It looks like all the abstracts from the Vision Sciences Society conference we were at a few months ago have now been published in the Journal of Vision. Below is the text of the abstract for the poster I presented. If you have any questions, please leave a comment!
We have previously demonstrated that transcranial magnetic stimulation (TMS) can cause the re-perception of recently presented visual stimuli. Here we find that such replays can be experienced for natural scene stimuli, with a level of detail suggesting low-level rather than iconic representations.
TMS was administered using a Magstim dual-pulse setup sending pulses with 50 ms separation through a figure-8 coil. The coil position over occipital cortex was optimized to elicit vivid flashes of brightness (phosphenes) in a darkened room. We screened subjects to find those that perceived large, bright phosphenes near fixation. To these subjects (N=7), we presented pictures of natural scenes and animals for 100 ms, followed by TMS at various ISIs. Subjects provided verbal descriptions, subjective ratings, and drew figures on the screen.
While TMS in a stable visual environment generally elicits phosphenes that are consistent across trials, colorless, and internally featureless, we found that TMS delivered shortly after image presentation led to the perception of clearly defined forms that varied according to the content of the flashed image.
In this experiment, five out of the seven subjects reported percepts that drew from the preceding images. In the most vivid cases, these would appear to be nearly photographic repetitions in portions of the display. In other cases, subjects would perceive uniformly-filled, phosphene-like figures whose outlines matched, in detail, contours drawn from the preceding image (abstract by Wu et. al. describes double-blind validation of these effects). In early trials, subjects experienced the most vivid replay effects within narrow time windows, which varied from subject to subject between 150-250 ms. With continued stimulation, longer ISIs (as much as one second) became effective.
This study indicates that rich, detailed visual information remains encoded well after visual perception has ended, and that TMS can allow conscious access to these nascent low-level representations.
Also, here’s the abstract for our complementary poster: TMS “instant replay” validated using novel double-blind stimulation technique