion as the basis for such evaluative changes. A subsequent set of studies employed a simple gambling task, in which the decisions of ostensible previous gamblers were indicated below available options on each trial, to assess the trade-off between social and non-social currencies and the transfer of social valence to interpersonal stimuli. In spite of demonstrating near-perfect knowledge of objective reward probabilities, participants reliably preferred gambling options and previous gamblers associated with conformity over those associated with reward. Formally, we found that a reinforcement learner that treated conformity as a surrogate reward provided a better account of choice preferences than did a conventional model. Finally, we investigated the relationship between social conformity and a tendency to explore the environment for potentially greater, yet unknown, rewards. We found that the degree to which participants adjusted their ratings of subjective food preferences to match the aggregate ratings of ostensible previous participants was negatively correlated with exploratory behavior in a multi-armed bandit task. In summary, we provide evidence for a common value-scale for social and non-social currencies, an ability of conforming decisions to imbue concomitant stimuli with affective significance, and a negative relationship between social conformity and reward exploration. By characterizing conformity as reinforcement learning, the framework advanced in this dissertation provides a mechanism for how apparently inconsequential consensus decisions may be motivated by previously acquired valence, as well as for how that valence may in turn be transferred to concomitant stimuli, both contextual and interpersonal. This novel approach will bridge a critical gap, at neural and behavioral levels, between complex socio-cognitive representations and basic mechanisms of reward-based learning and decision-making.