Objective: To assess patterns of novelty P3 elicited by the subject's own name (SON) in comatose patients and to compare SON novelty P3 prognostic value with that of mismatch negativity (MMN).
Methods: A passive oddball paradigm, previously validated in healthy subjects, including duration deviants and SON presented as a novel was applied in 50 severe comatose patients on average 20 days after coma onset. The outcome was assessed 3 months after coma onset.
Results: MMN to deviants was found in 14/50 patients and a central-parietal P3 to SON was found in 21/50 patients. In 12 patients, a parietal component (P3b) was also clearly present in the late part of P3. Four patients showed an MMN but no P3. Eleven patients had a novelty P3, with a late parietal component for 5 of them, but no MMN. The presence of a P3 was highly correlated with awakening. Compared to MMN, P3 showed as large a specificity for awakening (0.85). It showed a much higher sensitivity (0.71 versus 0.42). All but one patient with P3b woke up.
Conclusions: The use of novelty P3 elicited by SON increases the prognostic value of MMN alone and improves the assessment of comatose patients by demonstrating the activation of higher-level cognitive functions in some of them. It shows that unconsciously perceived stimuli are processed and activate brain areas similarly to consciously perceived stimuli.
Significance: SON as a novel in an MMN design can be used to increase the prognostic value of ERPs in comatose patients and to assess unconscious cognitive processes in uncommunicative patients.