Abstract
During smooth pursuit eye movements, retinal motion can be reafferent (caused by the eye movement) or represent motion in the world. An efference copy can be used to derive a reference signal, or amount of reafference to subtract from the image. However, the accuracy of this computation can be ensured through an image-based adaptive mechanism. Indeed, exposure to systematic background motion relative to the pursuit direction leads to a shift in the point of subjective stationarity (PSS). We aimed at elucidating the nature of this adaptive mechanism. A recalibration account predicts a shift in reference signal (i.e. predicted reafference), resulting in a shift of PSS, but no change in discrimination thresholds. An effect akin to motion adaptation contingent on pursuit direction would affect discrimination thresholds and the PSS. Participants performed a two-alternative forced-choice directional judgement task (left or right) as well as a confidence judgement task. Confidence in the directional judgements should be minimal at the PSS, unless responses are not driven by sensory evidence, allowing us to rule out a response bias [Gallagher, Suddendorf, & Arnold, (2019). Scientific Reports]. As shown before, exposure to background motion relative to the pursuit direction (random-dots, shown for 200 ms in the middle of the horizontal pursuit trajectory) shifted the PSS (test trials) towards the exposed motion, but only when tested on the exposed visual field and not the opposite visual field. Effect of exposure (same vs. opposite visual field) were of similar amplitude whether assessed with the PSS or the point of minimal confidence, indicating no response bias. Both judgements favour a recalibration account, whereby there is an adaptive shift in reference signal caused by the prevailing retinal motion during pursuit, as measured (at most) within a visual hemifield.