Neural correlates of asymmetric switch costs during affective task switching.

Task switching is widely used to investigate cognitive control, typically revealing switch costs-greater performance costs in switch than in repetition trials. Previous research reported affective asymmetric switch costs, with higher costs when switching towards an affective task compared with a switch towards a neutral task. This asymmetry can be explained by the inhibition of the affective task, which enables the performance of the neutral task, leading to increased costs for a switch from the neutral toward the affective task. In Experiment 1, we examined whether the affective task content interferes with the preparation of the switching process, as reflected in neural correlates of proactive control. In Experiment 2, we investigated whether the maintenance of task sets is modulated by affective task content indexed by mixing costs (repetition - single-task performance) at the behavioural and the neural levels. Participants completed a cued task-switching during electroencephalographic recording, and judged either the gender (neutral-task) or emotional expression (affective-task) of faces. Neurally, within the cue-stimulus interval, the switch-related posterior positivity was reduced in the affective compared with the neutral task, coinciding with increased switch costs for the affective task. This suggests an impaired task-set reconfiguration for the affective compared with the neutral task. For Experiment 2, we investigated whether mixing costs and the mixing-related centroparietal positivity are modulated by affective task content. Neurally, within the cue-stimulus interval, the mixing-related centroparietal positivity remained unchanged by affective task content, mirrored behaviourally by symmetrical mixing costs. These findings indicate that affective task content selectively interferes with the preparation of the switching process but not with goal-setting.