Proceedings of The Physiological Society

University College London December 2005 (2006) Proc Physiol Soc 1, C10

Oral Communications

Speeding up fast visually-guided human step adjustments with a startle

Reynolds, Raymond Francis; Day, Brian L;

1. Sobell Department, Institute of Neurology, London, United Kingdom.


  • Figure 1. Mean medio-lateral foot acceleration (minus control)

The time taken to respond as quickly as possible to a visual stimulus may be shortened by a simultaneous startling sound if the required motor response is known in advance [1]. However, there is conflicting evidence as to whether the same is true for reaction tasks in which the required movement is uncertain until it is cued by the visual imperative stimulus [2,3]. A major difference between these two situations is that in the second case the motor response may not be able to be fully prepared prior to the stimulus. If the speeding up occurs by the startle rapidly releasing an already prepared motor program, then the startle may be ineffective in the choice reaction situation. To investigate this further we studied the effects of a startle upon visually-guided mid-step adjustments evoked by a jumping foot target. This target-jumping task is equivalent to a choice reaction task except that responses occur much faster than in conventional visual reaction tasks (~120ms [4] compared to >200ms). Five subjects stepped 198 times to a rectangular 21x14cm target lit up ahead of their right foot. At the point of foot-off, the target was randomly made to jump 21cm medially or laterally in 1/3rd of trials. Of these trials, 12 included a startling acoustic stimulus delivered through headphones simultaneously with the target jump (120dB, 50ms duration). To examine any non-specific effects of the startle, there were also 4 acoustic stimuli given during control (no jump) trials. Medio-lateral foot acceleration was derived from an infra-red hallux marker recorded with a 3-D motion capture system (CODA mpx30). Control trials were subtracted from target-jump trials for statistical analyses. During all target-jump trials subjects accelerated the foot at short latency in the appropriate direction to intercept the target. At 125ms latency in non-startle trials, foot acceleration was just underway (0.47

Where applicable, experiments conform with Society ethical requirements