Rapid adaptive adjustments of selective attention following errors revealed by the time course of steady-state visual evoked potentials

Abstract

Directing attention to task-relevant stimuli is crucial for successful task performance, but too much attentional selectivity implies that new and unexpected information in the environment remains undetected. A possible mechanism for optimizing this fundamental trade-off could be an error monitoring system that immediately triggers attentional adjustments following the detection of behavioral errors. However, the existence of rapid adaptive post-error adjustments has been controversially debated. While preconscious error processing reflected by an error-related negativity (Ne/ERN) in the event-related potential has been shown to occur within milliseconds after errors, more recent studies concluded that error detection even impairs attentional selectivity and that adaptive adjustments are implemented, if at all, only after errors are consciously detected. Here, we employ steady-state visual evoked potentials elicited by continuously presented stimuli to precisely track the emergence of error-induced attentional adjustments. Our results indicate that errors lead to an immediate reallocation of attention towards task-relevant stimuli, which occurs simultaneously with the Ne/ERN. Single-trial variation of this adjustment was correlated with the Ne/ERN amplitude and predicted adaptive behavioral adjustments on the post-error trial. This suggests that early error monitoring in the medial frontal cortex is directly involved in eliciting adaptive attentional adjustments.

Keywords: Cognitive control; Error monitoring; Event-related potentials; Selective attention; Steady-state visual evoked potentials.

Fig. 1

Experimental task and behavioral data. A: Exemplary stimulus display. In each run, participants viewed a continuous stream of two random-dot kinematograms (RDKs) consisting of randomly and incoherently moving red and blue dots. Embedded in each run were 20 trials in which both RDKs moved independently in one of the four cardinal directions for 500 ms. Participants had to indicate whether the relevant stimuli (indicated by the color of the fixation cross) moved horizontally or vertically by pressing a button. B: Congruent trials implied that both blue and red dots moved either horizontally or vertically. Incongruent trials implied that blue dots moved vertically and red dots moved horizontally, or vice versa. C: Mean response times of correct responses revealed a general slowing after error trials irrespective of whether the current trial was congruent or incongruent. D: Mean error rates showed a reduction of the congruency effect, and thus a focusing of attention to relevant stimuli, following error trials. Error bars represent within-subject standard errors of the mean. ms = milliseconds.

Fig. 2

Amplitudes of steady-state visual evoked potentials (SSVEPs). A: Spatial distributions of SSVEP amplitudes at the two stimulated frequencies revealed the typical occipital distribution of the SSVEP signal. B: SSVEP amplitudes were enhanced at 10 Hz when the relevant stimulus was red but at 15 Hz when the relevant stimulus was blue.

Fig. 3

Time course of SSVEP amplitudes for relevant stimuli, irrelevant stimuli, and attentional selectivity. A: SSVEP amplitudes for relevant stimuli diverged significantly between errors and correct trials at 20 ms before the response, indicating a fast reallocation of attention to relevant stimuli on errors. B: SSVEP amplitudes for irrelevant stimuli were larger for errors than for correct trials across the whole epoch, although this difference reached significance only between −500 ms and −250 ms and later than 60 ms, presumably reflecting that errors predominantly occur when attention to irrelevant stimuli is increased. C: The net effect of these two patterns is represented by the difference between SSVEP amplitudes for relevant and irrelevant stimuli. The resulting measure of attentional selectivity is reduced before −340 ms and increased after 170 ms on errors relative to correct trials. Shaded areas reflect within-participant 95%-confidence intervals at the respective time point. Horizontal bars indicate significant differences between correct trials and errors in the pre-response phase (grey) and the post-response phase (black), as revealed by a cluster-based permutation test. ms = milliseconds.

Fig. 4

Error-related brain activity, post-error performance and their relationship with attentional adjustments in SSVEP amplitudes. A: Event-related potentials for correct and error trials revealed a larger Ne/ERN (0–30 ms, cluster around FCz) on high-adjustment trials than on low-adjustment trials. Error trials were categorized as high/low-adjustment based on a median split of SSVEP amplitudes for relevant stimuli between 100 and 200 ms (i.e., later than the time range of the Ne/ERN). Topography reflects the spatial distribution of the difference between errors and correct trials in the time range of the Ne/ERN. B: No comparable effect was obtained for the Pe (250–350 ms, cluster around Pz). Topography reflects the spatial distribution of the difference between errors and correct trials in the time range of the Pe. C: The Ne/ERN difference between high-adjustment and low-adjustment trials was recalculated for median splits on SSVEP amplitudes of relevant and irrelevant stimuli in consecutive time windows, thus creating a measure of the relationship between Ne/ERN and SSVEP signals over time. Black bars indicate significant time points in a cluster-based permutation test. The Ne/ERN was significantly related to SSVEP amplitudes of relevant stimuli starting at 110 ms, but unrelated to SSVEP amplitudes of irrelevant stimuli. Shaded areas reflect 95%-confidence intervals at the respective time point. D: To reveal how adjustments of SSVEP amplitudes for relevant stimuli affected performance on post-error trials, response times and error rates on post-error trials were calculated as a function of adjustment on the previous error trial (as defined above). Whereas no significant effect was obtained for response times, the congruency effect in the error rates was strongly reduced following high-adjustment error trials than following low-adjustment error trials. Error bars represent within-subject standard errors of the mean. ms = milliseconds.

Fig. 5

Summary of the sequence of events around errors. Each arrow represents a relationship revealed by our analyses. Impaired selectivity during stimulus processing in SSVEPs points to the origin of errors (red box). Errors are preconsciously detected as indicated by the Ne/ERN (left green box) which immediately leads to adaptive adjustments of attention (blue box) before conscious error processing reflected by the Pe takes place (right green box). Attentional adjustments finally lead to behavioral adjustments measured on the subsequent trial.