Reinforcement-related brain potentials from medial frontal cortex: origins and functional significance

Nieuwenhuis, Sander; Holroyd, Clay B.; Mol, Nisan; Coles, Michael

doi:10.1016/j.neubiorev.2004.05.003

Cited by 479 publications

(422 citation statements)

References 43 publications

Supporting

Mentioning

384

Contrasting

Unclassified

Order By: Relevance

“…Other evidence consistent with our results has been reported by recent studies of a reward-sensitive electrophysiological brain potential thought to be generated in the medial frontal cortex: the feedback negativity (reviewed in Nieuwenhuis et al, 2004). These studies have indicated that the evaluative system that produces the feedback negativity processes gains and losses in a contextdependent manner (Holroyd et al, 2004a).…”

Section: Discussionsupporting

confidence: 93%

Activity in human reward-sensitive brain areas is strongly context dependent

et al. 2005

Self Cite

View full text Add to dashboard Cite

Functional neuroimaging research in humans has identified a number of brain areas that are activated by the delivery of primary and secondary reinforcers. The present study investigated how activity in these reward-sensitive regions is modulated by the context in which rewards and punishments are experienced. Fourteen healthy volunteers were scanned during the performance of a simple monetary gambling task that involved a bwinQ condition (in which the possible outcomes were a large monetary gain, a small gain, or no gain of money) and a bloseQ condition (in which the possible outcomes were a large monetary loss, a small loss, or no loss of money). We observed reward-sensitive activity in a number of brain areas previously implicated in reward processing, including the striatum, prefrontal cortex, posterior cingulate, and inferior parietal lobule. Critically, activity in these reward-sensitive areas was highly sensitive to the range of possible outcomes from which an outcome was selected. In particular, these regions were activated to a comparable degree by the best outcomes in each condition-a large gain in the win condition and no loss of money in the lose condition-despite the large difference in the objective value of these outcomes. In addition, some rewardsensitive brain areas showed a binary instead of graded sensitivity to the magnitude of the outcomes from each distribution. These results provide important evidence regarding the way in which the brain scales the motivational value of events by the context in which these events occur. D

show abstract

Section: Discussionsupporting

confidence: 93%

Activity in human reward-sensitive brain areas is strongly context dependent

et al. 2005

Self Cite

View full text Add to dashboard Cite

show abstract

“…However, because our design only allowed us to compute this correlation across and not within subjects, further studies are needed to determine whether the FRN is also modulated in a similar way within subjects (across trials), and hence parametrically encodes the perceived severity of errors. Such a quantitative coding of the magnitude of an error would be consistent with a fine-tuned prediction error coding system (Fiorillo et al, 2003;Yasuda et al, 2004;Potts et al, 2010) and may depend on flexible reinforcement learning brain mechanisms Nieuwenhuis et al, 2004a).…”

Section: Using This New Task We Show That Commission and Accuracy Ermentioning

confidence: 69%

“…The reinforcement learning (RL) theory holds that the ERN and FRN indirectly reflect activity of a deep mesencephalic system projecting to the basal ganglia and then the ACC, where these ERP responses are eventually generated Nieuwenhuis et al, 2004a;Frank et al, 2005). According to this model, neurons in the basal ganglia constantly monitor ongoing behavior.…”

Section: Introductionmentioning

confidence: 99%

Delayed monitoring of accuracy errors compared to commission errors in ACC

et al. 2012

View full text Add to dashboard Cite

Error detection is essential for monitoring performance and preparing subsequent behavioral adjustments, and is associated with specific neural responses in the anterior cingulate cortex (ACC). To investigate whether different brain mechanisms subserve the processing of commission vs. accuracy errors, we recorded EEG in adult participants while they performed a novel speeded GO-NOGO aiming task ("the Shoot-NoShoot paradigm"). Our ERP results show that commission errors (responding during NOGO trials) elicited a classical error-related negativity (ERN) component, followed by an error-related positivity (Pe), as well as a negativity peaking before response onset (pre-ERN). By contrast, spatial accuracy errors elicited a feedback-related negativity (FRN), which correlated with the spatial discrepancy between response and target position across subjects. Fast hits also elicited a pre-ERN but no ERN, suggesting that this pre-response monitoring component might be related to the detection of error likelihood. Although source analysis revealed similar generators in ACC for these different error-related negativities, the respective timing differed, suggesting that commission errors are detected rapidly based on internal motor representations, whereas the detection of accuracy errors in ACC relies on the additional and swift processing of external visual information.

show abstract

“…The reinforcement learning model from Holroyd and Coles (2002), Holroyd et al (2005), Nieuwenhuis et al (2004Nieuwenhuis et al ( , 2002 argues that errors are detected primarily by the basal ganglia, which compares known stimulus-response relationships to stimuli perceived and responses made. Although the complex model is beyond description here, it does suggest that the ERN is the result of the basal ganglia enervating the mesencephalic dopamine system, which in turn disinhibits motor neurons in the ACC.…”

Section: Post-error Adaptationmentioning

confidence: 99%

Post-Error Behavior in Active Cocaine Users: Poor Awareness of Errors in the Presence of Intact Performance Adjustments

Hester

Simões‐Franklin

Garavan

2007

Neuropsychopharmacol

108

View full text Add to dashboard Cite

Active cocaine abusers have a diminished neural response to errors, particularly in the anterior cingulate cortex thought critical to error processing. The inability to detect, or adjust performance following errors has been linked to clinical symptoms including the loss of insight and perseverative behavior. We investigated the cognitive implications of this diminished error-related activity, using response inhibition tasks that required error awareness and performance adaptation. Twenty-one active cocaine users (six female subjects, mean age ¼ 40.3) and 22 non-drug using adults (six female subjects, mean ¼ 39.9) participated. The results indicated that cocaine users consistently demonstrated poorer inhibitory control, a deficit accompanied by reduced awareness of errors. Adaptation of post-error reaction times did not differ between groups, although a different measure of adaptive behavior: exerting inhibitory control on the trial immediately after failing to inhibit, was significantly poorer in the cocaine using sample. In summary, cocaine users demonstrated a diminished capacity for monitoring their behavior, but were able to perform post-error adjustment to processes not already suffering an underlying deficit. These difficulties are consistent with previous reports of cocaine-related hypoactivity in the neural system underlying cognitive control, and highlight the potential for cognitive dysfunction to manifest as behavioral deficits that likely contribute to the maintenance of drug dependence.

show abstract

Reinforcement-related brain potentials from medial frontal cortex: origins and functional significance

Cited by 479 publications

References 43 publications

Activity in human reward-sensitive brain areas is strongly context dependent

Activity in human reward-sensitive brain areas is strongly context dependent

Delayed monitoring of accuracy errors compared to commission errors in ACC

Post-Error Behavior in Active Cocaine Users: Poor Awareness of Errors in the Presence of Intact Performance Adjustments

Contact Info

Product

Resources

About