Abstract
Humans effortlessly compensate for feedback delays due to neural processing and physical transmission times in closed-loop motor control. With training, they can also adapt to additional latencies but the nature of adaptation is unclear. Three alternative processes have been suggested: (1) control stabilization by decreased control gain (2) spatial prediction of control signal (3) temporal recalibration of motor prediction (forward model). Only temporal recalibration can be expected to also recalibrate time perception (shift in perceived simultaneity). At present, most of the evidence points towards the first two options. Our hypothesis is that instead all three processes can occur and depend on the nature of the task. Particularly, we hypothesize that temporal recalibration is only possible with continuous feedback and a predictable control signal. To test this hypothesis, we trained participants with a 200ms feedback delay in a visually-guided manual-tracking task, varying the predictability of the reference signal between conditions, but keeping reference motion and delayed visual feedback constant. In Experiment 1, we focused on motor behavior (spatial error, temporal error, and spectral power of movement). We observed that only predictable training brings about negative aftereffects in all three measures. This combination of negative aftereffects indicates temporal recalibration of the forward model. By contrast, the results from unpredictable training indicate compensation by decreased control gain. We then tested whether delay adaptation in tracking also transfers to a perceptual task. After predictable but not unpredictable delay adaptation, we observed a 30ms shift of perceived simultaneity in a synchronization task (Experiment 2) and 38ms shift in an interval estimation task (Experiment 3). These results show that temporal recalibration of behavior and time perception is possible with a predictable control signal. The decreased control gain after unpredictable training confirms that different adaptive processes are evoked depending on the task.
Meeting abstract presented at VSS 2014