The Neural Basis of Error Detection: Conflict Monitoring and the Error-Related Negativity Nick Yeung Princeton University Matthew M. Botvinick University of Pennsylvania Jonathan D. Cohen Princeton University and University of Pittsburgh According to a recent theory, anterior cingulate cortex is sensitive to response conflict, the coactivation of mutually incompatible responses. The present research develops this theory to provide a new account of the error-related negativity (ERN), a scalp potential observed following errors. Connectionist simu- lations of response conflict in an attentional task demonstrated that the ERN—its timing and sensitivity to task parameters— can be explained in terms of the conflict theory. A new experiment confirmed predictions of this theory regarding the ERN and a second scalp potential, the N2, that is proposed to reflect conflict monitoring on correct response trials. Further analysis of the simulation data indicated that errors can be detected reliably on the basis of post-error conflict. It is concluded that the ERN can be explained in terms of response conflict and that monitoring for conflict may provide a simple mechanism for detecting errors. Errors are an important source of information in the regulation of cognitive processes. The mechanism by which people detect and correct their errors has been the object of study for many years, but research interest has increased in recent years following the dis- covery of neural correlates of performance monitoring. In partic- ular, studies of event-related brain potentials (ERPs) have revealed a neural response following errors that has been labeled the error- related negativity (ERN or Ne; Falkenstein, Hohnsbein, Hoorman, & Blanke, 1990, 1991; Gehring, Goss, Coles, Meyer, & Donchin, 1993). The most likely neural generator of the ERN is anterior cingulate cortex (ACC), an area that in recent years has been implicated in another function related to the evaluation of perfor- mance, monitoring for competition (or conflict) during information processing. The present research attempts to provide an integrative account of error- and conflict-related activity observed in anterior cingulate cortex. Specifically, we propose a new account of the ERN and error processing in terms of the conflict monitoring theory of anterior cingulate function. Background Behavioral Studies of Error Monitoring Participants in reaction time (RT) experiments are typically aware of their errors, reacting to them with visible or audible frustration. When asked, they are also able to signal their errors more systematically with an appropriate key-press (Rabbitt, 1966, 1967, 1968). Using this method, Rabbitt and colleagues have found that participants can detect most, though rarely all, of the errors they make in simple choice RT tasks (Rabbitt, 1968, 2002). However, these error-signaling responses can be quite slow and unreliable. In a study by Rabbitt (2002), for example, young adults detected 79% of their errors, taking an average of about 700 ms to do so, when they were given a second to respond before the next stimulus appeared. However, when the subsequent stimulus ap- peared 150 ms after an incorrect response, the same participants showed a limited ability to ignore this stimulus, as they were instructed to do, and their error-detection rate dropped to 56%. Whereas explicit error detection and signaling appear to be slow and effortful, error correction is fast and relatively automatic. Participants deal with errors more quickly and efficiently by pro- ducing a correcting response—that is, making the response they should have made—than by making a common detection response to all errors (Rabbitt, 1968, 1990, 2002). Indeed, errors are often immediately followed by a correcting response even when partic- ipants are instructed to suppress such responses (Maylor & Rab- Nick Yeung, Department of Psychology, Princeton University; Matthew M. Botvinick, Department of Psychiatry and Center for Cognitive Neuro- science, University of Pennsylvania; Jonathan D. Cohen, Department of Psychology, Princeton University, and Department of Psychiatry, Univer- sity of Pittsburgh. Preliminary versions of this research were presented in poster format at the conference on Executive Control, Errors, and the Brain (Jena, Ger- many, September 2000), at the 7th International Conference on Functional Mapping of the Human Brain (Brighton, England, June 2001), and at the 31st Annual Meeting of the Society of Neuroscience (San Diego, Califor- nia, November 2001). We thank Mike Coles, Clay Holroyd, John Kounios, Sander Nieuwenhuis, and Leigh Nystrom for comments on drafts of this article; Joe Bussiere, Jack Gelfand, Mike Scanlon, and Dan Zook for help in running the event-related brain potential experiment; and Richard Greenblatt for technical advice. Correspondence concerning this article should be addressed to Nick Yeung, who is now at the Department of Psychology, Carnegie Mellon University, Pittsburgh, PA 15213. E-mail: nyeung@cmu.edu Psychological Review Copyright 2004 by the American Psychological Association 2004, Vol. 111, No. 4, 931–959 0033-295X/04/$12.00 DOI: 10.1037/0033-295X.111.4.931 931