Abstract
This experiment was designed to examine the possibility for the detection of deception by ERPs. Thirteen subjects were studied. Phonetic stimuli, the subject's own name (SN) and other typical Japanese names (TN) were used as S1, and photic stimuli were employed as S2. The subjects were required to prevent detection of relevant stimulus.
ERPs and CNVs data were analyzed by ANOVA. A P3b showed larger amplitude for the SN than for the TN, while an N400 showed larger amplitude for the TN than for the SN. When the SN was used as a relevant stimulus, a late CNV amplitude decreased.
The P3b for the SN and the N400 for the TN indicated that the subjects detected not only the target stimulus but also the TN. These results suggested that if the SN among the other names could be detected by using the indexes of the P3b and the N400, it might be possible to detect the physiological responses to the relevant stimulus which has strong relevance to the subject. In addition, the late CNV amplitude can be valid for an index of detection of deception.
