The relative timing of active and passive touch

Abstract

Tactile stimulation usually occurs as a combination of an active movement (reaching out to touch a surface) and a sensation (actually feeling the surface against the skin). The brain has information about the active component (the motor command) before it occurs because of efference copy, while the passive component must be transduced before it can be processed. Since the active and passive tactile components are available to the brain at different times, determining the time of touch requires calculation worked backwards from the passive sensation, and/or worked forward from the active motor command. In order to determine which touch process is perceived more quickly, we varied the relative delay between an active and a passive touch signal and determined the relative time percieved as simultaneous. A passive touch needed to be presented before an active key was pressed in order for the two touches to be perceived as simultaneous, but this timing difference was not significant. In order to test the plasticity of the active and passive touch systems, we exploited the fact that the point of subjective simultaneity between two stimuli can sometimes be altered by repeated exposure to asynchronous presentation. We exposed subjects to an active key press/ passive touch pair delayed by 250 ms. This exposure increased the range of relative delays between active and passive touches at which the pairs were judged as simultaneous. This is consistent with an adaptive change in the processing of active touch.

Introduction

When a person reaches out and touches something there are two aspects to the event: the plan to reach out, and the sensory feedback once the action is completed. The plan consists of a decision to make the movement and an intention to move a part of the body from one place to another. This plan is then converted into a series of muscle activations which eventually carry out the movement. A copy of the motor command is available to many parts of the brain even before the movement occurs. This copy is known as the efference copy (von Holst and Mittelstaedt, 1950). In contrast, sensory feedback can only occur following the movement.

Libet et al. (1983) suggest that the efference copy for an active movement occurs as long as 250 ms before the movement. Efference copy can therefore be used to predict the timing of an action and to make anticipatory compensations (Duhamel et al., 1992, Morrone et al., 2005). In touch, an “active movement”, with its accompanying efference copy, may have an advantage in being able to predict the time of contact, over “passive touch” which only has sensory feedback (Kornhuber and Deecke, 1965). The anticipation of the active movement might allow for compensation in the neural delays which are inherent in the tactile system when determining the timing of a touch sensation.

When trying to determine when sensory stimuli were experienced (such as when making temporal order judgments), the relative timing of the stimuli needs to be reconstructed in memory (Libet et al., 1983, Lau et al., 2004). Obviously this reconstruction occurs after the event, but perhaps it could be done more accurately with the aid of efference copy. Therefore we want to determine the relative timing of the perception of the time of occurrence of an active key press (with efference copy) and a passive touch (with only the feedback from the touch). Would the active key press be perceived to be felt before the passive touch as a result of its efference copy advantage? Or would the sensory component of an active key press be unaffected by the fact that it is the consequence of a motor act? Currently, there are no published studies that have tested the perceived timing between active and passive touches. We hypothesize that, in order for an active and passive touch to be perceived as simultaneous, the passive touch will need to occur before the active key press to make up for the hypothesized backwards displacement in time enabled by the efference copy.

Since the efference copy is available up to 250 ms before the event and the sensory feedback arrives in the brain some 40 ms after the event (Macefield et al., 1989), matching the timing of the two requires a flexible system. Previous experiments have found that after repeated exposure to asynchronously presented multisensory stimulus pairs, the timing at which the stimuli are perceived as simultaneous can change (Fujisaki et al., 2004, Vroomen et al., 2004, Harrar and Harris, 2005). Changes can be found in both the PSSs (point of subjective simultaneity) and JNDs (just noticeable differences: the range of relative times perceived as simultaneous). A JND increase corresponds to a widening of the temporal window of stimulus staggers that are accepted as simultaneous. A PSS shift corresponds to a new estimate of simultaneity.

As the efference copy provides these active movements with additional information, it might be expected that active movements would be less responsive than other multisensory systems to sensory feedback, indicating a need for recalibration. Stetson et al. (2006) paired an active touch with a light presented with a delay of 135 ms and found a PSS shift towards light first, treating this new delay as the new simultaneity. Similarly, Cunningham et al. (2001) found the temporal perception of an active touch system to be adaptable in response to realistic and complex delayed stimuli. These observations suggest that the active system may indeed be as flexible as the processing of multisensory stimuli.

The perceived timing of passive touches however does seem to be determined more rigidly (Harrar and Harris, 2008). Navarra et al. (2007) found only a small JND increase after repeated exposure to audio-tactile pairs with an auditory stimulus leading by 75 ms. Although Keetels and Vroomen (2008) found a small PSS shift in the direction of the exposed temporal stagger following exposure to staggered visuo-tactile pairs, it is unknown if pairing a time-staggered active and passive touch will result in a recalibration. Previous experiments have shown that recalibration can result in the misordering of two passive stimuli. Therefore, here, a recalibration could result in the perception that subjects hit the key before they actually did.

Can the temporal perception of an active touch be recalibrated in a similar way to what has been demonstrated for the temporal perception of passive multisensory stimuli? Is the temporal perception of the efference copy as flexible as the temporal perception of sensory feedback?

Given that previous experiments show some flexibility in both the active and passive touch systems, we hypothesize that following exposure there will be some change in either JND or PSS of an active and passive touch pair.