Backward-masking: the effect of the duration of the second stimulus on recognition of the first stimulus.

OBJECTIVE: We recorded event-related magnetic fields following a target stimulus followed by a masking stimulus to investigate the visual backward masking effect using a helmet-type magnetoencephalography system in humans. METHODS: In the target stimulus with masking stimulus conditions, duration of the target stimulus was constant at 16 ms, and duration of the masking stimulus was altered (16, 48 and 144 ms). The target stimulus was masked by the 144-ms masking stimulus, but not by the 16-ms masking stimulus, and was obscured by the 48-ms masking stimulus. For control conditions (Single-condition), event-related magnetic fields were recorded following the sole presentation of the masking stimulus for 32, 64 or 160 ms. RESULTS: One major response was obtained at 180 ms after the onset of the stimulation in each condition. The equivalent current dipole of one major response was estimated to lie in the occipital lobe, but there was a relatively large inter-individual difference. There was no significant difference in latency between the target stimulus with masking stimulus conditions and Single-conditions. In the target stimulus with masking stimulus conditions with the 48- and 144-ms masking stimulus, the root mean square value did not differ from that in the respective Single-condition, while the root mean square value for the target stimulus with masking stimulus conditions with the 16-ms masking stimulus was significantly smaller than that in the Single-condition with the 32-ms masking stimulus, but not different from that in the Single-condition with the 16-ms masking stimulus. CONCLUSIONS: The peak latency of one major response depended on the onset of the first stimulus for both the target stimulus with masking stimulus conditions and Single-condition, but the root mean square value depended on the duration of the masking stimulus. We concluded that the temporal information for the target stimulus was preserved during the masking effect, while the figural information was interrupted by the masking stimulus. Our results suggested that temporal factors for the stimulus were processed differently from those responsible for the object's recognition during backward masking.