Treadmill experience alters treadmill effects on perceived visual motion

Abstract

Information on ongoing body movements can affect the perception of ambiguous visual motion. Previous studies on "treadmill capture" have shown that treadmill walking biases the perception of ambiguous apparent motion in backward direction in accordance with the optic flow during normal walking, and that long-term treadmill experience changes the effect of treadmill capture. To understand the underlying mechanisms for these phenomena, we conducted Experiment 1 with non-treadmill runners and Experiment 2 with treadmill runners. The participants judged the motion direction of the apparent motion stimuli of horizontal gratings in front of their feet under three conditions: walking on a treadmill, standing on a treadmill, and standing on the floor. The non-treadmill runners showed the presence of downward bias only under the walking condition, indicating that ongoing treadmill walking but not the awareness of being on a treadmill biased the visual directional discrimination. In contrast, the treadmill runners showed no downward bias under any of the conditions, indicating that neither ongoing activity nor the awareness of spatial context produced perception bias. This suggests that the long-term repetitive experience of treadmill walking without optic flow induced the formation of a treadmill-specific locomotor-visual linkage to perceive the complex relationship between self and the environment.

Figure 1. The schematic illustrations of the experimental conditions and hypotheses.

(a) Experimental conditions, W (left), Str (center), and Sfl (right) show the conditions of walking, standing on a treadmill, and standing on the floor, respectively. The apparent motion stimulus of horizontal gratings is presented in front of the participants' feet in each condition. (b) Hypotheses 1 (above) and 2 (below) of Experiment 1. The bold striped arrow indicates the participants' assumption of optic flow of ground surface that produces downward bias to the perceived direction of the stimuli under the condition. The bold striped square indicates the participants' assumption of zero optic flow of ground surface that produces zero bias to the perceived direction of the stimuli. The thin striped arrow indicates the participants' assumption of being on a treadmill, which produces downward bias to the perceived direction of the stimuli. In Hypothesis 1, the downward bias under Condition W is larger than that under the other two standing conditions, Str and Sfl. In Hypothesis 2, the downward bias under Condition W is larger than that under Condition Str. In addition, the downward bias under Condition Str is larger than that under Condition Sfl. (c) Hypotheses 1 (above) and 2 (below) of Experiment 2. In Hypothesis 1, there is no downward bias under the three conditions. In Hypothesis 2, the downward biases under Conditions W and Str are larger than that under Condition Sfl.

Figure 2. Experimental setup.

(a) Orthographic drawing of top view (top left), front view (bottom left), and right-side view (bottom right) of the monitor, the table, and the mirrors (depicted in blue). Each mirror was arranged to show the stimulus in front of the observer's feet. The light ray from the CRT monitor is drawn as yellow solid lines. The line of sight to the mirror image is drawn as yellow dashed lines. (b) Condition W. (c) Condition Str. (d) Condition Sfl.

Figure 3. Procedure.

Upper part represents the design of the sessions. The square with ‘W’, ‘Str’, and ‘Sfl’ represents the block of Condition W, Str, and Sfl, respectively. The order of the three conditions was randomized in every session. They took rest for about 6 min between sessions. Lower part represents the design of a block. Each block which takes approximately 5 min consisted of 22 trials and presented 11 distinct stimulus displays shown two times in a random order. Participant reported the direction of motion during interval (10 s) following stimulus display (3 s).

Figure 4. Results of Exp. 1 with non-treadmill runners (nTRs).

(a) Probability that the nTRs perceived the direction of motion of the stimulus as “downward” is plotted against the shift of the gratings. Circles, crosses, and triangles show the response under Conditions W, Str, and Sfl, respectively. The probabilities of “downward” responses as a function of the grating shift under each condition are fitted with a psychometric function. The solid, dashed, and dotted lines show the fitted curve under Conditions W, Str, and Sfl, respectively. (b) Mean ratio of “downward” response (RDR) under each condition. Error bars indicate standard deviation (SD). Asterisks on the top represent a significant difference between the conditions (* p<.05 and *** p<.001). Asterisk on the bottom represents a significant difference from the chance level (p<.05). (c) Mean point of subjective equality (PSE) under each condition. Error bars indicate SD. Asterisks and a dagger on the top represent significant and marginally significant differences between the conditions, respectively (** p<.01 and † p = .04). Asterisk on the bottom represents a significant difference from the chance level (p<.05).

Figure 5. Results of Exp. 2 with treadmill runners (TRs).

(a) Probability that the TRs perceived the direction of motion of the stimulus as “downward” is plotted against the grating shift. The probabilities of “downward” responses as a function of the shift of the gratings under each condition are fitted with a psychometric function. See the caption of Figure 4a for a description of the symbols and lines. (b) Mean RDR under each condition. Error bars indicate SD. (c) Mean PSE under each condition. Error bars indicate SD.