No abstract

No abstract

In a choice reaction time (RT) task, electromyographic (EMG) recordings allowed us to fractionate RT into two subcomponents, namely premotor time and motor time. This has been done for correct trials and errors. The analysis of the EMG burst and motor time (between EMG onset and overt response) showed that the EMG burst amplitude was reduced and the motor time was longer for errors than for correct responses. In the same way as posterror slowing on the RT was interpreted as revealing between-trials changes in executive control, the present data provide direct evidence for an on-line, within-trial, executive control.

No abstract

The event-related potential called "Error Negativity" (Ne, ERN), which appears when subjects commit errors in choice reaction time tasks, is a marker of response monitoring. By introducing a response probability bias, we show that the Ne is sensitive to response expectancy. We further show that the small negativity evoked by correct responses (Ne-like, CRN) is also sensitive to response expectancy: On unexpected responses, the former decreases while the latter increases to such an extent that the amplitudes of the two components are in the same range of magnitude. Although the sensitivity of the Ne to response expectancy is compatible with the current models accounting for the Ne, the common sensitivity of the Ne and the Ne-like supports the idea that they reflect functionally similar monitoring processes.