A three-armed bandit task measuring social exploratory/exploitative behavior in older adults
Poster Session D - Monday, March 31, 2025, 8:00 – 10:00 am EDT, Back Bay Ballroom/Republic Ballroom
Christian Valtierra1 (christian.valtierra@ucsf.edu), Lorenzo Pasquini1, Adam Gazzaley1, Gabriella Mace1, Avery Ostrand1, Maria Auil1, Patrick McConnell1, Sydney Griffith1, Jeremy Hogeveen2, Nathan Spreng3; 1Neuroscape, Neurology Department, University of California, San Francisco, CA, USA, 2University of New Mexico, Albuquerque, NM, USA, 3Montreal Neurological Institute, McGill University, Montreal, Canada
Accomplishing one's goals in life requires a balancing act between exploiting known strategies, and exploring novel options that may be more fruitful down the line. This exploration-exploitation trade-off changes with age, with older adults typically displaying more exploitative behavior when compared to younger adults. Here, we used a social variation of the three-armed bandit task to explore social exploratory/exploitative behavior in older adults. We adapted a three-armed bandit task presenting faces of older adults. Three faces were presented at the same time, each with distinct probabilities of reward. Participants were asked to select the image that is most likely to result in a reward. At regular intervals, a novel image with a new reward probability was inserted pseudo-randomly, thus providing the participant with an opportunity to explore a new source of potential reward. Individual rates for best choice, worst choice, and novel choice were derived from this behavioral data. Participants were significantly more likely to choose the best option than the worst option. Participants were most likely to follow their current best strategy, although they engaged with the novel stimulus at a greater frequency than the known worst option. In a previous study applying a non-social version of the three-arm bandit task in younger adults, participants chose novel stimuli at a much higher rate when compared to older adults. Future research is needed to evaluate whether this reflects an expected effect of age or whether the social nature of our task resulted in a lower exploration rate.
Topic Area: EMOTION & SOCIAL: Development & aging