Scarcity is central to economics because limited resources force trade-offs. But can the mere feeling of scarcity affect economic behavior beyond these constraints? I study this question using theory, experimental evidence, and administrative data. I develop a model that defines perceived scarcity as resources falling below subjective needs and shows how it leads individuals to optimally reallocate limited attention toward the scarce domain and away from competing tasks. The model predicts when these shifts improve, worsen, or do not affect performance, depending on how attention translates into outcomes in each task. I test the model in a preregistered experiment in which 501 participants allocate attention between a consumption and a production task. To induce differences in perceived scarcity, I randomly assign participants to high or low consumption budgets before giving them the same intermediate budget. Despite facing identical constraints, participants accustomed to higher budgets are substantially more likely to feel scarce (77% vs. 45%). They devote 10 percentage points more attention to consumption (measured through mouse-tracking) and search more extensively, yet do not make significantly better consumption decisions. Meanwhile, their performance in the production task declines, and their total earnings are 15% lower. Exploiting a reform in Hungarian unemployment benefits, I examine the model's implications for job search. I find a pattern consistent with the model: despite facing identical current benefits, workers who experienced a larger benefit drop have a 56% higher job-finding rate but 9% lower reemployment wages.

Presented at: Internal Theory Seminar Caltech (2026),  BSPA Harvard (2026), Krajbich Lab UCLA (2026), EEA-ESEM Dublin (2026), Berlin PhD Conference (2026), ESA Europe UPF (2026), University of Bath (2026),  Behavioral Economics Spring School UCSD (2026), FUR Alicante (2026), Cognitive Foundations of Decision-Making SS Ghent (2025), Choice and Welfare Queen Mary (2025), Summer Institute in Theory-Based Experiments Caltech (2024), FUR Queensland (2024), Invited Seminar Sydney (2024), SNE (2024), EADM UPF (2022)

We develop a theoretical framework for analyzing how learning occurs directly from data in the absence of prior information. We interpret learning as the process through which decision makers identify a model, defined as the set of variables that matter for the outcome, and use those variables to form behavioral rules. The comparative statics show that extraction of the optimal rule depends not only on its own value, but also on how clearly it outperforms competing alternatives. Differences in cognitive costs then make agents respond differently to changes in relative rule values, generating systematic and predictable disagreement even when they face the same data. Finally, the framework makes rule-specific costs identifiable from behavior, allowing us to determine which observable measures of complexity can be interpreted as entering the cost function. We test the framework in our experiment and in a reanalysis of data from Kendall and Oprea (2024), finding strong support for its predictions in both settings.

Presented at:  ESA Europe UPF (2026)*, EEA-ESEM Dublin (2026)*,  HEC PhD Workshop (2026)*, FUR Alicante (2026)*, ASFEE Nancy (2025)*, Summer Institute on Bounded Rationality Max Planck Institute Berlin (2023), The 33rd Advanced School in Economic Theory Jerusalem (2023)

*coauthor presentations

Coordination failures are pervasive in economic and social life, yet their behavioral determinants remain difficult to isolate because beliefs, risk attitudes, and other traits often move together. We study these determinants using transcranial static magnetic stimulation (tSMS) over the left dorsolateral prefrontal cortex (DLPFC) in two randomized, placebo-controlled, double-blind experiments (N=112). Modulating the left DLPFC increased selection of the risky action required for efficient coordination in repeated Stag Hunt games, raising the rate of choosing this action by 16-17 percentage points. The effect was concentrated among participants with pessimistic beliefs about their partner’s future actions, indicating a greater willingness to bear strategic risk. This interpretation is supported by an independent lottery task, in which stimulated participants also became more risk tolerant. By contrast, we do not find evidence that the treatment affected measured reciprocity, patience, or cognitive reflection.

Presented at: Society for Neuroeconomics (Boston) (2025) - POSTER LINK, NeuroPsychoEconomics University College Dublin (2025), Annual Conference of the French Association of Experimental Economics (2025), Neuroeconomics of Disadvantage Workshop University of Sydney (2024), Neuroeconomics Summer School University of Pennsylvania (2023)