The aim of this study was to test the hypothesis that visual mismatch negativity (vMMN) is based on memory trace formation. Special care was taken to distinguish between memory mismatch and rareness effect. Subjects were seated in front of a monitor and asked to listen to a story. The standard sequence block consisted of nine consecutive 24-vane windmill patterns with an 800-ms inter-stimulus interval. The deviant sequence block consisted of one 24-vane pattern with eight six-vane patterns. Inter-train intervals (ITIs) varied among 1, 6, and 12 s in the changing ITI experiment, while the deviant stimulus occupied the eighth position in the sequence. In the changing order experiment, the position of the deviant stimulus was varied among the second, fourth and eighth position with a 12-s constant ITI. vMMN was ascertained from the difference in responses to standard and deviant stimulus. vMMN appeared in the occipital region 150-300 ms after stimulus onset. It was significantly modulated by the ITI, and more than four preceding stimuli were needed to reinstate the implicit memory trace. These results suggest that memory-based change detection underlies vMMN. Therefore, vMMN is useful to study visual sensory memory function.
All Science Journal Classification (ASJC) codes