Abstract
Numerous cortical areas are active while participants relate together visual and egocentric vestibular information. Although dorsal parietal and medial prefrontal cortical activity correlate with visual-vestibular task performance, it is not clear what causal role these dorsal cortical areas may play in the human. We used transcranial magnetic stimulation (TMS) to interfere with activity in parietal or medial prefrontal cortices in groups of between 16-20 healthy people, while they performed the subjective visual vertical (SVV) task. Participants reported with a button press whether a flashed line was tilted counterclockwise or clockwise of true vertical. By fitting psychometric functions, we measured perceptual performance in terms of bias (also referred to as accuracy) versus precision (or sensitivity, threshold, reliability, sigma). In the first study, participants were sorted into two groups of 16 according to their baseline bias at SVV i.e. those with either a slight counterclockwise versus clockwise bias when judging a line to be truly vertical. Right parietal TMS facilitated verticality perception, reducing the difference between groups - affecting bias, with no effect on precision. ERPs suggested that the behavioural TMS effect occurred through normalizing individual SVV biases. No such effects occurred with control stimulation and tasks. In the second study, to ensure a high perceptual demand (putatively necessary to demonstrate a dorsal medial involvement) SVV lines were presented inside pop-out targets within a visual search array. Perceptual performance was analysed before and after theta-burst TMS stimulation of the medial frontal cortex, a control site, or no stimulation, in three groups of 20 people. Medial frontal stimulation improved the precision of verticality judgments with no effects on bias. Taken together, we suggest that human dorsal cortical regions play roles in SVV perception which are causal, dissociable, and attentional.