The occurrence of such "partial errors" shows that incorrect response activations could be corrected online, before turning into overt errors.
In the present study, we showed that, unlike overt errors, such "partial errors" are poorly consciously detected by participants, who could report only one third of their partial errors.
Incorrect responses (i.e. overt errors), response failures (i.e. RT > 2000 ms) and premature responses (i.e. RT < 100 ms; regardless of accuracy) were excluded a priori from RT data analyses.
First, overt errors are known to elicit two brain responses: early fronto-medial brain activity, known as the "error negativity" (Ne: Falkenstein, Hohnsbein, Hoormann, & Blanke, 1991; also termed the "error-related negativity," or ERN: Gehring, Goss, Coles, Meyer, & Donchin, 1993), followed by a later positive deflection termed the "error positivity" (Pe: Falkenstein et al., 1991).
Such "partial errors" have been shown to induce a brain response initially reported for overt errors (Burle et al., 2008; Masaki & Segalowitz, 2004; Scheffers, Coles, Bernstein, Gehring, & Donchin, 1996; Vidal et al., 2000), called the "error negativity" (Falkenstein et al., 1991) or "error-related negativity" (Gehring et al., 1993).
On the other hand, it has been shown that cognitive adjustments do occur after partial errors, such as increased incorrect-response suppression (Burle et al., 2002) or post-partial-error slowing (although this is weaker than slowing after overt errors; Allain, Burle, Hasbroucq, & Vidal, 2009).
Concerning the latter comparison, as has already been reported (Allain, Carbonnell, Burle, Hasbroucq, & Vidal, 2004), the mean EMG burst has a smaller amplitude on overt errors than on pure-correct responses [ t(14) = 3.55, p <.002], whereas the initial slopes do not differ [ t(14) = 0.3, p =.77; see Fig. 2b, inset].
Indeed, Fig. 2 (panels c and d) shows the cumulative density functions (CDFs) of the EMG surface (panel c) for undetected and detected partial errors, along with the CDF for the overt-error surface, as an inset.
This rating procedure basically replaced the verification phase (i.e. overt registration of errors whenever the participant felt he or she had made a response error), as implemented in the main EEG experiment.
Besides, trials with tapping standard deviations greater than 1000 ms were treated as missing values as they potentially reflect overt omissions and errors due to e.g. key mal-functions.
