Investigating the relationship between visual working memory and visual awareness: evidence from TMS and OSM

Human vision seems so fast and effortless that we often don’t consider the processes that take place before we have the perceptual experience of “seeing”. However, by the time a stimulus first emerges into consciousness it has already been coded at various stages within the visual system. Recent evidence from patients with visual extinction has revealed that the active maintenance of an item in working memory facilitated enhanced awareness of targets that matched a previously viewed memory cue (Soto & Humphreys, 2006). Here, we asked whether the contents of working memory could facilitate efficient visual perception in normal subjects by triggering top-down signals that bias the visual system in favour of a preactivated object representation. Specifically, we hypothesised that target stimuli that matched a previously presented memory cue would be less susceptible to visual masking. In Experiments 1 and 2, we used transcranial magnetic stimulation (TMS) to the occipital pole as a method to degrade visual perception. However, we were unable to adequately mask targets and we instead provide a detailed discussion of the difficulties in localising area V1 using external anatomical landmarks. In Experiment 3, we used object substitution masking (OSM) to reduce awareness of target stimuli. In a typical OSM paradigm, visual masking occurs when a four-dot mask persists on the screen after target offset, whereas target detection is unimpaired by the simultaneous offset of the target and mask (Enns & Di Lollo, 1997). As predicted, we observed a significant main effect of Mask Offset. Detection sensitivity was higher on simultaneous-offset trials, relative to delayed-offset trials. In contrast, the interaction between Memory Match and Mask Offset was not significant. To overcome the limitations of inferential statistics in interpreting null findings, we implemented Bayesian statistics and obtained substantial evidence in favour of the null hypothesis. We conclude that merely holding something in working memory is not sufficient to enhance awareness of degraded visual objects.