The Generation Mechanism Of Post-error Adjustment And The Influencing Factors

Continuous monitoring of action outcomes is essential to achieve the goal-directed behaviors and to implement behavioral adjustments in response to the erroneous actions.It protects us from potentially harmful circumstances.Post-error slowing is a typical post-error adjustment,behaving as slower subsequent response after errors.However,the generation mechanism of post-error adjustment is still in debate.And how the influencing factors modulate post-error adjustment is still lack of evidence.To explore the above issues,we investigated the neural activity of post-error adjustment in the study 1 and study 2,and investigated how the social factors influence the post-error adjustment in the study 3 and study 4.Since resting-state f MRI is a promising technology to investigate the spontaneous brain activity,the study 1(experiment 1)explored the neural correlates of post-error adjustment by evaluating the amplitude of low-frequency fluctuations(ALFF)in a modified Go/No-go task across 75 subjects.The behavioral results showed that accelerated responses and improved accuracy were observed following No-go errors.The ALFF-behavior analysis revealed that the degree of post-error adjustment was positivity with the ALFF in bilateral insula.Thus,we consider the bilateral insula were the neural indicator of individual difference in post-error adjustment.Conscious error perception is a critical processing for the error monitoring.Although extensive studies have focused on error awareness,how error awareness emerges and influences post-error adjustment remains unclear.We investigated this issue using a Go/No-go error awareness task in 36 participants(study 2 or experiment 2).Post-error adjustments indicated two dissociated behavior patterns,with speeding up following aware errors and slowing down following unaware errors.ERN was comparable between two error types,but theta-ERN was larger for aware errors than unaware errors.Moreover,theta(4-7 Hz)was stronger for aware errors than unaware errors,even when the non-phase-locked component was removed from the EEG.Increased Pe and reduced alpha(8-14 Hz,a reversed index)were observed for aware errors compared with unaware errors.Altogether,these findings suggest that the early stage of error monitoring(ERN and theta)engages in preconscious preparation,whereas the late stage(Pe and alpha)reflects the accumulated error evidences,which drives the emergence of error awareness.Additionally,during the subjective error perception,alpha was activated up to 400 ms before perception response only for perceptive errors,suggesting that the adjustment following aware errors was induced by proactive control and the adjustment following unaware errors was induced by reactive control.Although previous studies have attempted to investigate the post-error slowing effect in the context of observed errors,the issue as to whether the post-error slowing effect generalizes across different tasksets remains unclear.And,the generation mechanism of post-error slowing after observed errors is necessary to examine.To explore the above issues,we utilized an observation –execution task in three experiments(study 3).During each trial,participants were instructed to first observe the response outcomes of their “partners” in the observation task,and then to complete their own key-press according to the response rules in the execution task.In Experiment 3,the letter Flanker task was employed in the both observation task and execution task.Moreover,three error rate conditions(20%,50% and 80%)were preset in the observation task.The results showed that the slowing effect after observed errors occurred in three error rate conditions.In particular,the proportion of observed errors and observed correct was equal in the 50% error rate condition,in which the event frequency information was effectively controlled.Therefore,these findings suggest the error information plays an important role in the slowing effect after observed errors,consistent with the assumption of the cognitive control account.In Experiment 4,the stimuli and response rules were different.A color discrimination task was utilized in the observation task,but a letter Flanker task was utilized in the execution task.The results pattern was similar to the experiment 3,suggesting the slowing effect after observed errors was a generic adjustment process,even the stimuli and response rules were distinct.However,for the accuracy,the post-observed error and post-observed correct was not significant in both experiment 3 and experiment 4.To rule out the potential influence of the ceiling effect,we conducted a control experiment 5.In Experiment 5,we shortened the response deadline in the execution task to increase the error rate.Moreover,to ensure that sufficient trials could be utilized for statistical analyses in the infrequent event condition and that the length of this experiment was the same as experiment 3 and experiment 4,we removed the 20% error rate condition and added a block to the 50% and 80% error rate conditions.The results showed that the slowing effect after observed errors were obtained in both 50% and 80% error rate condition,replicated the result in experiment 4.Additionally,the error rate increased in the experiment,but the accuracy between the post-observed errors and the post-observed correct was still not significant.These findings may suggest that the slowing effect and improved accuracy after observed errors do not rely on the same generation mechanism.Except that one can monitor the other's outcomes,the other also can monitor the individual's outcomes in the social life.Previous studies have demonstrated that observed errors influence individual's performance.However,ta date,there is no study investigating whether and how the other's monitoring influence individual performance.In addition,the issue as to whether social context modulates the post-error adjustment for females and males differently should be focused.To explore the above issues,we employed a color Flanker task in two experiment(study 4).In Experiment 6,we instructed the participant to interact with a same-sex or opposite-sex partner to perform the task together.They needed to sit next to each other and shared a computer.One participant was responsible for performing the Flanker task(performer),and the other was responsible for monitoring the error responses of the performer(monitor).They alternated their roles after completing one block.The results showed that the post-error slowing was significantly larger in females than in males in the interaction context.In particular,the sex difference of post-error slowing mainly benefited from the opposite-sex interaction.In Experiment 7,to verify the role of interaction context,we conducted this control experiment in the individual context.The result showed that the post-error slowing between females and males was not significant in the individual context.The contrast analysis of results between experiment 6 and experiment 7 showed the post-error slowing effect was significantly larger in the interaction context than in the individual context.Moreover,the females' performance was influenced by partner,no matter whether the observer was same-sex or opposite-sex,whereas the males' performance was influenced merely by the opposite-sex observer.These findings may suggest that interaction context influences the degree of post-error slowing and modulated the females and males differently.Moreover,females are more sensitive to the social information,and thus they are more susceptible by the task-irrelevant information.In sum,we investigated the generation mechanism of post-error adjustment and the influencing factors by employing various technology.We first utilized the resting-state f MRI technology to investigate the basic brain activity of post-error adjustment and found that bilateral insula was involved in the post-error adjustment.This result demonstrated that the ACC alone cannot fulfill the post-error adjustment,the error processing is the combined effects of multiple regions.In the error awareness study,although the ERN was not different between aware errors and unaware errors,the activity in the theta-ERN and theta band was significantly larger in aware errors than unaware errors,suggesting the early stage following errors was engaged in the preparation of error awareness processing.All these findings provide novel evidences for the deeper understanding of post-error adjustment.In addition,the observed errors and interaction context contents social factors,which is conductive to increase the ecological validity of post-error adjustment and supplement the exist theories of post-error adjustment.We can set relatively high error rate condition without increased task difficulty in the context of observed errors,which can help to investigate the role of error information and frequency information in the post-error adjustment.Moreover,we found the post-error slowing occurred in distinct tasksets,suggesting the adjustment induced by the observed error was generic.We attempted to investigate the post-error slowing in the interaction context and found the post-error slowing effect was significantly larger in females than inmales.Moreover,females' performance could be influenced by the observer,no matter whether the observer was same-sex or opposite-sex,whereas the males' performance was influenced merely by the opposite-sex observer.These findings suggest females are sensitive to the social information and are susceptible to be influenced by the task-irrelevant information.However,male's performance is merely influenced by the information of opposite-sex.