The impact of reward and punishment on skill learning depends on task demands

Abstract

Reward and punishment motivate behavior, but it is unclear exactly how they impact skill performance and whether the effect varies across skills. The present study investigated the effect of reward and punishment in both a sequencing skill and a motor skill context. Participants trained on either a sequencing skill (serial reaction time task) or a motor skill (force-tracking task). Skill knowledge was tested immediately after training, and again 1 hour, 24–48 hours, and 30 days after training. We found a dissociation of the effects of reward and punishment on the tasks, primarily reflecting the impact of punishment. While punishment improved serial reaction time task performance, it impaired force-tracking task performance. In contrast to prior literature, neither reward nor punishment benefitted memory retention, arguing against the common assumption that reward ubiquitously benefits skill retention. Collectively, these results suggest that punishment impacts skilled behavior more than reward in a complex, task dependent fashion.

Intention to learn modulates the impact of reward and punishment on sequence learning

“…It is possible that these two processes interact competitively at the level of medial temporal lobe, thereby reducing the expression of early sequence knowledge. This possibility may be interesting to explore in the future using patients with medial temporal lobectomy 41 . Reward and punishment did not impact immediate retention Reward did not benefit performance on the immediate retention probes in the intentional or the unintentional learning conditions, which effectively replicates the similar null result found in prior work by our group 6 . Given the widely observed benefit of reward to visuomotor adaptation memory retention, it will be important to understand whether any specific conditions reliably foster reward-related memory benefits in the context of motor skill learning 4,11,22,23 .…”

“…Trials with RTs less than 150 ms and greater than 800 ms were excluded from the analysis. The bounds of this exclusion criteria are consistent with prior work from our lab 6,12 .…”

“…Yet despite the importance of valenced feedback for learning, the effects of reward and punishment on performance during the learning of motor skills have largely been unexplored 4 . In the context of motor sequence learning, two previous studies both demonstrated that punishment improves performance during learning 5,6 . However, punishment was not beneficial on a different motor sequencing task that required more fine motor control compared to the SRTT 6 , suggesting that generalizing the effect of reward and punishment across tasks may be limited.…”

“…Only two previous studies have used the SRTT to study the influence of reward and punishment on sequence learning. In our previous study, participants were not told about the presence of a sequence and demonstrated implicit acquisition when tested at 30-days post-learning 6 , and in the study by Wachter and colleagues, participants were excluded if they achieved explicit awareness of the presence of the sequence 5 .…”

Differential impact of reward and punishment on functional connectivity after skill learning

“…The potential to use reward and punishment, collectively referred to as valenced feedback, during training has been pursued in recent years as a potential method to increase skill learning and retention (Abe et al, 2011;Galea et al, 2015;Steel et al, 2016;Wachter et al, 2009). Prior behavioral studies of motor adaptation suggest that reward and punishment have differing effects on motor learning.…”