Whereas sensory processing latency is altered by many factors including stimulus intensity and processing channel, the simultaneity judgment between two sensory signals is generally much more accurate (i.e., more consistent with the original event timings) than predicted from response latencies to those signals. This suggests that apparent simultaneity is based on the "time markers" of the input signals, which are different from the timings of signal detection in the brain. The neural correlate of time marker however remains unknown. Here we measured brain activity with magnetoencephalography (MEG) while the subjects performed one of two tasks. In the simple reaction time (RT) task, subjects responded to an onset of random-dots coherent motion. In the synchrony judgment task, they judged simultaneity between coherent motion onset and a beep sound for several SOAs. For both tasks, coherence of random-dots motion changed abruptly from 0% to 30, 40 or 90% (step) or gradually from 0% to 90% at the rate of 80, 120 and 200 %/s (ramp). We found that motion coherence affected RT and the point of subjective simultaneity (PSS) differently. For the step stimuli, increasing motion coherence decreased RT but had no effect on PSS. For the ramp stimuli, on the other hand, motion coherence affected not only RT but also PSS. These modulations of RT and PSS could be predicted by the timing when leaky-integrated hMT+ response crossed certain thresholds. The threshold for PSS was smaller than that for RT, indicating that the time marker for simultaneity judgment is located at the timing earlier than the detection latency. We suggest that synchronous perception is based on a comparison of the time marker assigned at the timing when leaky-integrated sensory response crossed a relatively low threshold. This time marker is considerably independent of stimulus amplitude, and thus contributes to accurate timing perception.
展开▼
