Memory-reliant Post-error Slowing Is Associated with Successful Learning and Fronto-occipital Activity

Abstract

Abstract■ Negative feedback after an action in a cognitive task can lead to devaluing that action on future trials as well as to more cautious responding when encountering that same choice again. These phenomena have been explored in the past by reinforcement learning theories and cognitive control accounts, respectively. Yet, how cognitive control interacts with value updating to give rise to adequate adaptations under uncertainty is less clear. In this fMRI study, we investigated cognitive control-based behavioral adjustments during a probabilistic reinforcement learning task and studied their influence on performance in a later test phase in which the learned value of items is tested. We provide support for the idea that functionally relevant and memory-reliant behavioral adjustments in the form of post-error slowing during reinforcement learning are associated with test performance. Adjusting response speed after negative feedback was correlated with BOLD activity in right inferior frontal gyrus and bilateral middle occipital cortex during the event of receiving the feedback. Bilateral middle occipital cortex activity overlapped partly with activity reflecting feedback deviance from expectations as measured by unsigned prediction error. These results suggest that cognitive control and feature processing cortical regions interact to implement feedback-congruent adaptations beneficial to learning. ■

The Sustained Influence of an Error on Future Decision-Making

“…Finally, more accurate future trials could be differentiated from less accurate trials through latent process components of the decision during the first post-error trial. This shows that even in the absence of a direct relation of RT and accuracy on the post-error trial, post-error adaptations could still be beneficial with regard to future trials, as has been previously observed (Schiffler et al, 2016). …”

“…This result illustrates how the post-error adaptation process may influence accuracy of future responses (Hester et al, 2007; Schiffler et al, 2016). Given that task-relevant neural activation increases and decreases have previously been found in relation to post-error adaptations (King et al, 2010; Danielmeier et al, 2011; Steinhauser et al, 2017), this finding is not surprising.…”

“…Beyond the immediate error-correction on the first trial after an error, there are indications that error monitoring processes may improve accuracy on responses that occur later in time (Hester et al, 2007, 2008, 2009; Klein et al, 2007; Schiffler et al, 2016). For instance, Hester and colleagues demonstrated that posterior medial frontal cortex activity both on error (Hester et al, 2009) and post-error trials (Hester et al, 2007) was associated with decision accuracy several trials later.…”

“…For instance, Hester and colleagues demonstrated that posterior medial frontal cortex activity both on error (Hester et al, 2009) and post-error trials (Hester et al, 2007) was associated with decision accuracy several trials later. Further, a recent study showed that memory-reliant PES may benefit learning in a reinforcement learning context (Schiffler et al, 2016). Here, the slowing was related to stimulus specific errors that occurred on average 22 s earlier, which demonstrates that post-error response time adaptations can serve a function in learning on a longer timescale.…”

Errors and Action Monitoring: Errare Humanum Est Sed Corrigere Possibile

“…It is worth mentioning that, although the discovery of post-error behaviors dates back from the 60s, current research is still very active in this domain and largely aimed, now, at identifying the brain structures that are targeted by post-error action monitoring processes (e.g., Danielmeier et al, 2015;Beatty et al, 2018; and also at identifying the brain mechanisms (e.g., Purcell and Kiani, 2016;Schiffler et al, 2016;Perri et al, 2017), brain structures (e.g., Schiffler et al, 2016;Zhang et al, 2017;Fu et al, 2019), or neurochemical systems (e.g., Danielmeier et al, 2015;Sellaro et al, 2015) that are involved in or responsible for post-error behaviors.…”