I propose a framework of competing motivations in which different motives to exert effort can interfere with one another. Consequently, combining a strong incentive with a weaker one may be less motivating than offering the strong incentive alone. I test this hypothesis in three pre-registered experiments on Amazon MTurk via CloudResearch (N = 3,412). In the first study, participants completed real-effort tasks for monetary rewards, charitable rewards, or both; those working solely for money completed 23 percent more tasks than participants offered the same monetary reward combined with a charitable incentive. In the second study, participants worked for monetary rewards, a lottery reward, or both. While increasing the monetary piece rate raised effort, combining money with a lottery reward did not. The third study isolates the weaker incentives previously paired with money and shows that both the charity and lottery rewards independently increased effort relative to a no-incentive control. These findings support the competing motivations framework and challenge the assumption that motivations from different incentives combine additively.

We investigate the leading motivations for indirect reciprocity by experimentally studying the role of type-based, intentions-based, and outcome-based preferences. We design a novel experimental setting in which participants can help other participants under varying probability of being seen by third parties. Those third-party observers can then reward participants based on both their history of helpful behavior and the observability of that behavior. Because good deeds done in public are often strategically motivated to build reputation, while only truly altruistic people help in private, we can identify whether indirect reciprocity is directed towards good deeds or good people. We find that indirect reciprocity towards an individual is influenced by their previous attempts to be helpful and their impact on others, but is surprisingly unaffected by the authenticity of past benevolent actions. That is, good deeds are rewarded regardless of whether these actions reflect true altruism or are likely a strategic play to encourage future reciprocal kindness. We discuss implications for theories of indirect reciprocity and sustained cooperation in large groups.

We examine whether people reward helpful behavior because it reveals the helper’s general helpfulness or because the behavior personally benefited them. In a simple game, a helper makes separate decisions about whether to help two recipients, without knowing that one recipient will later have an opportunity to reward her. That recipient then observes both decisions and chooses whether to help the original helper. In a preregistered experiment, we vary the helper’s and reciprocator’s endowments so that their material payoffs are held constant across all four helping profiles. Helping the third party increases the reciprocator’s willingness to help regardless of whether the reciprocator was also helped. Yet, when the helper performs exactly one equally costly act, the helping rate is 13.4 percentage points higher when the reciprocator rather than the third party benefited. The results provide evidence for both general helpfulness and personal directedness. The former is consistent with type-based reciprocity, whereas the latter reflects the personally directed kindness emphasized by intentions-based theories. Neither account is sufficient by itself when direct and indirect interactions occur together. Our findings therefore clarify how reciprocity operates when people observe both how someone treats them and how that person treats others. 

This field experiment investigates the effect of volunteering on future donation behaviour, aiming to understand if volunteering and donating are substitutes or complements.

I investigate the interplay between the inherent cost of a prosocial activity and the feelings of personal satisfaction (often termed "warm glow") that arise from engaging in such behaviours. Using a 2×2 experimental design, participants are divided into groups working either 'for themselves' or 'for charity', undergoing tasks that are either 'easy' or 'hard'. The study is designed to test the hypothesis that participants working for charity will experience greater warm glow when they work on more difficult tasks.