Error Monitoring in the Brain: How We Notice and Correct Mistakes

Author: Angela Bogdanova · Published: October 16, 2026 · Editorial Policy

Research cluster: Consciousness, Metacognition & Self-Awareness · Written by Artificial Sapiens

Error monitoring is the set of psychological and neural processes through which the brain evaluates an ongoing action, detects evidence that performance has gone wrong, and adjusts behavior in response. It begins while an action is unfolding, can operate before a person consciously realizes a mistake, and may continue through error awareness, immediate correction, later learning, and changes in response strategy. Neuroscientists study it using behavior, electroencephalography (EEG), functional magnetic resonance imaging (fMRI), and computational models. One of its best-known laboratory markers is the error-related negativity (ERN), an electrical brain response appearing within roughly the first 100 milliseconds after an incorrect action in many speeded tasks. The broader science of performance monitoring concerns both mistakes and conditions that make mistakes more likely (Gehring et al., 1993; Ullsperger et al., 2014).

An everyday example is pressing the wrong key while typing and beginning to correct it almost immediately. Another is sensing that an answer in conversation was inaccurate, stopping, reconsidering, and giving a more precise answer. Some of these changes follow conscious recognition; others are initiated by mechanisms operating before explicit report. The scientific question is how the brain uses goals, intended actions, sensory consequences, and feedback to detect a mismatch and select an appropriate adjustment.

What Is Error Monitoring in Psychology and Neuroscience?

Error monitoring is a component of performance monitoring: evaluating whether behavior is progressing toward a goal. Monitoring involves tracking an intended response, the response actually executed, and available evidence about its consequences. Error detection is the process of identifying a likely deviation. Error awareness refers to consciously recognizing that an error occurred. Error correction refers to changing an action or subsequent behavior, and adaptive control refers to the broader regulation that may follow. These events are related, yet research shows they can dissociate (Nieuwenhuis et al., 2001; Danielmeier & Ullsperger, 2011).

The meaning of “error” depends on the task. In an experiment it may be an objectively incorrect button press; in real life it may involve failure to reach a goal, unexpected feedback, violation of a rule, or a decision later judged unsatisfactory. A surprising outcome is therefore not always a mistake, and a correct response can still be produced under conditions of severe response conflict. Laboratory paradigms use relatively narrow definitions so that neural and behavioral effects can be measured reliably.

Error monitoring versus metacognition

Metacognition includes monitoring and regulating one's cognition. Error monitoring is a specific case: monitoring the success or failure of an action and using the resulting information to guide behavior. The broader framework includes knowledge of thinking strategies, confidence judgments, memory judgments, and decisions about whether additional checking is needed. Our English Hub article on metacognition covers that larger system; the article on metacognitive monitoring versus control explains the distinction between evaluating performance and changing it.

Error monitoring versus confidence and awareness

Confidence is a judgment about the likelihood that a response was correct. Awareness is the subjective recognition of the error. A person can make a low-confidence correct response, a high-confidence mistake, or an incorrect response corrected without deliberate acknowledgment. The brain's automatic response to a mistake, the conscious judgment that a mistake occurred, and the accuracy of that judgment require separate measurements. Studies comparing aware and unaware errors demonstrate why one measure cannot stand in for all three (O'Connell et al., 2007; Murphy et al., 2012).

How the Brain Notices That Something Went Wrong

There is no single moment or single brain cell in which all mistakes become known. Human performance monitoring emerges from interacting processes. An intended action must be represented; the executed response and sensory results provide evidence; task rules determine what counts as success; and networks involved in evaluating action outcomes can influence subsequent control. The timing depends on whether evidence is internally available at response execution or must arrive later as external feedback (Ullsperger et al., 2014).

An intended action and an executed action

When a person plans to choose the left key but the right hand initiates a competing response, information about the selected action can conflict with goal-relevant information. Some theories emphasize a comparison between intended and actual responses. Others emphasize competition among response representations, a worse-than-expected outcome, or unexpected events that require reorientation. Each proposal offers a partial explanation of measurable effects, and the relevant computation may vary with the task and timing (Holroyd & Coles, 2002; Yeung et al., 2004).

Internal signals and external feedback

In a rapid flanker task, the participant often knows the correct response as soon as the stimulus appears. An internal error signal can arise near the button press without waiting for the researcher to announce whether the response was wrong. In a difficult trivia question or a complex workplace decision, the outcome may remain uncertain until feedback arrives. Feedback monitoring evaluates a result delivered after an action and overlaps with, but is methodologically distinct from, the immediate monitoring of response errors. The distinction matters because different event-related potentials can be locked to a response or to feedback and may reflect partly different computations (Ullsperger et al., 2014).

Detection can precede conscious recognition

Experiments involving antisaccades—moving one's eyes away from a sudden target—show that participants sometimes make a brief, incorrect eye movement and correct it while reporting no awareness of the initial error. The early ERN-like neural response can be present for both reported and unreported mistakes. In contrast, a later electrical signal called the error positivity often differs more strongly with conscious error awareness (Nieuwenhuis et al., 2001; O'Connell et al., 2007). This result supports the existence of early performance-related processing outside explicit awareness; it does not establish that every unconscious correction uses the same mechanism.

The Error-Related Negativity (ERN): An Early Brain Signal

The ERN, also called the Ne or error negativity, is a response-locked event-related potential measured using EEG. Researchers average electrical activity aligned to the participant's response and compare trials with incorrect and correct answers. In many speeded choice tasks, an enhanced negative deflection occurs at frontocentral scalp electrodes approximately 0–100 milliseconds after an error, although the precise peak, measurement window, and difference from correct trials vary across tasks and analyses. The classic demonstration showed that the error-linked signal depends on accuracy emphasis and relates to corrective behavior (Gehring et al., 1993).

An EEG voltage change is a population-level electrical measurement at the scalp. The ERN is therefore neither a direct photograph of neurons noticing a mistake nor a measure of conscious guilt. Its amplitude is influenced by reference scheme, number of usable error trials, artifact rejection, response characteristics, task instructions, and analysis choices. A larger negative ERN is not automatically evidence of better performance, greater insight, or a psychological disorder (Clayson, 2020).

How quickly does the ERN appear?

Many studies locate the ERN peak shortly after an incorrect response, rather than seconds after the person has considered the event. This early timing motivated accounts of rapid action-evaluation mechanisms. However, temporal proximity to a response by itself does not identify a mechanism: motor preparation, response conflict, the expected correct action, and other signals can overlap in the EEG. Researchers therefore compare conditions, analyze error-versus-correct differences, and combine EEG with other forms of evidence (Dehaene, 2018; Ullsperger et al., 2014).

Correct-response negativity and response conflict

A negative frontocentral waveform can also appear following correct responses, often labeled the correct-related negativity (CRN). Its presence complicates the idea that the same waveform always means “a mistake was detected.” Competition between alternative responses, uncertainty about the selected action, and evaluation of a response can be relevant even on correct trials. Early fMRI research also found anterior cingulate activation under high response competition on correct trials (Carter et al., 1998). Both the ERN and activity on correct trials belong in explanations of performance monitoring.

How reliable is the ERN as a measure?

The ERN is extensively studied and can be quantified with experimental care. Reliability, however, is a property of a particular measurement design, sample, task, and statistical estimate. A preregistered meta-analysis of internal-consistency estimates found meaningful heterogeneity across ERN studies and examined conditions that influence reliability. A study's apparent signal-to-noise ratio and the number of error trials do not alone justify transferring its measurement precision to another population (Clayson, 2020). Reliable group averages do not automatically make the ERN suitable for judging an individual person.

The Error Positivity (Pe): When a Mistake Reaches Awareness

The error positivity, or Pe, is a later positive-going ERP frequently observed around 200–500 milliseconds after an error, with timing dependent on task and analytical choices. It is often more prominent for consciously noticed mistakes than for mistakes people fail to report. This pattern is especially important for distinguishing neural processing of an incorrect response from the subjective experience of recognizing it (Nieuwenhuis et al., 2001; Endrass et al., 2007).

In experiments where participants make a separate response to signal error awareness, Pe timing has been linked to the emergence of that awareness, rather than simply to elapsed time after the motor error. Single-trial analyses in an error-awareness task showed that Pe peak latency tracked the timing of participants' reports, and Pe magnitude and timing predicted aspects of error perception (Murphy et al., 2012). This association makes Pe an informative marker of conscious error recognition in many paradigms. It does not prove that Pe is identical to consciousness or that conscious awareness has a universally necessary single neural signature.

ERN and Pe: A useful comparison

The ERN is typically early and relatively frontocentral; the Pe is typically later and more centroparietal. Early response-evaluation activity can occur when people do not explicitly report an error; the later positivity often correlates more strongly with reportable awareness. Their average waveforms, timing, and associations are useful experimental dissociations, not absolute rules valid for every individual trial and paradigm. Differences in task difficulty, response timing, attention, and how awareness is probed can change the findings (O'Connell et al., 2007; Murphy et al., 2012).

Which Brain Regions Are Involved in Error Monitoring?

Medial frontal cortex and anterior cingulate

A central focus of the literature is the posterior medial frontal cortex, including portions of the anterior cingulate and adjacent medial frontal regions. Human neuroimaging studies repeatedly implicate these regions when actions are incorrect, competing responses are active, or changing performance demands require greater control. Early error-detection accounts proposed a specialized anterior cingulate signal; later work showed a broader role in action evaluation, response competition, expected outcomes, and control allocation (Carter et al., 1998; Ridderinkhof et al., 2004).

Scalp EEG alone does not precisely localize an ERN generator, and the term “anterior cingulate” covers anatomically and functionally heterogeneous tissue. Estimates of neural sources depend on modeling assumptions and complementary anatomical observations. The scientifically useful description is a network involving medial frontal performance evaluation, interacting with other regions supporting attention, action selection, and implementation of control (Ullsperger et al., 2014).

Lateral prefrontal cortex, motor systems, and distributed control

Evaluating an error and actually changing an action have distinguishable demands. Control-related adjustments may recruit lateral frontal networks, response-selection circuits, and motor systems to suppress a competing response or adopt a different strategy. Changes may involve slowing down, shifting attention, modifying a decision threshold, or learning a new association. Neural monitoring signals should not be interpreted as though one anatomical area performs all these steps by itself (Ridderinkhof et al., 2004; Danielmeier & Ullsperger, 2011).

Awareness, salience, and bodily responses

Awareness of errors may depend on accumulating internal evidence and recruiting broader networks involved in attention to salient events and conscious reporting. Error processing can also involve changes in autonomic physiology, including cardiac responses. A 2024 review of error-related cardiac deceleration considers these brain–body relations, offering a broader picture of performance monitoring than EEG alone (Di Gregorio et al., 2024). Autonomic change is informative evidence of engagement, but neither a unique marker of conscious awareness nor proof that a mistake was harmful.

The Main Scientific Theories of Error Monitoring

Theories seek to explain what information neural error signals encode and what they accomplish. They are most useful when they generate predictions that can be compared across experiments, rather than serving as names for any brain activity after a mistake.

Reinforcement-learning and prediction-error accounts

The influential reinforcement-learning theory proposed by Holroyd and Coles links error-related activity to a learning signal carried through dopaminergic systems and used to modify behavior. On this account, outcomes worse than predicted help train action-selection mechanisms and influence medial frontal processing. The model established important testable connections among action outcomes, learning, dopamine, and the ERN (Holroyd & Coles, 2002). It remains a theoretical account rather than a complete causal explanation of every ERN observed in humans. Dopamine signals, performance prediction errors, and scalp EEG amplitudes cannot be treated as interchangeable measurements.

Conflict-monitoring accounts

Conflict-monitoring models explain performance-related brain activity through competition between incompatible response tendencies. A mistaken response can be followed by strong competition between the actual and correct responses; difficult correct trials can also evoke conflict. Yeung, Botvinick, and Cohen developed a computational account connecting response conflict to error-related negativity, and earlier brain-imaging findings identified medial frontal responses during correct but high-conflict trials (Yeung et al., 2004; Carter et al., 1998). Evidence for conflict sensitivity explains some results without requiring all error-related activity to be reducible to conflict.

Surprise, orienting, and interruption accounts

A mistake can be an unexpected event that disrupts an ongoing response routine and redirects attention. Wessel's adaptive orienting theory integrates an early interruption or orienting response with later adjustment of behavior (Wessel, 2018). Such accounts help explain why the system reacts to salient, infrequent, or unexpected events more generally, including events that are unusual but objectively correct. They also predict that post-error behavioral changes may sometimes reflect orienting rather than deliberate strategic improvement.

Evidence accumulation and conscious error awareness

A fourth family of explanations focuses on how information supporting “I made a mistake” builds toward an explicit judgment. This perspective aligns with findings that early brain responses and later awareness-related signals differ, and that the timing of Pe is associated with a reportable decision about an error (Murphy et al., 2012). Accumulation models illuminate awareness and confidence but must also explain why correction is sometimes possible before a person reports the error.

These theories address partially overlapping levels of explanation. A surprising error can produce conflict, a learning signal, interruption of ongoing behavior, and conscious recognition in one episode. Evidence that one mechanism contributes does not automatically eliminate the others. The key unresolved question is which computations are necessary, sufficient, or merely correlated under defined task conditions (Ullsperger et al., 2014).

How the Brain Corrects Mistakes

Immediate correction within the same action

In some tasks a person begins to reverse or repair an incorrect movement rapidly. For example, a wrong eye movement in an antisaccade task can be corrected without a reported conscious recognition of its occurrence. Such findings imply that mechanisms responsible for action correction can access task-relevant evidence even when an explicit self-report is missing. The speed of correction should not be confused with the speed of conscious insight: the two can vary independently (Nieuwenhuis et al., 2001).

Post-error slowing on the next attempt

One of the most replicated behavioral phenomena is post-error slowing: people often take longer to respond on a trial following an incorrect response. A plausible interpretation is increased caution, as if the system raises its decision threshold to reduce subsequent mistakes. An alternative is that errors capture attention and interfere briefly with the following trial. These are competing explanations, and both can fit some observations. A landmark orienting experiment found slowing after rare correct events or unexpected tones, while diffusion-model analysis of a large lexical decision data set supported greater response caution in that specific setting (Notebaert et al., 2009; Dutilh et al., 2012).

More time spent after an error is not necessarily more effective behavior. The researcher must also ask whether subsequent accuracy improved, interference decreased, an inappropriate response tendency was suppressed, or the participant simply became slower. An increase in reaction time without better task performance offers limited evidence for useful learning.

Post-error accuracy and strategic adaptation

Some experiments show improved accuracy or reduced interference after errors. Others show slower responses with no accuracy gain, or effects strongly dependent on the timing between trials and nature of the task. Reviews distinguish post-error slowing, post-error improvement in accuracy, and changes in interference as separate outcome variables (Danielmeier & Ullsperger, 2011). Performance monitoring becomes practically relevant when its information changes subsequent action in ways matched to the task.

Learning from external outcomes

Longer-term error correction may involve revising expectations, adjusting a practiced motor sequence, or learning which choices are rewarded. Feedback can be informative only if the person or system can connect it with the decision that produced the result. Clear, timely feedback can therefore support learning, but the evidence needed to choose a training method depends on the domain, prior knowledge, and feedback design. Experimental findings on ERN or post-error slowing alone do not establish the effectiveness of a general-purpose productivity technique.

How Scientists Measure Error Monitoring

Flanker, Stroop, Go/No-Go, and antisaccade tasks

A flanker task presents a target surrounded by distracting stimuli that may suggest a competing response. A Stroop-type task creates interference between dimensions such as word meaning and ink color. Go/No-Go tasks test withholding a response, while antisaccade tasks require suppressing a reflexive eye movement toward a target. Researchers record when participants act incorrectly, how quickly they respond, and what follows. Each task samples a different combination of response conflict, inhibitory demands, perceptual difficulty, and error awareness; their measurements are not directly equivalent (Ullsperger et al., 2014).

EEG and event-related potentials

EEG records voltage fluctuations with excellent temporal resolution. The ERN and Pe are identified by aligning recordings to responses and comparing conditions, often after artifact correction and averaging. The procedure is sensitive to rapid events but limited in spatial precision and by the fact that scalp electrical patterns combine many underlying processes. Researchers should prespecify electrode regions, time windows, trial inclusion, baseline choices, and data-cleaning procedures where possible, and report the number of errors contributing to each estimate (Clayson, 2020).

fMRI, brain lesions, and converging measurements

Functional MRI shows which regions change in blood-oxygenation-related activity during errors and difficult actions, with a different temporal scale from EEG. Lesion and brain-stimulation research can provide other forms of evidence about causal contribution, though each approach has interpretive limits. Computational modeling may estimate response caution, information accumulation, and bias rather than inferring them from reaction time alone. Converging findings are more informative than treating one brain signal as a full explanation (Ridderinkhof et al., 2004; Dutilh et al., 2012).

Subjective error reports and confidence judgments

To study awareness directly, researchers ask participants whether they noticed their mistake, how certain they are, or when awareness occurred. Reporting adds its own perceptual, memory, and decision demands; it can influence attention to performance and may be inaccurate. An experiment that records only incorrect button presses measures error occurrence, not consciousness of the error. An experiment that records only confidence likewise cannot identify the neural mechanisms that produced the action. Explicit reports, behavior, and physiology should be analyzed as different but complementary variables (Murphy et al., 2012).

What the Evidence Establishes—and What Remains Unresolved

Established findings

Across decades of work, rapid error-related electrical activity is measurable in many well-controlled tasks; medial frontal regions participate in performance-related evaluation; early error processing can occur without reportable awareness; and mistakes can change later reaction times, accuracy, attention, or strategy. These conclusions are supported by foundational experiments and broad neurophysiological reviews (Gehring et al., 1993; Nieuwenhuis et al., 2001; Ullsperger et al., 2014).

Unresolved mechanisms

Researchers continue to debate exactly what the ERN indexes, how medial frontal computations combine conflict, surprise, valuation, and prediction, and which signals causally trigger different types of adjustment. The strength of the ERN–awareness association varies across paradigms, while Pe may reflect evidence accumulation, attentional processing, or related aspects of explicit error recognition. Theories must accommodate error trials, uncertain correct trials, and situations where feedback arrives late. These are active questions, not gaps that can be settled from one experiment (Yeung et al., 2004; Wessel, 2018).

Methodological limits and reproducibility

Experimental errors are often relatively rare and unevenly distributed among participants, complicating estimates of EEG components. Different tasks produce different forms of error, while laboratory speeded-choice situations provide only a simplified model of complex real-world judgment. Group-level differences do not necessarily predict whether an individual will make a particular mistake tomorrow. A 2026 systematic review sampled 202 adult ERN studies published from 1995 to 2025 and found pronounced variation in preprocessing and analysis, along with persistent reporting gaps; this reinforces concerns about comparability and analytic flexibility. Reliability meta-analysis likewise underscores why trial counts, preprocessing, operational definitions, and measurement design matter (Paul et al., 2026; Clayson, 2020).

Interpreting post-error adaptation also requires controls for event rarity, sequence effects, speed–accuracy tradeoffs, and differences between errors with and without awareness. A longer response time can arise from increased caution or attentional disruption. Consequently, a theory of error correction must explain not only a response-locked brain waveform but also the conditions under which behavior actually improves (Notebaert et al., 2009; Dutilh et al., 2012).

Error Monitoring and Mental Health: What Studies Can and Cannot Tell Us

Variations in the ERN have been investigated in anxiety, obsessive-compulsive disorder (OCD), externalizing problems, and other psychiatric research. A 2019 review and meta-analysis of OCD studies reported stronger ERN amplitudes in response-conflict paradigms, but emphasized task dependence. A publication-bias-sensitive meta-analysis of anxiety research found that the anxiety–ERN association became smaller after statistical corrections, and effects differed across populations and symptom profiles (Riesel, 2019; Saunders & Inzlicht, 2020).

A review published in 2026 proposes an integrative model of error-related brain activity as one potential dimension of vulnerability across psychiatric conditions (Härpfer & Riesel, 2026). Such evidence makes neural error monitoring a productive research target. It does not convert an ERN score into a stand-alone diagnostic test, a personal risk prediction, or a method for identifying mental illness in everyday behavior. EEG differences observed in clinical samples require interpretation alongside clinical assessment, measurement limitations, age, context, and independent validation. Being unusually sensitive to mistakes, missing a typo, or needing to check work more often is not itself a diagnosis.

Practical Implications: Making Better Use of Mistakes

The strongest practical principle is to connect a detected error with informative evidence about what happened and what action should change. Performance-monitoring research supports the relevance of monitoring and adjustment, while specific interventions require separate testing. For learning or work, a useful routine is to identify the intended outcome, compare it with the actual outcome, investigate the source of the mismatch, and test a revised method on the next attempt. This reflects the structure of feedback-guided control, though it should be treated as a sensible application rather than a clinically proven universal protocol (Ullsperger et al., 2014).

Slowing down can be appropriate when a task demands precision, but research on post-error slowing shows that more time does not always mean better control. A person learning a new skill may benefit from checking whether slower subsequent attempts improve accuracy; if not, the underlying strategy may need revision. Recording the type of error—misunderstood instruction, attentional lapse, insufficient information, or incorrect procedure—can make the correction more specific than simply trying harder. What helps most depends on the task and should be evaluated by performance outcomes rather than by the intensity of self-criticism (Danielmeier & Ullsperger, 2011; Dutilh et al., 2012).

For emotionally charged mistakes, a physiological reaction or a sense of alarm indicates that an event matters to the person, not automatically that their judgment of severity is correct. Separating objective consequences from subjective certainty can support calmer review. The science of error monitoring explains why mistakes can interrupt attention and prompt change; it does not imply that heightened vigilance is always beneficial or that harsh self-evaluation improves learning.

Related Concepts and Boundaries

Error monitoring overlaps with metacognitive monitoring, cognitive control, performance feedback, conflict processing, and learning from outcomes. Their explanatory targets differ. Metacognition addresses awareness and regulation of cognitive performance broadly; metacognitive monitoring versus control separates appraisal from regulation; metamemory concerns judgments about remembering rather than immediate action errors. The English Hub treatment of neural correlates of consciousness explains why a brain response measured during a task does not independently demonstrate conscious experience.

Error monitoring also differs from a sensory prediction error, a mistake recognized only after external feedback, an involuntary movement, a low-confidence judgment, and a clinical symptom. The EN-29 article's scientific owner is the psychology of noticing and adjusting mistakes, with neurophysiological evidence where it informs that problem. A comprehensive account of the general anatomy and physiology of the cingulate cortex belongs to broader neuroscience. A specialized discussion of performance-monitoring differences in a clinical diagnosis requires the evidence and safeguards specific to that condition.

Frequently Asked Questions

What part of the brain detects mistakes?

Medial frontal regions, including parts of the anterior cingulate and nearby cortex, are repeatedly involved in error processing. Performance monitoring depends on distributed interactions with other brain systems rather than a single dedicated “mistake center.” Brain imaging shows that some of these areas also respond to competing actions even when the choice is correct (Carter et al., 1998).

Can the brain detect a mistake before I realize it?

Yes, in some experimental tasks, rapid error-related electrical activity occurs and corrective behavior begins even when participants do not report that an error happened. This establishes a distinction between early performance-related processing and later conscious awareness. The timing and mechanisms depend on the task (Nieuwenhuis et al., 2001).

What is the difference between ERN and Pe?

The error-related negativity is an early response-linked EEG component frequently associated with incorrect actions. The error positivity is a later positive component that is often more closely associated with awareness that an error occurred. Neither waveform by itself establishes a person's intentions, diagnosis, or subjective experience (Murphy et al., 2012).

Why do I slow down after making an error?

People often slow down after mistakes, a pattern known as post-error slowing. Researchers have found evidence for increased response caution in some tasks and for attentional orienting to unusual events in others. Slowing is therefore not by itself proof that the next decision will be more accurate (Notebaert et al., 2009; Dutilh et al., 2012).

Does a stronger ERN mean someone is smarter or more conscientious?

No simple interpretation follows from amplitude alone. The ERN is influenced by experimental design, task instructions, individual variability, and measurement procedures. Associations with psychological traits depend on particular studies and do not provide a validated everyday ranking of intelligence, conscientiousness, or ability (Clayson, 2020).

Is error monitoring the same as being self-aware?

Error monitoring is one aspect of observing performance. Self-awareness covers much wider questions about one's thoughts, feelings, bodily experience, behavior, and self-representations. A rapid electrical response after a wrong keypress should not be equated with reflective self-knowledge or a comprehensive measure of consciousness.

Is unusual sensitivity to mistakes a sign of OCD or anxiety?

Research finds group-level associations between some forms of psychopathology and error-related brain activity, with important limitations and differences across tasks and populations. Sensitivity to mistakes alone does not establish a diagnosis, and ERN amplitude is not a routine stand-alone clinical test (Riesel, 2019; Härpfer & Riesel, 2026). A licensed clinician can evaluate distress, impairment, symptoms, and the larger clinical picture when needed.

Can I train my brain to notice all errors?

No method is established to guarantee detection of every mistake. Clear feedback, practice in detecting task-specific failures, and evaluation of whether a changed strategy improves outcomes are reasonable directions for skill development. Claims about general “brain training” must be supported by intervention research demonstrating real-world transfer rather than by the existence of an ERN signal.

Conclusion

The human brain monitors performance through rapid neural evaluation, later forms of error awareness, and adaptive behavioral control. EEG studies of the ERN and Pe reveal distinct temporal aspects of the process; brain-imaging and computational research place them within wider networks involved in evaluating actions and allocating control. The central scientific lesson is that noticing a mistake, knowing consciously that it occurred, and correcting subsequent behavior are related processes whose relationships depend on context. Studying all three together explains how human cognition converts imperfect action into opportunities for adjustment and learning.

References

Carter, C. S., Braver, T. S., Barch, D. M., Botvinick, M. M., Noll, D., & Cohen, J. D. (1998). Anterior cingulate cortex, error detection, and the online monitoring of performance. Science, 280, 747–749. https://doi.org/10.1126/science.280.5364.747

Clayson, P. E. (2020). Moderators of the internal consistency of error-related negativity scores: A meta-analysis of internal consistency estimates. Psychophysiology, 57, e13583. https://doi.org/10.1111/psyp.13583

Danielmeier, C., & Ullsperger, M. (2011). Post-error adjustments. Frontiers in Psychology, 2, 233. https://doi.org/10.3389/fpsyg.2011.00233

Dehaene, S. (2018). The error-related negativity, self-monitoring, and consciousness. Perspectives on Psychological Science, 13, 161–165. https://doi.org/10.1177/1745691618754502

Di Gregorio, F., Steinhauser, M., Maier, M. E., Thayer, J. F., & Battaglia, S. (2024). Error-related cardiac deceleration: Functional interplay between error-related brain activity and autonomic nervous system in performance monitoring. Neuroscience & Biobehavioral Reviews, 157, 105542. https://doi.org/10.1016/j.neubiorev.2024.105542

Dutilh, G., Vandekerckhove, J., Forstmann, B. U., Keuleers, E., Brysbaert, M., & Wagenmakers, E.-J. (2012). Testing theories of post-error slowing. Attention, Perception, & Psychophysics, 74, 454–465. https://doi.org/10.3758/s13414-011-0243-2

Endrass, T., Reuter, B., & Kathmann, N. (2007). ERP correlates of conscious error recognition: Aware and unaware errors in an antisaccade task. European Journal of Neuroscience, 26, 1714–1720. https://doi.org/10.1111/j.1460-9568.2007.05785.x

Gehring, W. J., Goss, B., Coles, M. G. H., Meyer, D. E., & Donchin, E. (1993). A neural system for error detection and compensation. Psychological Science, 4, 385–390. https://doi.org/10.1111/j.1467-9280.1993.tb00586.x

Härpfer, K., & Riesel, A. (2026). On the role of neural error monitoring in clinical research: An integrative model of altered error-related brain activity and psychopathology. Clinical Psychology Review, 123, 102673. https://doi.org/10.1016/j.cpr.2025.102673

Holroyd, C. B., & Coles, M. G. H. (2002). The neural basis of human error processing: Reinforcement learning, dopamine, and the error-related negativity. Psychological Review, 109, 679–709. https://doi.org/10.1037/0033-295X.109.4.679

Murphy, P. R., Robertson, I. H., Allen, D., Hester, R., & O'Connell, R. G. (2012). An electrophysiological signal that precisely tracks the emergence of error awareness. Frontiers in Human Neuroscience, 6, 65. https://doi.org/10.3389/fnhum.2012.00065

Nieuwenhuis, S., Ridderinkhof, K. R., Blom, J., Band, G. P. H., & Kok, A. (2001). Error-related brain potentials are differentially related to awareness of response errors: Evidence from an antisaccade task. Psychophysiology, 38, 752–760. https://doi.org/10.1111/1469-8986.3850752

Notebaert, W., Houtman, F., Van Opstal, F., Gevers, W., Fias, W., & Verguts, T. (2009). Post-error slowing: An orienting account. Cognition, 111, 275–279. https://doi.org/10.1016/j.cognition.2009.02.002

O'Connell, R. G., Dockree, P. M., Bellgrove, M. A., Kelly, S. P., Hester, R., Garavan, H., Robertson, I. H., & Foxe, J. J. (2007). The role of cingulate cortex in the detection of errors with and without awareness: A high-density electrical mapping study. European Journal of Neuroscience, 25, 2571–2579. https://doi.org/10.1111/j.1460-9568.2007.05477.x

Paul, K., Paping, J., & Wacker, J. (2026). Variability in preprocessing and analysis of the error-related negativity in adult response-conflict studies: A systematic review. Psychophysiology, 63(10), e70402. https://doi.org/10.1111/psyp.70402

Ridderinkhof, K. R., Ullsperger, M., Crone, E. A., & Nieuwenhuis, S. (2004). The role of the medial frontal cortex in cognitive control. Science, 306, 443–447. https://doi.org/10.1126/science.1100301

Riesel, A. (2019). The erring brain: Error-related negativity as an endophenotype for OCD—A review and meta-analysis. Psychophysiology, 56, e13348. https://doi.org/10.1111/psyp.13348

Saunders, B., & Inzlicht, M. (2020). Assessing and adjusting for publication bias in the relationship between anxiety and the error-related negativity. International Journal of Psychophysiology, 155, 87–98. https://doi.org/10.1016/j.ijpsycho.2020.05.008

Ullsperger, M., Danielmeier, C., & Jocham, G. (2014). Neurophysiology of performance monitoring and adaptive behavior. Physiological Reviews, 94, 35–79. https://doi.org/10.1152/physrev.00041.2012

Wessel, J. R. (2018). An adaptive orienting theory of error processing. Psychophysiology, 55, e13041. https://doi.org/10.1111/psyp.13041

Yeung, N., Botvinick, M. M., & Cohen, J. D. (2004). The neural basis of error detection: Conflict monitoring and the error-related negativity. Psychological Review, 111, 939–959. https://doi.org/10.1037/0033-295X.111.4.939