It is no longer possible to use an elliptical construction, because one subject (the patient) is singular and the other (chemotherapy sessions) is plural. Therefore, it is necessary to include a second verb that corresponds to the second subject: for the type of offense in which the verb occurred flawlessly (failure of omission) and where the verb occurred with s (control error), we found no interaction between the form of the verb and the grammar in any time window. In other words, the participants also seemed sensitive to errors of omission and commission. To our knowledge, this is the first ERP study that directly compares errors and commission errors under the S-V agreement (previous studies have either looked at one type of error or collapsed in both types of errors). Here too, it is interesting to note that our participants were all adults who listened in their first language in a pristine auditory environment. It remains to be seen whether omission and commission errors are also important for other groups in the population, such as children or the latter, or whether adults in L1 have different sensitivities when they have hearing loss or if the acoustic environment is more difficult. Visual study of waveforms shows that non-grammatical verbs versus grammatical verbs produced a negative bilateral waveform on the front central electrodes, followed by a positive waveform on the central-posterior electrode. Statistical analyses carried out with cluster-based permutation tests, that the contrasts observed for grammatical and non-grammatical verbs revealed a significant negative group (p-0.036) between 130 and 210 ms in the forward central electrodes and a significant positive group (p < 0.0001) between 350 and 590 ms with a centcentcentcent-rear distribution. The authors explain that the research was conducted in the absence of commercial or financial relationships that could be interpreted as a potential conflict of interest. All patients were satisfied (patients are an accounting noun, so the verb is plural) Sentences were grouped into media and final lists where each list had two blocks of 10 minutes.