Fig. 2. Task behavioral metrics as a function of trial type, group, and drug conditions.

(A) Task difficulty was individually tailored to achieve consistent percentages of error trials across participants and sessions. Following both error and correct responses, informative feedback was presented on 73% (red/green) and noninformative feedback on 27% of trials (yellow) receiving a participant response. The no-response trials (white) reflected momentary lapses of attention and were not under the control of the task’s adaptive algorithm. (B) Average RT for errors was longer than that for correct trials [RESPONSE main effect (*P < 0.001)]. (C) Nicotine and varenicline decreased the percent of no-response trials among both smokers (PATCH * PILL: P = 0.02) and nonsmokers (PATCH * PILL: P = 0.049). *Significant post hoc pairwise comparisons at a Bonferroni-corrected threshold (n = 3 comparisons). Error bars, SEM.