Neuropsychologia 47 (2009) 663–670 Contents lists available at ScienceDirect Neuropsychologia journal homepage: www.elsevier.com/locate/neuropsychologia Neural time course of conﬂict adaptation effects on the Stroop task Michael J. Larson a,b,∗ , David A.S. Kaufman b , William M. Perlstein b,c,d a Department of Psychology, Brigham Young University, Provo, UT, United States b Department of Clinical and Health Psychology, University of Florida, Gainesville, FL, United States c Department of Psychiatry, University of Florida, Gainesville, FL, United States d McKnight Brain Institute, University of Florida, Gainesville, FL, United States article info Article history: Received 3 June 2008 Received in revised form 24 October 2008 Accepted 17 November 2008 Available online 21 November 2008 Keywords: Event-related potential ERP Conﬂict Adaptation Sequential trial Stroop Cognitive control Executive function Anterior cingulate ACC Gratton effect abstract Cognitive control theory suggests conﬂict effects are reduced following high- relative to low-conﬂict trials. Such reactive adjustments in control, frequently termed “conﬂict adaptation effects,” indicate a dynamic interplay between regulative and evaluative components of cognitive control necessary for adaptable goal-directed behavior. The current study examined conﬂict adaptation effects while 36 neurologically- normal participants performed a single-trial color-naming Stroop task. Trials preceded by incongruent (high conﬂict) and congruent (low conﬂict) trials were compared for behavioral (response time [RT] and error rate) and electrophysiological (N450 and conﬂict SP components of the event-related poten- tial [ERP]) concomitants of cognitive control. A conﬂict adaptation effect was present for RTs that could not be accounted for by associative or negative priming. ERPs revealed a parietal conﬂict slow poten- tial (conﬂict SP) that differentiated incongruent from congruent trials and monotonically differentiated current trial congruency on the basis of previous-trial context (i.e., showed conﬂict adaptation); the fronto-medial N450 was sensitive to current trial congruency but not to previous-trial context. Direct comparison of normalized conﬂict SP and N450 amplitudes showed the conﬂict SP was sensitive to the effects of previous-trial context, while the N450 was so to a lesser extent and in a different pattern. Find- ings provide clariﬁcation on the neural time course of conﬂict adaptation and raise further questions regarding the relative roles of the parietal conﬂict SP and fronto-medial N450 in conﬂict detection and processing. © 2008 Elsevier Ltd. All rights reserved. Goal-directed behavior requires a ﬂexible and adaptive cogni- tive control system for recognizing appropriate or inappropriate task completion and dynamically adjusting performance when con- trol is misdirected or inadequate. The cognitive control mechanisms required to monitor for performance errors or response conﬂict (i.e., the simultaneous activation of two competing response options) and to signal for subsequent adjustments are critical to adap- tive behavior and efﬁcient task completion (see Botvinick, Carter, Braver, Barch, & Cohen, 2001). Performance adjustments include slowing of response time (RT) following an error to increase accu- racy (i.e., the Rabbitt effect; Rabbitt, 1966, 1968) and facilitation of RTs following high- relative to low-conﬂict trials (i.e., the Grat- ton or “conﬂict adaptation” effect; Gratton, Coles, & Donchin, 1992). Several authors have utilized the Stroop color-naming task to examine such performance adjustments. In the Stroop task (Stroop, 1935), conﬂict is greater for incongruent (e.g., the word BLUE writ- ∗ Corresponding author at: Department of Psychology, Brigham Young University, 244 TLRB, Provo, UT 84602, United States. Tel.: +1 801 422 6125. E-mail address: michael larson@byu.edu (M.J. Larson). ten in red) than congruent (RED written in red) color-naming trials due to the simultaneous activation of competing represen- tations. Behavioral adjustments in control following high conﬂict include faster RTs on incongruent trials preceded by incongruent trials (hereafter referred to as iI trials) than on incongruent trials preceded by congruent trials (cI), and slower RTs for congruent tri- als preceded by incongruent trials (iC) relative to congruent trials preceded by congruent trials (cC) (Kerns et al., 2004). The expla- nation offered for this pattern of performance adjustments is that high conﬂict detected on an incongruent trial leads to recruit- ment of greater cognitive resources than on congruent trials; the cognitive resources are then utilized on the subsequent trial to enhance performance (Botvinick et al., 2001; Egner, 2007; Gratton et al., 1992; Kerns, 2006; Kerns et al., 2004; Ullsperger, Bylsma, & Botvinick, 2005). In consequence, RTs on iI trials are faster than cI trials because the preceding incongruent trial results in increased signaling for cognitive control, while when the preceding trial is congruent fewer cognitive resources are allocated for use on the following trial. Response times for iC trials tend to be longer than those for cC trials due to switching between congruencies and because conﬂict-driven control reduces the facilitating effect of consecutive repetition of congruent trials (see Egner, 2007 for 0028-3932/$ – see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.neuropsychologia.2008.11.013