fMRI and discounting of gains and losses

David Hardisty, Elke Weber, Daphna Shohamy
Project Complete

Recent advances in neuroeconomics have revealed the processes by which people discount future rewards. This research is investigating how and why losses are discounted differently from gains. Study 1 (a behavioral pilot) used real financial gains and losses at a delay of four weeks, manipulating outcome sign (gain vs. loss), magnitude (~$5 or ~$70), and default (default now vs. future). Study 2 was identical except that all participants completed the task while undergoing functional magnetic resonance imaging. We predicted that whereas activation of the ventromedial prefrontal cortex (VMPFC), ventral striatum and posterior cingulate cortex (PCC) would encode the decision utility of both delayed losses and gains, anterior cingulate cortex (ACC) activation would be greater for losses than gains, thus corroborating and explaining the behavioral differences between discounting of gains and losses (especially losses being discounted less than gains). We have finished data collection on this project, and data analysis is ongoing.


Major Findings

  • In the behavioral pilot, real gains and losses were discounted less than hypothetical gains and losses, but were subject to the same contextual influences (i.e., even real outcomes showed the sign effect, magnitude effect, default effect, and their interactions). This is the first rigorous comparison of real and hypothetical gains and losses in intertemporal choice.
  • Analysis of the fMRI data is ongoing, but preliminary results show that intertemporal gains show greater VMPFC and ventral striatum activation, while intertemporal losses show greater activation in the habenula.


Broader Impacts

  • This project has been successful in training young researchers Dave Hardisty was a PhD student working on the project and is now a professor at University of British Columbia. Randi Scott was an RA on both the “Tradeoffs” project and the fMRI project and is now a master’s student in the Quantitative Methods in Social Sciences program at Columbia. Randi is an African American female, so this contributes to the training and development of underrepresented groups in science. Galen Treuer was an RA on the project and is now a PhD student at University of Miami.
  • Other broader impacts will be forthcoming as data analysis proceeds.