The effect of trial number on the emergence of the 'broken escalator' locomotor aftereffect.

Walking onto a stationary platform, which had been previously experienced as moving generates a locomotor aftereffect (LAE), which resembles the 'broken escalator' phenomenon. Experimentally, this is achieved by having subjects walk initially onto a stationary sled (BEFORE condition), then onto a moving sled (MOVING condition, or adaptation trials) and then again onto the stationary sled (AFTER condition). Subjects are always appropriately warned of the change in conditions. In this paper, we ask how many adaptation trials are needed to produce such a LAE. Thus, in two experiments, the number of MOVING trials was varied between 20 and 5 (Experiment 1) and between 8 and 1 (Experiment 2). Gait velocity, trunk position, foot contact timing and EMG of the ankle flexor-extensors muscles were measured. In comparison with BEFORE trials all groups in the AFTER trials walked inappropriately fast, experienced a large overshoot of the trunk and showed increased leg EMG, indicating that all groups showed a LAE. In each experiment, and for all variables, no significant difference between the groups (i.e. 20 down to one MOVING trials) was found. The study shows that this LAE, in contrast to other motor aftereffects reported in the literature, can be generated with only one or two adaptation trials and without requiring unexpected 'catch' trials. The fast aftereffect generation observed is likely to depend on two types of mechanisms: (1) the nature of the sensorimotor adaptation process, involving multiple sensory feedbacks (visual, vestibular and proprioceptive), anticipatory control and large initial task errors and (2) the involvement of two phylogenetically old neural mechanisms, namely locomotion and fear. Fear-relevant mechanisms, which are notably resistant to cognitive control, may be recruited during the adaptation trials and contribute to the release of this LAE.