Abstract
Localizing touch on the skin requires integration of multiple spatial signals, including reference landmarks and motion cues. It is well known that motion patterns can bias the perceived endpoint of motion. However, it is unknown whether static touch presented post-motion is also distorted. To investigate this, we presented space-changing motion patterns and tested position perception 1 s and 10 s post-motion. We used a brush moving along the forearm at 15 cm s(-1), brushing 4.5 cm skin patches near the elbow and near the wrist, skipping a 10-cm long metal-shielded patch in the middle ('numb spot'). It accelerated to 100 cm s(-1) across the shielded gap in an attempt to create an illusion of continuous motion between the separate brushed areas. After several such deceptive motions, 12 participants indicated the locations touched by a von Frey filament near the elbow and wrist, all within the previously brushed areas. Localization responses shifted 4-10 mm towards the numb spot in the skipped-patch condition compared with controls with either continuous brushing across the full forearm, or brushing the same patches without acceleration. This spatial distortion was equally strong 1 s and 10 s after motion offset with only isolated location-specific differences between delays. In addition, participants' sketches indicated a reduction in perceived gap size. We propose that participants used the brushed fields as reference frame for localisation, with the high-velocity motion compressing the perceived space between them. This means that motion-defined boundaries can serve as spatial landmarks for static touch.