Sensorimotor adaptation to delayed visual feedback of the whole field of view

If humans wear prism goggles that spatially displace their visual field, they adapt their behaviour and perception with training and compensate for the visuomotor mismatch. This adaptation is of a semi-permanent nature, i.e., it persists for some time after the goggles are removed (aftereffect) and transfers to tasks that participants were not explicitly trained on. Here we tested whether this kind of adaptation is also possible in the temporal domain. Participants wore a head-mounted display with two cameras in front of the eyes that relay images in real time to the display. They were trained for one hour to perform sensorimotor tasks wearing this device (e.g., arranging blocks, walking down a corridor, playing musical instruments). In the delay condition, visual feedback from the cameras was artificially delayed by 150 ms during training. In the control condition, no additional visual lag was present. Participants improved their coordination markedly with training and reported that interaction with the world felt more coherent after training. To measure temporal aftereffects, participants were tested in three sensorimotor tasks without delay before and after training: manual tracking, rotational synchronization, and target interception. These test tasks were not used during training. We observed a transfer of delay adaptation to these tasks in the order of 50-100 ms. As expected, the aftereffects were stronger in the delay condition than in the control condition. These results show that adaptation to delayed feedback of the whole visual field is semi-permanent, with aftereffects and generalization across tasks, as in spatial prism adaptation