Adult age differences in frontostriatal representation of prediction error but not reward outcome

Abstract

Emerging evidence from decision neuroscience suggests that although younger and older adults show similar frontostriatal representations of reward magnitude, older adults often show deficits in feedback-driven reinforcement learning. In the present study, healthy adults completed reward-based tasks that did or did not depend on probabilistic learning, while undergoing functional neuroimaging. We observed reductions in the frontostriatal representation of prediction errors during probabilistic learning in older adults. In contrast, we found evidence for stability across adulthood in the representation of reward outcome in a task that did not require learning. Together, the results identify changes across adulthood in the dynamic coding of relational representations of feedback, in spite of preserved reward sensitivity in old age. Overall, the results suggest that the neural representation of prediction error, but not reward outcome, is reduced in old age. These findings reveal a potential dissociation between cognition and motivation with age and identify a potential mechanism for explaining changes in learning-dependent decision making in old adulthood.

Fig. 1

Reward-learning (MIL) task, time-course-based results: Time courses of activation comparing positive to negative prediction errors during the learning task in younger (age range 22–44; N = 12), middle-aged (age range 45–60; N = 13), and older (age range 64–85; N = 14) adults. Black lines are positive prediction errors, and gray lines are negative prediction errors. MPFC = medial prefrontal cortex; ACC = anterior cingulate cortex; NAcc = nucleus accumbens. Error bars indicate SEMs. White regions highlight feedback intervals adjusted for hemodynamic lag

Fig. 2

Reward-learning (MIL) task, whole-brain results. (A) Regions of the brain where activation was significantly modulated by prediction error at outcome across age during learning. (B) Regions of the brain where the modulation of activation by prediction error showed age differences. Cooler colors correspond to negative z scores, which indicate a reduced modulation of activation as age increased. R = right. A right/left, anterior/posterior, or superior/inferior value is listed in the upper corner of each statistical map. The anatomical underlay is an average of all subjects’ spatially normalized structural scans

Fig. 3

Simple-reward (MID) task, time-course-based results: Time courses of activation comparing gains to nongains during the task that did not require learning in younger (age range 22–44; N = 12), middle-aged (age range 45–60; N = 13), and older (age range 64–85; N = 12) adults. Black lines are gain outcomes (+$0.50, +$5.00), and gray lines are nongain (+$0) outcomes. MPFC = medial prefrontal cortex; ACC = anterior cingulate cortex; NAcc = nucleus accumbens. Error bars indicate SEMs. White regions highlight feedback intervals adjusted for hemodynamic lag

Fig. 4

Simple-reward (MID) task, whole-brain results. (A) Regions of the brain where activation was significantly greater for monetary gains than for nongains at outcome across age. (B) No significant age differences in reward modulation were apparent at outcome in the MID task. R = right. A right/left, anterior/posterior, or superior/inferior value is listed in the upper corner of each statistical map. The anatomical underlay is an average of all subjects’ spatially normalized structural scans

Fig. 5

Study 2 expanded MIL task behavior. Although younger adults outperformed older adults in the short block condition (25 trials), the older adults performed as well as the younger adults in the long block condition (75 trials). Error bars indicate SEMs. YA = younger adults; OA = older adults. ^* p < .05