Neural edelmanism

Abstract

As with some patients with primary visual cortex (V1) damage, transcranial magnetic stimulation (TMS) over V1 reliably induces blindsight, whereby observers can correctly discriminate the attributes of visual stimuli despite being unable to detect them. This TMS-induced blindsight has been demonstrated to reflect a form of unconscious vision that relies upon different neural pathways than with conscious vision. However, the timing of the neural processes mediating TMS-induced blindsight has been unclear, especially when considering suggestions that TMS interferes with feedback processes to V1 that mediate conscious visual perception. To better elucidate the neural mechanisms that give rise to blindsight, we tested TMS-induced blindsight for the orientation of visual stimuli across a range of stimulus onset asynchronies (SOAs) to assess how different latencies of visual cortex disruption, relative to a visual stimulus, affect detection rates and forced-choice discrimination accuracy. At all TMS latencies, including at SOAs with substantial visual suppression from TMS, discrimination performance was significantly above-chance, demonstrating the consistency of TMS-induced blindsight. Crucially, we observed two windows of maximum visual suppression from TMS at SOAs between 65 and 105 ms, but consistent above-chance discrimination performance accuracy across these windows. However, at longer SOAs, detection and discrimination covaried, suggesting a dependency of discrimination performance on detection only when detection rates exceed threshold levels of normal vision. Taken together, these results indicate that unconscious discrimination occurs independently of detection, including at TMS intervals that optimally interfere with conscious visual perception. They further suggest that forced-choice discrimination is less dependent on feedback processes to V1 than visual awareness and that TMS-induced blindsight is not the same as near-threshold vision.