Abstract
Binocular masking is a particularly interesting means of suppressing human visual awareness, as the images it renders subjectively 'invisible' nonetheless excite robust activity in human cortex. Recently, binocular masking has been leveraged to show people can be trained to better interact with inputs that remain, subjectively, invisible. Here we ask what is learnt in such circumstances. Do people gain better insight into their' true level of sensitivity, allowing motor planning to be better guided by an unchanged sensory code? Or is signal encoding modulated by training, resulting in a heightened level of objective sensitivity? Our data suggest the latter. We had people train for five consecutive days, to poke targets presented in one of three masked locations. Target intensity was set to a fraction of the participants' pre-training objective detection threshold. During training, people categorised targets as visible or invisible (providing a confidence rating concerning task performance in the latter circumstance). Feedback was provided when the participant had reported not having seen the target, and statistical analyses were restricted to data from these trials. We found that people became better at selecting the target location with training, even when they insisted they could not see it. More important, post-training we found objective thresholds for target detection had improved. We regard our data as evidence that the strength of binocular masking is malleable – reduced by training to localise masked targets, resulting in an enhanced objective sensitivity to masked target location.
Meeting abstract presented at VSS 2016