Testing theories of post-error slowing

Abstract task sequence initiation deficit dissociates anxiety disorders from obsessive-compulsive disorder and healthy controls

In everyday life, humans perform sequences of tasks. These tasks may be disrupted in people with obsessive-compulsive disorder (OCD). Symptoms, such as compulsions, can be considered sequential and often cause repetitions of tasks that disrupt daily living (e.g., checking the stove while cooking). Motor sequences have been used to study behavioral deficits in OCD. However, not all sequences are motor sequences. Some are more "abstract" in that they are composed of a series of tasks (e.g., chopping and stirring) rather than being dependent on individual actions or stimuli. These abstract task sequences require cognitive control mechanisms for their execution. Although theory has proposed deficits in these sequences in OCD as well, they have not been directly investigated. We tested the hypotheses that OCD participants exhibit deficits in the control mechanisms specific to abstract task sequences and more general flexible behavior (measured with task switching within the sequences), relative to health controls (HCs) and clinical controls (participants with anxiety disorders [ANX]). A total of 112 participants completed abstract task sequences consisting of simple categorization tasks. Surprisingly, participants with OCD did not perform worse than HCs or ANX. However, ANX participants showed impairments specific to sequential control that did not extend to more general flexible control. Thus, we showed a novel behavioral dissociation between OCD and ANX specific to abstract task sequential control. These results also implicate deficits in specific frontal sequential control neural circuitry in ANX and not in OCD, where implicit sequential deficits may more closely align with striatal circuits.

The Effect of 6 weeks Physical Activity on Cognitive Control and Trait Impulsivity in Multi-problem Young Adults: First Findings of an RCT-study

This study aimed to report the effect of a 6-week light-active versus moderate-active physical activity intervention embedded in a multimodal day treatment program on selected measures of cognitive control (i.e., response inhibition, error processing, and cognitive interference) and trait impulsivity. A randomized controlled design was implemented, including male multi-problem young adults (aged 18-27) assigned to either light-active (N = 12) or moderate-active physical activity lessons (N = 11). A repeated measures design was used to examine treatment effects between the two groups over time on response inhibition, error processing, and cognitive interference (measured respectively with a Go/NoGo task, a Flanker task, and the Stroop) and trait impulsivity (measured with the Dutch Baratt Impulsiveness Scale). Cognitive control, but not trait impulsivity, improved over time. Specifically, enhancements in inhibition and reduced cognitive interference were observed after 6 weeks. Error processing did not improve, but we did observe improved performance on an error-processing task. No interaction with physical activity intensity was found, suggesting similar treatment effects regardless of intensity. Results should be interpreted with caution due to several limitations, including the small sample size. Overall, due to current limitations (i.e., physical activity embedded in a larger treatment program, small sample size at follow-up, and low intervention adherence), it is not possible to draw any definite conclusions. However, the current findings contribute to a growing body of evidence suggesting potential benefits of physical activity (embedded in a multi-modal day treatment program) in the enhancement of cognitive control deficits in at-risk populations, independent of exercise intensity.

Event-related brain potentials to typing errors in transparent and intransparent German words

Error detection in typing is crucial for assessing the adequacy of ongoing actions, leveraging both predictive mechanisms for early detection and sensory feedback for late detection. Neurophysiological studies have supported the anticipation of errors through predictive models. This research extends the understanding of error detection in typing, focusing on the neurocognitive mechanisms underlying errors in transparent and intransparent German words. Thirty-six volunteer students typed out aurally presented words, classified as either orthographically transparent or intransparent, on a computer keyboard without the possibility of correction. Because of poor spelling or excessive artifacts, the final sample comprised 27 participants. Event-related potentials (ERPs) were obtained time-locked to key presses, and behavioral data on typing correctness and speed were collected. A higher error rate and longer latency for intransparent words compared to transparent ones were found, suggesting the complexity of spelling impacts typing correctness. Post-error slowing was observed, aligning with increased cognitive control following errors. ERPs revealed a negative component akin to the error-related negativity (ERN) for typing errors, with a pronounced ERN-like negativity preceding erroneous key-presses, particularly for intransparent words. The study provides evidence of the cognitive and neural mechanisms underlying typing errors, highlighting the impact of orthographic transparency. The detection of an ERN-like negativity before erroneous key-presses, especially in typing intransparent words, underscores the brain's use of predictive mechanisms for error detection.

Task-specific relationships between error-related ERPs and behavior: Flanker, Stroop, and Go/Nogo tasks

Performance monitoring has been widely studied during different forced-choice response tasks. Participants typically show longer response times (RTs) and increased accuracy following errors, but there are inconsistencies regarding the connection between error-related event-related brain potentials (ERPs) and behavior, such as RT and accuracy. The specific task in any given study could contribute to these inconsistencies, as different tasks may require distinct cognitive processes that impact ERP-behavior relationships. The present study sought to determine whether task moderates ERP-behavior relationships and whether these relationships are robustly observed when tasks and stimuli are treated as random effects. ERPs and behavioral indices (RTs and accuracy) recorded during flanker, Stroop, and Go/Nogo tasks from 180 people demonstrated a task-specific effect on ERP-behavior relationships, such that larger previous-trial error-related negativity (ERN) predicted longer RTs and greater likelihood of a correct response on subsequent trials during flanker and Stroop tasks but not during Go/Nogo task. Additionally, larger previous-trial error positivity (Pe) predicted faster RTs and smaller variances of RTs on subsequent trials for Stroop and Go/Nogo tasks but not for flanker task. When tasks and stimuli were treated as random effects, ERP-behavior relationships were not observed. These findings support the need to consider the task used for recording performance monitoring measures when interpreting results across studies.

A Diffusion Model Account of Cognitive Variability in Healthy Aging and Mild Cognitive Impairment

Within-person variation in cognitive performance is linked to pathological aging. Cognitive fluctuations have not been analyzed using cognitive process models, such as the diffusion model, to characterize which cognitive processes contribute to variability in cognition. We collected 21 daily assessments of attention and personality in younger adults, healthy older adults, and those with mild cognitive impairment. We employed mixed-effects location scale models to analyze group differences on mean diffusion parameters and daily variability. Discussion focuses on how these methods extend our understanding of how cognitive deficits might appear in aging and disease and the moderating influence of daily personality.

A theoretical perspective on action consequences in action imagery: internal prediction as an essential mechanism to detect errors

Acting in the environment results in both intended and unintended consequences. Action consequences provide feedback about the adequacy of actions while they are in progress and when they are completed and therefore contribute to monitoring actions, facilitate error detection, and are crucial for motor learning. In action imagery, no actual action takes place, and consequently, no actual action consequences are produced. However, imagined action consequences may replace actual action consequences, serving a similar function and facilitating performance improvements akin to that occurring with actual actions. In this paper, we conceptualize action imagery as a simulation based on internal models. During that simulation, forward models predict action consequences. A comparison of predicted and intended action consequences sometimes indicates the occurrence of action errors (or deviations from optimal performance) in action imagery. We review research indicating that action errors are indeed sometimes imagined in action imagery. These results are compatible with the view that action imagery is based on motor simulation but incompatible with the view that action imagery is solely based on abstract knowledge. The outlined framework seems suitable to cover a wide range of action imagery phenomena and can explain action imagery practice effects.

Recording Neural Reward Signals in a Naturalistic Operant Task Using Mobile-EEG and Augmented Reality

The electrophysiological response to rewards recorded during laboratory tasks has been well documented, yet little is known about the neural response patterns in a more naturalistic setting. Here, we combined a mobile-EEG system with an augmented reality headset to record event-related brain potentials (ERPs) while participants engaged in a naturalistic operant task to find rewards. Twenty-five participants were asked to navigate toward a west or east goal location marked by floating orbs, and once participants reached the goal location, the orb would then signify a reward (5 cents) or no-reward (0 cents) outcome. Following the outcome, participants returned to a start location marked by floating purple rings, and once standing in the middle, a 3 s counter signaled the next trial, for a total of 200 trials. Consistent with previous research, reward feedback evoked the reward positivity, an ERP component believed to index the sensitivity of the anterior cingulate cortex to reward prediction error signals. The reward positivity peaked ∼230 ms with a maximal at channel FCz (M = -0.695 μV, ±0.23) and was significantly different than zero (p < 0.01). Participants took ∼3.38 s to reach the goal location and exhibited a general lose-shift (68.3% ±3.5) response strategy and posterror slowing. Overall, these novel findings provide support for the idea that combining mobile-EEG with augmented reality technology is a feasible solution to enhance the ecological validity of human electrophysiological studies of goal-directed behavior and a step toward a new era of human cognitive neuroscience research that blurs the line between laboratory and reality.

People are more error-prone after committing an error

Humans tend to slow down after making an error. A longstanding account of this post-error slowing is that people are simply more cautious. However, accuracy typically does not improve following an error, leading some researchers to suggest that an initial 'orienting' response may initially impair performance immediately following error. Unfortunately, characterizing the nature of this error-based impairment remains a challenge in standard tasks that use free response times. By exerting control over the timing of responses, we reveal the time course of stimulus-response processing. Participants are less accurate after an error even when given ample time to make a response. A computational model of response preparation rules out the possibility that errors lead to slower cognitive processing. Instead, we find that the efficacy of cognitive processing in producing an intended response is impaired following errors. Following an error, participants commit more slips of action that tend to be a repetition of the previous mistake. Rather than a strategic shift along a single speed-accuracy tradeoff function, post-error slowing observed in free response time tasks may be an adaptive response to impaired cognitive processing that reflects an altered relationship between the speed and accuracy of responses.

Motor inhibition errors and interference suppression errors differ systematically on neural and behavioural features of response monitoring

Action inhibition and error commission are prominent in everyday life. Inhibition comprises at least two facets: motor inhibition and interference suppression. When motor inhibition fails, a strong response impulse cannot be inhibited. When interference suppression fails, we become distracted by irrelevant stimuli. We investigated the neural and behavioural similarities and differences between motor inhibition errors and interference suppression errors systematically from stimulus-onset to post-response adaptation. To enable a direct comparison between both error types, we developed a complex speeded choice task where we assessed the error types in two perceptually similar conditions. Comparing the error types along the processing stream showed that the P2, an early component in the event-related potential associated with sensory gating, is the first marker for differences between the two error types. Further error-specific variations were found for the parietal P3 (associated with context updating and attentional resource allocation), for the lateralized readiness potential (LRP, associated with primary motor cortex activity), and for the Pe (associated with error evidence accumulation). For motor inhibition errors, the P2, P3 and Pe tended to be enhanced compared to successful inhibition. The LRP for motor inhibition errors was marked by multiple small response impulses. For interference suppression errors, all components were more similar to those of successful inhibition. Together, these findings suggest that motor inhibition errors arise from a deficient early inhibitory process at the perceptual and motor level, and become more apparent than interference suppression errors, that arise from an impeded response selection process.

Neural and Behavioral Measures of Stress-induced Impairment in Error Awareness and Post-error Adjustment

Exposure to stress negatively affects error processing, but the impact of stress on error awareness remains to be determined. In the present study, we examined the temporal dynamics of error awareness and post-error adjustment following acute stress. Forty-nine healthy men were randomly assigned to the control (n = 26) or stress group (n = 23). After stress induction, participants completed the error awareness task, and their brain activity was assessed by electroencephalography. Compared to the control group, the stress group demonstrated lower error awareness accuracy and smaller Pe (error positivity) and ΔPe amplitudes following aware error responses, which indicated impairment of error awareness following stress. Furthermore, the stress group had lower accuracy in post-aware error responses than in post-unaware error responses and the control group, which indicated poor post-error adjustment following stress. Our results showed a stress effect on sequential stages of error processing. Stress induces impaired error identification, which further generates maladaptive post-error performance.

Finding Pattern in the Noise: Persistent Implicit Statistical Knowledge Impacts the Processing of Unpredictable Stimuli

Humans can extract statistical regularities of the environment to predict upcoming events. Previous research recognized that implicitly acquired statistical knowledge remained persistent and continued to influence behavior even when the regularities were no longer present in the environment. Here, in an fMRI experiment, we investigated how the persistence of statistical knowledge is represented in the brain. Participants (n = 32) completed a visual, four-choice, RT task consisting of statistical regularities. Two types of blocks constantly alternated with one another throughout the task: predictable statistical regularities in one block type and unpredictable ones in the other. Participants were unaware of the statistical regularities and their changing distribution across the blocks. Yet, they acquired the statistical regularities and showed significant statistical knowledge at the behavioral level not only in the predictable blocks but also in the unpredictable ones, albeit to a smaller extent. Brain activity in a range of cortical and subcortical areas, including early visual cortex, the insula, the right inferior frontal gyrus, and the right globus pallidus/putamen contributed to the acquisition of statistical regularities. The right insula, inferior frontal gyrus, and hippocampus as well as the bilateral angular gyrus seemed to play a role in maintaining this statistical knowledge. The results altogether suggest that statistical knowledge could be exploited in a relevant, predictable context as well as transmitted to and retrieved in an irrelevant context without a predictable structure.

Error modulates categorization of subsecond durations in multitasking contexts

Monitoring errors consumes limited cognitive resources and can disrupt subsequent task performance in multitasking scenarios. However, there is a dearth of empirical evidence concerning this interference with prospective estimation of time. In this study, we sought to investigate this issue through a serial multitasking experiment, employing a temporal bisection task as the primary task. We introduced two task contexts by implementing two different concurrent tasks. In one context, participants were tasked with discriminating the size difference between two visual items, while in the other context, they were required to judge the temporal order of similar visual items. The primary task remained the same for the entire experiment. Psychophysical metrics, including subjective bias (determined by the bisection point) and temporal sensitivity (measured by the Weber ratio), in addition to reaction time, remained unaltered in the primary task regardless of the perceptual context exerted by the concurrent tasks. However, commission of error in the concurrent tasks (i.e., non-specific errors) led to a right-ward shift in the bisection point, indicating underestimation of time after errors. Applying a drift-diffusion framework for temporal decision making, we observed alterations in the starting point and drift rate parameters, supporting the error-induced underestimation of time. The error-induced effects were all diminished with increasing a delay between the primary and concurrent task, indicating an adaptive response to errors at a trial level. Furthermore, the error-induced shift in the bisection point was diminished in the second half of the experiment, probably because of a decline in error significance and subsequent monitoring response. These findings indicate that non-specific errors impact the prospective estimation of time in multitasking scenarios, yet their effects can be alleviated through both local and global reallocation of cognitive resources from error processing to time processing.

Task-general or specific: The alertness modulates post-error adjustment

Previous studies have shown that alertness is closely related to executive control function, but its impact on components of post-error adjustment is unknown. This study applied the Attentional Networks Test and the Four-choice Flanker task with three response stimulus intervals (RSIs) to explore the correlation between alertness and post-error adjustment. The linear mixed-effects model of alertness and RSI on the post-error processing indicators showed a significant negative correlation between the alertness and post-error slowing (PES) under 200 ms RSI , as well as between alertness and post-error improvement in accuracy (PIA) under both 700 ms RSI and 1200 ms RSI. Participants with lower alertness showed larger post-error slowing in the early stages, while those with higher alertness had smaller PIA in later stages. This study revealed the effects of alertness on different processing components of post-error adjustment. The control strategies utilized by individuals with high and low levels of alertness differed in preparation for performance monitoring. Alertness improved post-error response speed in a task-unspecific manner, but not post-error adaptation.

Rethinking Norm Psychology

Norms permeate human life. Most of people's activities can be characterized by rules about what is appropriate, allowed, required, or forbidden-rules that are crucial in making people hyper-cooperative animals. In this article, I examine the current cognitive-evolutionary account of "norm psychology" and propose an alternative that is better supported by evidence and better placed to promote interdisciplinary dialogue. The incumbent theory focuses on rules and claims that humans genetically inherit cognitive and motivational mechanisms specialized for processing these rules. The cultural-evolutionary alternative defines normativity in relation to behavior-compliance, enforcement, and commentary-and suggests that it depends on implicit and explicit processes. The implicit processes are genetically inherited and domain-general; rather than being specialized for normativity, they do many jobs in many species. The explicit processes are culturally inherited and domain-specific; they are constructed from mentalizing and reasoning by social interaction in childhood. The cultural-evolutionary, or "cognitive gadget," perspective suggests that people alive today-parents, educators, elders, politicians, lawyers-have more responsibility for sustaining normativity than the nativist view implies. People's actions not only shape and transmit the rules, but they also create in each new generation mental processes that can grasp the rules and put them into action.

Need for Cognition is associated with a preference for higher task load in effort discounting

When individuals set goals, they consider the subjective value (SV) of the anticipated reward and the required effort, a trade-off that is of great interest to psychological research. One approach to quantify the SVs of levels of difficulty of a cognitive task is the Cognitive Effort Discounting Paradigm by Westbrook and colleagues (2013). However, it fails to acknowledge the highly individual nature of effort, as it assumes a unidirectional, inverse relationship between task load and SVs. Therefore, it cannot map differences in effort perception that arise from traits like Need for Cognition, since individuals who enjoy effortful cognitive activities likely do not prefer the easiest level. We replicated the analysis of Westbrook and colleagues with an adapted version, the Cognitive and Affective Discounting (CAD) Paradigm. It quantifies SVs without assuming that the easiest level is preferred, thereby enabling the assessment of SVs for tasks without objective order of task load. Results show that many of the 116 participants preferred a more or the most difficult level. Variance in SVs was best explained by a declining logistic contrast of the [Formula: see text]-back levels and by the accuracy of responses, while reaction time as a predictor was highly volatile depending on the preprocessing pipeline. Participants with higher Need for Cognition scores perceived higher [Formula: see text]-back levels as less effortful and found them less aversive. Effects of Need for Cognition on SVs in lower levels did not reach significance, as group differences only emerged in higher levels. The CAD Paradigm appears to be well suited for assessing and analysing task preferences independent of the supposed objective task difficulty.Protocol registrationThe stage 1 protocol for this Registered Report was accepted in principle on August 19, 2022. The protocol, as accepted by the journal, can be found at: https://doi.org/10.17605/OSF.IO/CPXTH .

Feeling and deciding: Subjective experiences rather than objective factors drive the decision to invest cognitive control

When presented with the choice to invest cognitive control in a task, several signals are monitored to reach a decision. Leading theoretical frameworks argued that the investment of cognitive control is determined by a cost-benefit computation. However, previous accounts remained silent on the potential role of subjective experience in this computation. We experience confidence when giving an answer, feel the excitement of an anticipated reward, and reflect on how much effort is required for successful task performance. Two questions are investigated in the present work: how objective task parameters give rise to subjective experience and whether these drive the decision to allocate cognitive control. To this end, we designed a task in which we manipulated three objective parameters in the same sequence of events (stimulus uncertainty, physical effort, and reward prediction error). We asked participants to report their subjective experiences associated with these manipulations: confidence, subjective physical effort, and reward satisfaction. At the end of each trial, participants indicated whether they wanted to repeat that trial on the next day. In response to the first question, we demonstrate that subjective ratings are reliable and selective. Subjective experiences closely mirrored their objective manipulations. In response to the second question, we demonstrate that subjective experiences provide a better fit for the decisions on future control investments. While objective task parameters are considered when deciding, they do not always produce the expected changes in subjective experience, and when dissociations occur, it is the subjective experience that better explains the decision to allocate cognitive control.

Examining post-error performance in a complex multitasking environment

Previous work on indices of error-monitoring strongly supports that errors are distracting and can deplete attentional resources. In this study, we use an ecologically valid multitasking paradigm to test post-error behavior. It was predicted that after failing an initial task, a subject re-presented with that task in conflict with another competing simultaneous task, would more likely miss their response opportunity for the competing task and stay 'tunneled' on the initially errored task. Additionally, we predicted that an error's effect on attention would dissipate after several seconds, making error cascades less likely when subsequent conflict tasks are delayed. A multi-attribute task battery was used to present tasks and collect measures of both post-error and post-correct performance. Results supported both predictions: post-error accuracy on the competing task was lower compared to post-correct accuracy, and error-proportions were higher at shorter delays, dissipating over time. An exploratory analysis also demonstrated that following errors (as opposed to post-correct trials), participants clicked more on the task panel of the initial error regardless of delay; this continued task-engagement provides preliminary support for errors leading to a cognitive tunneling effect.

Recording neural reward signals in the real-world using mobile-EEG and augmented reality

The electrophysiological response to rewards recorded during laboratory-based tasks has been well documented over the past two decades, yet little is known about the neural response patterns in 'real-world' settings. To address this issue, we combined a mobile-EEG system with an augmented reality headset (which blends high definition "holograms" within the real-world) to record event-related brain potentials (ERP) while participants navigated an operant chamber to find rewards. 25 participants (age = 18-43, Male=6, Female=19) were asked to choose between two floating holograms marking a west or east goal-location in a large room, and once participants reached the goal location, the hologram would turn into a reward (5 cents) or no-reward (0 cents) cue. Following the feedback cue, participants were required to return to a hologram marking the start location, and once standing in it, a 3 second counter hologram would initiate the next trial. This sequence was repeated until participants completed 200 trials. Consistent with previous research, reward feedback evoked the reward positivity, an ERP component believed to index the sensitivity of the anterior cingulate cortex to reward prediction error signals. The reward positivity peaked around 235ms post-feedback with a maximal at channel FCz (M=-2.60μV, SD=1.73μV) and was significantly different than zero (p < 0.01). At a behavioral level, participants took approximately 3.38 seconds to reach the goal-location and exhibited a general lose-shift (68.3% ± 3.5) response strategy and were slightly slower to return to the start location following negative feedback (2.43 sec) compared to positive feedback (2.38 sec), evidence of post-error slowing. Overall, these findings provide the first evidence that combining mobile-EEG with augmented reality technology is a feasible solution to enhance the ecological validity of human electrophysiological studies of goal-directed behavior and a step towards a new era of human cognitive neuroscience research that blurs the line between laboratory and reality.

Prepotent response inhibition in autism: Not an inhibitory deficit?

Research outcomes on prepotent response inhibition in neurodevelopmental conditions during adulthood seem inconsistent, especially in autism. To gain further insight in these inconsistencies, the current study investigates inhibitory performance, as well as task strategies such as adaptive behavior during inhibitory tasks in autistic adults. As Attention-Deficit/Hyperactivity Disorder (ADHD) is often co-occurring in autism and associated with differences in both inhibition and adaptation, the role of ADHD symptoms is explored. Additionally, prior research is extended to middle- and late-adulthood, and the role of cognitive aging is assessed. Hundred-and-five autistic adults and 139 non-autistic adults (age: 20-80 yrs) were compared on a Go-NoGo task. No significant group differences in inhibitory difficulties (commission errors) or adaptation (post error slowing) were observed, and both did not relate significantly to ADHD symptoms. However, when controlling for reaction time autistic individuals made significantly more inhibitory errors than non-autistic individuals, yet the effect size was modest (Cohen's d = .27). Exploratory analyses showed that adaption significantly related to inhibition in non-autistic individuals only, possibly hinting at altered adaptive behavior during inhibitory tasks in autistic adults. ADHD symptoms related to response variability in the autism group only. Furthermore, task strategy changed with older age in both groups, with slower and more cautious responses at older age. Taken together, although minor differences may exist, autistic and non-autistic people show largely similar patterns of inhibitory behavior throughout adulthood. Differences in task timing and strategy seem relevant for future longitudinal studies on cognitive aging across neurodevelopmental conditions.

Both reactive and proactive control are deficient in children with ADHD and predictive of clinical symptoms

Cognitive control deficits are a hallmark of attention deficit hyperactivity disorder (ADHD) in children. Theoretical models posit that cognitive control involves reactive and proactive control processes but their distinct roles and inter-relations in ADHD are not known, and the contributions of proactive control remain vastly understudied. Here, we investigate the dynamic dual cognitive control mechanisms associated with both proactive and reactive control in 50 children with ADHD (16F/34M) and 30 typically developing (TD) children (14F/16M) aged 9-12 years across two different cognitive controls tasks using a within-subject design. We found that while TD children were capable of proactively adapting their response strategies, children with ADHD demonstrated significant deficits in implementing proactive control strategies associated with error monitoring and trial history. Children with ADHD also showed weaker reactive control than TD children, and this finding was replicated across tasks. Furthermore, while proactive and reactive control functions were correlated in TD children, such coordination between the cognitive control mechanisms was not present in children with ADHD. Finally, both reactive and proactive control functions were associated with behavioral problems in ADHD, and multi-dimensional features derived from the dynamic dual cognitive control framework predicted inattention and hyperactivity/impulsivity clinical symptoms. Our findings demonstrate that ADHD in children is characterized by deficits in both proactive and reactive control, and suggest that multi-componential cognitive control measures can serve as robust predictors of clinical symptoms.

Updating the relationship of the Ne/ERN to task-related behavior: A brief review and suggestions for future research

The error negativity/error-related negativity (Ne/ERN) is one of the most well-studied event-related potential (ERP) components in the electroencephalography (EEG) literature. Peaking about 50 ms after the commission of an error, the Ne/ERN is a negative deflection in the ERP waveform that is thought to reflect error processing in the brain. While its relationships to trait constructs such as anxiety are well-documented, there is still little known about how the Ne/ERN may subsequently influence task-related behavior. In other words, does the occurrence of the Ne/ERN trigger any sort of error corrective process, or any other behavioral adaptation to avoid errors? Several theories have emerged to explain how the Ne/ERN may implement or affect behavior on a task, but evidence supporting each has been mixed. In the following manuscript, we review these theories, and then systematically discuss the reasons that there may be discrepancies in the literature. We review both the inherent biological factors of the neural regions that underlie error-processing in the brain, and some of the researcher-induced factors in analytic and experimental choices that may be exacerbating these discrepancies. We end with a table of recommendations for future researchers who aim to understand the relationship between the Ne/ERN and behavior.

EEG Dynamics of Error Processing and Associated Behavioral Adjustments in Preschool Children

Preschool children show neural responses and make behavioral adjustments immediately following an error. However, there is a lack of evidence regarding how neural responses to error predict subsequent behavioral adjustments during childhood. The aim of our study was to explore the neural dynamics of error processing and associated behavioral adjustments in preschool children from unsatisfied basic needs (UBN) homes. Using EEG recordings during a go/no-go task, we examined within-subject associations between the error-related negativity (ERN), frontal theta power, post-error slowing, and post-error accuracy. Post-error accuracy increased linearly with post-error slowing, and there was no association between the neural activity of error processing and post-error accuracy. However, during successful error recovery, the frontal theta power, but not the ERN amplitude, was associated positively with post-error slowing. These findings indicated that preschool children from UBN homes adjusted their behavior following an error in an adaptive form and that the error-related theta activity may be associated with the adaptive forms of post-error behavior. Furthermore, our data support the adaptive theory of post-error slowing and point to some degree of separation between the neural mechanisms represented by the ERN and theta.

Neurophysiological mechanisms of error monitoring in human and non-human primates

Performance monitoring is an important executive function that allows us to gain insight into our own behaviour. This remarkable ability relies on the frontal cortex, and its impairment is an aspect of many psychiatric diseases. In recent years, recordings from the macaque and human medial frontal cortex have offered a detailed understanding of the neurophysiological substrate that underlies performance monitoring. Here we review the discovery of single-neuron correlates of error monitoring, a key aspect of performance monitoring, in both species. These neurons are the generators of the error-related negativity, which is a non-invasive biomarker that indexes error detection. We evaluate a set of tasks that allows the synergistic elucidation of the mechanisms of cognitive control across the two species, consider differences in brain anatomy and testing conditions across species, and describe the clinical relevance of these findings for understanding psychopathology. Last, we integrate the body of experimental facts into a theoretical framework that offers a new perspective on how error signals are computed in both species and makes novel, testable predictions.

Hearts, flowers, and fruits: All children need to reveal their post-error slowing

Slowing down responses after errors (i.e., post-error slowing [PES]) is an established finding in adults. Yet PES in young children is still not well understood. In this study, we investigated (a) whether young children show PES in tasks with different types of cognitive conflict and differing demands on executive functions, (b) whether PES is adaptive and efficient in the sense that it is associated with better task performance, and (c) whether PES correlates between tasks. We tested 4- to 6-year-old children on the Funny Fruits task (FF; n = 143), a Stroop-like task that incorporates semantic conflict and taxes children's inhibition skills, and the Hearts and Flowers task (HF; n = 170), which incorporates spatial conflict and taxes children's inhibition skills in its incongruent block and taxes both inhibition and cognitive flexibility (rule-switching) skills in its mixed block. A subgroup of children were tested on both FF and HF (n = 74). Results revealed that, first, children showed PES in FF and both blocks of HF, indicating that PES occurs in both types of conflict and under varying executive demands. Second, PES was associated with task accuracy, but only for FF and the mixed HF. Third, a between-task association in PES emerged only between FF and the mixed HF. Together, these findings indicate that PES is still a developing strategy in young children; it is present but only adaptive for, and correlates between, semantic inhibition and spatial flexibility.

Endorsement and embodiment of cautiousness-related age stereotypes

Endorsement of implicit age stereotypes was assessed with the propositional evaluation paradigm (PEP) in a high-powered, preregistered study, comprising samples of young (n = 89) and older (n = 125) adults. To investigate whether implicit age stereotypes shape the behavior via self-stereotyping ("embodiment"), we examined whether implicit endorsement of the belief of older (young) people being cautious (reckless) predicts older (young) individuals' spontaneous behavior in a speeded response time task. In both age groups, we found significant implicit endorsement effects of age stereotypical beliefs. However, implicit endorsement effects of the cautiousness-related age stereotypes were unrelated to our indicators of spontaneous cautious/reckless behavior in the speeded RT task (as assessed with the parameter a of a diffusion model analysis) for both age groups. The same pattern of results (endorsement of age stereotypic beliefs but no relation with behavioral indicators) was found for explicit measures of age stereotypes. Replicating previous findings, implicit and explicit measures of cautiousness-related age stereotypes were uncorrelated. In sum, our findings provide evidence for the implicit and explicit endorsement of cautiousness-related stereotypical beliefs about old and young people; individual differences in belief endorsement, however, did not predict differences in spontaneous cautiousness-related behavior in a speeded RT task.

Imbalanced weighting of proactive and reactive control as a marker of risk-taking propensity

According to the dual mechanisms of control (DMC), reactive and proactive control are involved in adjusting behaviors when maladapted to the environment. However, both contextual and inter-individual factors increase the weight of one control mechanism over the other, by influencing their cognitive costs. According to one of the DMC postulates, limited reactive control capacities should be counterbalanced by greater proactive control to ensure control efficiency. Moreover, as the flexible weighting between reactive and proactive control is key for adaptive behaviors, we expected that maladaptive behaviors, such as risk-taking, would be characterized by an absence of such counterbalance. However, to our knowledge, no studies have yet investigated this postulate. In the current study, we analyzed the performances of 176 participants on two reaction time tasks (Simon and Stop Signal tasks) and a risk-taking assessment (Balloon Analog Risk Taking, BART). The post-error slowing in the Simon task was used to reflect the spontaneous individuals' tendency to proactively adjust behaviors after an error. The Stop Signal Reaction Time was used to assess reactive inhibition capacities and the duration of the button press in the BART was used as an index of risk-taking propensity. Results showed that poorer reactive inhibition capacities predicted greater proactive adjustments after an error. Furthermore, the higher the risk-taking propensity, the less reactive inhibition capacities predicted proactive behavioral adjustments. The reported results suggest that higher risk-taking is associated with a smaller weighting of proactive control in response to limited reactive inhibition capacities. These findings highlight the importance of considering the imbalanced weighting of reactive and proactive control in the analysis of risk-taking, and in a broader sense, maladaptive behaviors.

The role of affective interference and mnemonic load in the dynamic adjustment in working memory

The amount of cognitive control required during a task may fluctuate over the course of performing a task. The dynamic upregulation of cognitive control has been proposed to occur in response to conflict or in response to the need for additional control during demanding cognitive tasks. Specifically, upregulation in cognitive control results in improved performance on trials that follow more demanding trials. Recent work has demonstrated that upregulation occurs during following trials with high (vs. low) mnemonic load and negative (vs. neutral) affective interference during working memory tasks. Although dynamic upregulation appears to be a robust phenomenon, less is known about individual difference factors that may alter the likelihood to engage in upregulation as a result of a signal to increase cognitive control. The current study attempted to replicate prior findings that suggest upregulation may occur following higher load trials or affective interference during a working memory task. Further, the study examined anxiety, depressive symptomatology, working memory capacity, mood, and dispositional mindfulness and possible moderators for upregulation of cognitive control. Participants (N = 150) completed a delayed recognition working memory task with mnemonic load (High vs. Low) and affective interference (Negative vs. Neutral) parametrically manipulated. Participants completed measures of the individual difference factors. The current findings replicate prior work demonstrating an upregulation of cognitive control following high load trials and negative affective interference. Individual difference factors did not moderate the upregulation findings, suggesting that upregulation is a robust phenomenon that can be triggered by both affective interference and mnemonic load.

What is left after an error? Towards a comprehensive account of goal-based binding and retrieval

The cognitive system readily detects and corrects erroneous actions by establishing episodic bindings between representations of the acted upon stimuli and the intended correct response. If these stimuli are encountered again, they trigger the retrieval of the correct response. Thus, binding and retrieval efficiently pave the way for future success. The current study set out to define the role of the erroneous response itself and explicit feedback for the error during these processes of goal-based binding and retrieval. Two experiments showed robust and similar binding and retrieval effects with and without feedback and pointed towards sustained activation of the unbound, erroneous response. The third experiment confirmed that the erroneous response is more readily available than a neutral alternative. Together, the results demonstrate that episodic binding biases future actions toward success, guided primarily through internal feedback processes, while the erroneous response still leaves detectable traces in human action control.

Capturing Dynamic Performance in a Cognitive Model: Estimating ACT-R Memory Parameters With the Linear Ballistic Accumulator

The parameters governing our behavior are in constant flux. Accurately capturing these dynamics in cognitive models poses a challenge to modelers. Here, we demonstrate a mapping of ACT-R's declarative memory onto the linear ballistic accumulator (LBA), a mathematical model describing a competition between evidence accumulation processes. We show that this mapping provides a method for inferring individual ACT-R parameters without requiring the modeler to build and fit an entire ACT-R model. Existing parameter estimation methods for the LBA can be used, instead of the computationally expensive parameter sweeps that are traditionally done. We conduct a parameter recovery study to confirm that the LBA can recover ACT-R parameters from simulated data. Then, as a proof of concept, we use the LBA to estimate ACT-R parameters from an empirical dataset. The resulting parameter estimates provide a cognitively meaningful explanation for observed differences in behavior over time and between individuals. In addition, we find that the mapping between ACT-R and LBA lends a more concrete interpretation to ACT-R's latency factor parameter, namely as a measure of response caution. This work contributes to a growing movement towards integrating formal modeling approaches in cognitive science.

Two Types of Motor Inhibition after Action Errors in Humans

Adaptive behavior requires the ability to appropriately react to action errors. Post-error slowing (PES) of response times is one of the most reliable phenomena in human behavior. It has been proposed that PES is partially achieved through inhibition of the motor system. However, there is no direct evidence for this link, or indeed, that the motor system is physiologically inhibited after errors altogether. Here, we used transcranial magnetic stimulation and electromyography to measure corticospinal excitability (CSE) across four experiments using a Simon task, in which female and male human participants sometimes committed errors. Errors were followed by reduced CSE at two different time points and in two different modes. Shortly after error commission (250 ms), CSE was broadly suppressed (i.e., even task-unrelated motor effectors were inhibited). During the preparation of the subsequent response, CSE was specifically reduced at task-relevant effectors only. This latter effect was directly related to PES, with stronger CSE suppression accompanying greater PES. This suggests that PES is achieved through increased inhibitory control during post-error responses. To provide converging evidence, we then reanalyzed an openly available EEG dataset that contained both Simon- and Stop-signal tasks using independent component analysis. We found that the same neural source component that indexed action cancellation in the stop-signal task also showed clear PES-related activity during post-error responses in the Simon task. Together, these findings provide evidence that post-error adaptation is partially achieved through motor inhibition. Moreover, inhibition is engaged in two modes (first nonselective, then selective), aligning with recent multistage theories of error processing.SIGNIFICANCE STATEMENT It is a common observation that humans implement a higher degree of caution when repeating an action during which they just committed a mistake. In the laboratory, such increased "caution" is reflected in post-error slowing of response latencies. Many competing theories exist regarding the precise neural mechanisms underlying post-error slowing. Using transcranial magnetic stimulation, we show that, after error commission, the human corticomotor system is momentarily inhibited, both immediately after an error and during the preparation of the next action. Moreover, motor inhibition during the latter time period is directly predictive of post-error slowing. This shows that inhibitory control is a key mechanism humans engage to regulate their own behavior in the aftermath of error commission.

Reduced Error Recognition Explains Post-Error Slowing Differences among Children with Attention Deficit Hyperactivity Disorder

OBJECTIVE

Youth with attention deficit hyperactivity disorder (ADHD) often show reduced post-error slowing (PES) compared to typically developing controls. This finding has been interpreted as evidence that children with ADHD have error recognition and adaptive control impairments. However, several studies report mixed results regarding PES differences in ADHD, and among healthy controls, there is considerable debate about the cognitive-behavioral origin of PES.

METHODS

We tested competing hypotheses aimed at clarifying whether reduced PES in children with ADHD is due to impaired error detection, deficits in adaptive control, and/or attention orienting to novelty. Children aged 7-11 years with a diagnosis of ADHD (n = 74) and controls (n = 30) completed four laboratory-based computer tasks with variable cognitive loads and error types.

RESULTS

ADHD diagnosis was associated with shorter PES only on a task with high cognitive load and low error-cuing, consistent with impaired error recognition. In contrast, there was no evidence of impaired adaptive control or heightened novelty orienting among children with ADHD.

CONCLUSIONS

The cognitive-behavioral origin of PES is multifactorial, but reduced PES among children with an ADHD diagnosis is due to impaired error recognition during cognitively demanding tasks. Behavioral interventions that scaffold error recognition may facilitate improved performance among children with ADHD.

Modeling Distraction: How Stimulus-driven Attention Capture Influences Goal-directed Behavior

The importance of paying attention to a task at hand is emphasized from an early age and extends throughout life. The costs of attentional focus, however, include the potential to miss important changes in the environment, so some process for monitoring nontask information is essential. In this study, a model of latent cognitive variables was applied to data obtained from a two-alternative forced-choice task where participants identified the longer of two sounds. Using an adaptive procedure task, accuracy was maintained at a higher or lower level creating two difficulties, and the sounds were heard either where frequency changes in the sound were rare or common (oddball and multistandard conditions, respectively). Frequency changes created stimulus-driven "distraction" effects in the oddball sequence only, and cognitive modeling (using the linear ballistic accumulator) attributed these effects to slowed accumulation of evidence about tone length on these trials. Concurrent recording of auditory ERPs revealed these delays in evidence accumulation to be related to the amplitude of N2 or mismatch negativity period and P300 response components. In contrast, the response time on trials after a rare frequency change was associated with increased caution in decision-making. Results support the utility of mapping behavioral and ERP measures of performance to latent cognitive processes that contribute to performance and are consistent with a momentary diversion of resources to evaluate the deviant sound feature and remodel predictions about sound.

Post-error Slowing Reflects the Joint Impact of Adaptive and Maladaptive Processes During Decision Making

Errors and their consequences are typically studied by investigating changes in decision speed and accuracy in trials that follow an error, commonly referred to as “post-error adjustments”. Many studies have reported that subjects slow down following an error, a phenomenon called “post-error slowing” (PES). However, the functional significance of PES is still a matter of debate as it is not always adaptive. That is, it is not always associated with a gain in performance and can even occur with a decline in accuracy. Here, we hypothesized that the nature of PES is influenced by one’s speed-accuracy tradeoff policy, which determines the overall level of choice accuracy in the task at hand. To test this hypothesis, we had subjects performing a task in two distinct contexts (separate days), which either promoted speed (hasty context) or cautiousness (cautious context), allowing us to consider post-error adjustments according to whether subjects performed choices with a low or high accuracy level, respectively. Accordingly, our data indicate that post-error adjustments varied according to the context in which subjects performed the task, with PES being solely significant in the hasty context (low accuracy). In addition, we only observed a gain in performance after errors in a specific trial type, suggesting that post-error adjustments depend on a complex combination of processes that affect the speed of ensuing actions as well as the degree to which such PES comes with a gain in performance.

A little doubt saves many mistakes: Early and late error detection in copy-typing

Based on internal predictions, action-errors can be detected relatively early. Different kinds of sensory feedback further provide information about the occurrence of errors later on. To investigate the mechanisms underlying error detection in copy-typing, ten-finger-typists and hunt-and-peck-typists copy typed with and without visibility of the screen and keyboard. We expected that error detection evolves in slower typing before, during, and after an error. Results showed that more errors were reported with visible screen than with covered screen in both groups underpinning the importance of distal action-effects for error detection. Importantly, ten-finger-typists showed pre-error-slowing in the inter-keystroke-intervals (IKIs) before reported errors, but hunt-and-peck-typists did not. In both groups, error-slowing was observed in the last IKI before both reported and unreported errors. Hence, internal predictions play a role in error detection in both groups, but in ten-finger-typists, internal models may be more precise, leading to earlier error detection. Alternatively, slowing down may increase the probability of detecting errors. Finally, in both groups post-error-slowing indicates that sensory feedback from performing keystrokes contributes to error detection. In conclusion, feedback from distal action-effects (i.e., the screen), movement related feedback, and predictive mechanisms contribute to error detection in typing.

Age-related qualitative differences in post-error cognitive control adjustments

Detecting an error signals the need for increased cognitive control and behavioural adjustments. Considerable development in performance monitoring and cognitive control is evidenced by lower error rates and faster response times in multi-trial executive function tasks with age. Besides these quantitative changes, we were interested in whether qualitative changes in balancing accuracy and speed contribute to developmental progression during elementary school years. We conducted two studies investigating the temporal and developmental trajectories of post-error slowing in three prominent cognitive conflict tasks (Stroop, Simon, and flanker). We instructed children (8-, 10-, and 12-year-old) and adults to respond as fast and as accurately as possible and measured their response times on four trials after correct and incorrect responses to a cognitive conflict. Results revealed that all age groups had longer response times on post-error versus post-correct trials, reflecting post-error slowing. Critically, slowing on the first post-error trial declined with age, suggesting an age-related reduction in the orienting response towards errors. This age effect diminished on subsequent trials, suggesting more fine-tuned cognitive control adjustments with age. Overall, the consistent pattern across tasks suggests an age-related change from a relatively strong orienting response to more balanced cognitive control adaptations.

Cognitive Control as a Multivariate Optimization Problem

A hallmark of adaptation in humans and other animals is our ability to control how we think and behave across different settings. Research has characterized the various forms cognitive control can take-including enhancement of goal-relevant information, suppression of goal-irrelevant information, and overall inhibition of potential responses-and has identified computations and neural circuits that underpin this multitude of control types. Studies have also identified a wide range of situations that elicit adjustments in control allocation (e.g., those eliciting signals indicating an error or increased processing conflict), but the rules governing when a given situation will give rise to a given control adjustment remain poorly understood. Significant progress has recently been made on this front by casting the allocation of control as a decision-making problem. This approach has developed unifying and normative models that prescribe when and how a change in incentives and task demands will result in changes in a given form of control. Despite their successes, these models, and the experiments that have been developed to test them, have yet to face their greatest challenge: deciding how to select among the multiplicity of configurations that control can take at any given time. Here, we will lay out the complexities of the inverse problem inherent to cognitive control allocation, and their close parallels to inverse problems within motor control (e.g., choosing between redundant limb movements). We discuss existing solutions to motor control's inverse problems drawn from optimal control theory, which have proposed that effort costs act to regularize actions and transform motor planning into a well-posed problem. These same principles may help shed light on how our brains optimize over complex control configuration, while providing a new normative perspective on the origins of mental effort.

The influence of error detection and error significance on neural and behavioral correlates of error processing in a complex choice task

Error detection and error significance form essential mechanisms that influence error processing and action adaptation. Error detection often is assessed by an immediate self-evaluation of accuracy. Our study used cognitive neuroscience methods to elucidate whether self-evaluation itself influences error processing by increasing error significance in the context of a complex response selection process. In a novel eight-alternative response task, our participants responded to eight symbol stimuli with eight different response keys and a specific stimulus-response assignment. In the first part of the experiment, the participants merely performed the task. In the second part, they also evaluated their response accuracy on each trial. We replicated variations in early and later stages of error processing and action adaptation as a function of error detection. The additional self-evaluation enhanced error processing on later stages, probably reflecting error evidence accumulation, whereas earlier error monitoring processes were not amplified. Implementing multivariate pattern analysis revealed that self-evaluation influenced brain activity patterns preceding and following the response onset, independent of response accuracy. The classifier successfully differentiated between responses from the self- and the no-self-evaluation condition several hundred milliseconds before response onset. Subsequent exploratory analyses indicated that both self-evaluation and the time on task contributed to these differences in brain activity patterns. This suggests that in addition to its effect on error processing, self-evaluation in a complex choice task seems to have an influence on early and general processing mechanisms (e.g., the quality of attention and stimulus encoding), which is amplified by the time on task.

Confidence-Controlled Hebbian Learning Efficiently Extracts Category Membership From Stimuli Encoded in View of a Categorization Task

In experiments on perceptual decision making, individuals learn a categorization task through trial-and-error protocols. We explore the capacity of a decision-making attractor network to learn a categorization task through reward-based, Hebbian-type modifications of the weights incoming from the stimulus encoding layer. For the latter, we assume a standard layer of a large number of stimulus-specific neurons. Within the general framework of Hebbian learning, we have hypothesized that the learning rate is modulated by the reward at each trial. Surprisingly, we find that when the coding layer has been optimized in view of the categorization task, such reward-modulated Hebbian learning (RMHL) fails to extract efficiently the category membership. In previous work, we showed that the attractor neural networks' nonlinear dynamics accounts for behavioral confidence in sequences of decision trials. Taking advantage of these findings, we propose that learning is controlled by confidence, as computed from the neural activity of the decision-making attractor network. Here we show that this confidence-controlled, reward-based Hebbian learning efficiently extracts categorical information from the optimized coding layer. The proposed learning rule is local and, in contrast to RMHL, does not require storing the average rewards obtained on previous trials. In addition, we find that the confidence-controlled learning rule achieves near-optimal performance. In accordance with this result, we show that the learning rule approximates a gradient descent method on a maximizing reward cost function.

Linking neurophysiological processes of action monitoring to post-response speed-accuracy adjustments in a neuro-cognitive diffusion model

The cognitive system needs to continuously monitor actions and initiate adaptive measures aimed at increasing task performance and avoiding future errors. To investigate the link between the contributing cognitive processes, we introduce the neuro-cognitive diffusion model, a statistical approach that allows a combination of computational modelling of behavioural and electrophysiological data on a single-trial level. This unique combination of methods allowed us to demonstrate across three experimental datasets that early response monitoring (error negativity; Ne/c) was related to more response caution and increased attention on task-relevant features on the subsequent trial, thereby preventing future errors, whereas later response monitoring (error positivity, Pe/c) maintained the ability of responding fast under speed pressure. Our results suggest that Pe/c-related processes might keep Ne/c-related processes in check regarding their impact on post-response adaptation to reconcile the conflicting criteria of fast and accurate responding.

Punishment and reward normalize error-related cognitive control in PTSD by modulating salience network activation and connectivity

Posttraumatic Stress Disorder (PTSD) symptomatology disrupts inhibitory control during sustained attention. However, PTSD-related inhibitory control deficits are partially ameliorated when punishments and rewards are administered based on task performance, which suggests motivational processes contribute to these deficits. Additionally, PTSD may also impair error-related cognitive control following inhibitory control failures as measured by post-error slowing (PES). However, it remains unclear if motivational processes also contribute to impaired error-related cognitive control in PTSD. Using an incentivized sustained attention paradigm in two independent samples of post-9/11 veterans, we characterized PTSD-related differences in PES during both non-motivated conditions (no task-based incentives) and motivated conditions (task-based rewards and punishments). In Study 1 (n = 139), PTSD symptom severity was modestly associated with smaller PES in the non-motivated condition, whereas no PTSD-related association was observed in the motivated condition. In Study 2 (n = 35), we replicated and extended these results by using fMRI to characterize modulation of the triple network system comprised of the Salience Network (SN), Frontoparietal Control Network (FPCN), and Default Mode Network (DMN). In the non-motivated condition, PTSD symptom severity was associated with non-specific SN and FPCN hyperactivation during both failed and successful inhibitory control. In the motivated condition, PTSD symptom severity was associated with greater focal activation of both the SN and Superior Parietal Lobule cluster (an FPCN node) during punished inhibitory control failures and weaker SN-FPCN connectivity during rewarded inhibitory control successes. Together, these results suggest that dysregulated motivational processes in PTSD may contribute to impaired error-related cognitive control.

Post-error Slowing During Instrumental Learning is Shaped by Working Memory-based Choice Strategies

Post-error slowing (PES) - a relative increase in response time for a decision on trialtgiven an error on trialt - 1 - is a well-known effect in studies of human decision-making. Post-error processing is reflected in neural signatures such as reduced activity in sensorimotor regions and increased activity in medial prefrontal cortex. PES is thought to reflect the deployment of executive resources to get task performance back on track. This provides a general account of PES that cuts across perceptual decision-making, memory, and learning tasks. With respect to PES and learning, things are complicated by the fact that learning often reflectsmultiple qualitatively different processes with distinct neural correlates. It is unclear if multiple processes shape PES during learning, or if PES reflects a policy for reacting to errors generated by one particular process (e.g., cortico-striatal reinforcement learning). Here we provide behavioral and computational evidence that PES is influenced by the operation of multiple distinct processes. Human subjects learned a simple visuomotor skill (arbitrary visuomotor association learning) under low load conditionsmore amenable to simple working memory-based strategies, and high load conditions that were putatively more reliant on trial-by-trial reinforcement learning. PES decreased withload, even when the progress of learning (i.e., reinforcement history) was accounted for. This result suggested that PES during learning is influenced by the recruitment of working memory. Indeed, observed PES effects were approximated by a computational model with parallel working memory and reinforcement learning systems that are differentially recruited according to cognitive load.

Adaptation following errors: Error awareness predicts future performance

The ability to detect an error in performance is critical to ongoing and future goal-directed behaviour. Diminished awareness of errors has been associated with a loss of insight and poor functional recovery in several clinical disorders (e.g., attention-deficit/hyperactivity disorder, addiction, schizophrenia). Despite the clear imperative to understand and remediate such deficits, error awareness and its instantiation in corrective behaviour remains to be fully elucidated. The present study investigated the relationship between error awareness and future performance in order to determine whether conscious recognition of errors facilitates adaptive behaviour. Fifty-one healthy participants completed a motor Go/No-Go error awareness task that afforded the opportunity to learn from errors. A mixed-effects model was specified wherein awareness of an error was used to predict inhibitory performance on the following No-Go trial. The model revealed a significant predictive effect of error awareness on future performance, such that aware errors were more frequently followed by correct inhibitory performance. Notably, improvement in performance accuracy was not due to a temporary increase in conservatism of responding, but appeared to be a context-specific adaptation. These results highlight the adaptive role of error awareness and the relationship between error awareness and learning from errors that has the potential to contribute to clinical symptomatology.

Post-error slowing: Large scale study in an online learning environment for practising mathematics and language

The ability to monitor and adjust our performance is crucial for adaptive behaviour, a key component of human cognitive control. One widely studied metric of this behaviour is post-error slowing (PES), the finding that humans tend to slow down their performance after making an error. This study is a first attempt at generalizing the effect of PES to an online adaptive learning environment where children practise mathematics and language skills. This population was of particular interest since the major development of error processing occurs during childhood. Eight million response patterns were collected from 150,000 users aged 5 to 13 years old for 6 months, across 23 different learning activities. PES could be observed in most learning activities and greater PES was associated with greater post-error accuracy. PES also varied as a function of several variables. At the task level, PES was greater when there was less time pressure, when errors were slower, and in learning activities focusing on mathematical rather than language skills. At the individual level, students who chose the most difficult level to practise and had higher skill ability also showed greater PES. Finally, non-linear developmental differences in error processing were found, where the PES magnitude increased from 6 to 9-years-old and decreased from 9 to 13. This study shows that PES underlies adaptive behaviour in an educational context for primary school students.

Understanding neural signals of post-decisional performance monitoring: An integrative review

Performance monitoring is a key cognitive function, allowing to detect mistakes and adapt future behavior. Post-decisional neural signals have been identified that are sensitive to decision accuracy, decision confidence and subsequent adaptation. Here, we review recent work that supports an understanding of late error/confidence signals in terms of the computational process of post-decisional evidence accumulation. We argue that the error positivity, a positive-going centro-parietal potential measured through scalp electrophysiology, reflects the post-decisional evidence accumulation process itself, which follows a boundary crossing event corresponding to initial decision commitment. This proposal provides a powerful explanation for both the morphological characteristics of the signal and its relation to various expressions of performance monitoring. Moreover, it suggests that the error positivity -a signal with thus far unique properties in cognitive neuroscience - can be leveraged to furnish key new insights into the inputs to, adaptation, and consequences of the post-decisional accumulation process.

Timing-dependent differential effects of unexpected events on error processing reveal the interactive dynamics of surprise and error processing

When unexpected events occur during goal-directed behavior, they automatically trigger an orienting-related cascade of psychological and neural processes through which they influence behavior and cognition. If the unexpected event was caused by an action error, additional error-specific, strategic-related processes have been proposed to follow the initial orienting period. Little is known about the neural interactions between action errors and unexpected perceptual events, two instantiations of unexpected events, in these two putative stages of post-error processing. Here, we aimed to address this by investigating the electrophysiological dynamics associated with action errors and unexpected perceptual events using scalp EEG with a focus on the frontal midline (FM) delta-to-theta oscillations (1-8 Hz) indicative of the performance-monitoring system. Specifically, we examined how the timing of unexpected sounds would influence behavior and neural oscillations after action errors, depending on the length of the intertrial interval (ITI). Our data showed that unexpected sounds aggravated post-error decreases in accuracy when they occurred (1) immediately after errors (i.e., post-error orienting period), regardless of ITI and (2) immediately after the post-error stimulus (i.e., post-error strategic period), at short ITIs. Meanwhile, action errors and unexpected sounds independently produced increased FM delta-to-theta power during the post-error orienting period, regardless of ITIs. However, when unexpected sounds occurred during the post-error strategic period, action errors produced lower FM delta-to-theta power than correct responses, at short ITIs. These differential effects of unexpected events on behavior and FM delta-to-theta dynamics support the notion of the two post-error periods during which different processes are implemented.