Awareness of voluntary action, rather than body ownership, improves motor control

Abstract

Awareness of the body is essential for accurate motor control. However, how this awareness influences motor control is poorly understood. The awareness of the body includes awareness of visible body parts as one's own (sense of body ownership) and awareness of voluntary actions over that visible body part (sense of agency). Here, I show that sense of agency over a visible hand improves the initiation of movement, regardless of sense of body ownership. The present study combined the moving rubber hand illusion, which allows experimental manipulation of agency and body ownership, and the finger-tracking paradigm, which allows behavioral quantification of motor control by the ability to coordinate eye with hand movements. This eye-hand coordination requires awareness of the hand to track the hand with the eye. I found that eye-hand coordination is improved when participants experience a sense of agency over a tracked artificial hand, regardless of their sense of body ownership. This improvement was selective for the initiation, but not maintenance, of eye-hand coordination. These results reveal that the prospective experience of explicit sense of agency improves motor control, suggesting that artificial manipulation of prospective agency may be beneficial to rehabilitation and sports training techniques.

Figure 1

Apparatus, stimuli, and procedure. (a) Participants binocularly viewed the visual stimulus presented through a head-mounted display and placed their right hand on a table. Their right index finger was attached to the arm of a force-feedback device. (b) A computer graphics (CG) hand and two white points were presented on a gray surface in the virtual environment. The right index finger of the CG hand was placed 10° to the left of the participant’s unseen right index finger. One white point served as the starting point of eye and hand movements, and the other served as the turning point. (c) Sample result of eye and hand positions for one participant. (d) On a given trial, the participant’s hand was moved actively or passively to induce the moving rubber hand illusion. Then, the participants were instructed to track the CG hand movements with pursuit eye movements.

Figure 2

Results of experiment 1. (a, b) Strength of the rated ownership and agency of the computer graphics (CG) hand shown as box-and-whisker plots for the four conditions. The ownership and agency indexes were defined as the difference in ratings between the illusion and control questions. (a) The standard score (z-score) was calculated for each of the ownership (blue) and agency (pink) indexes, which were defined as the standardized index. Horizontal black bars represent medians, and the boxes denote the interquartile ranges. Whiskers extend to the farthest data points within 1.5 times the interquartile range from the top and bottom edges of the boxes. Individual open circles represent outliers. (b) The standardized ownership index was subtracted from the standardized agency index. Colored bars represent the medians, and the error bars represent the interquartile ranges. (c, d) Eye latency, as defined by eye onset time relative to CG hand onset time. Orange and purple symbols represent the active and passive hand movement conditions, respectively, for individual participants. Square symbols represent the mean ± standard deviation. (e, f) Pursuit gain, defined as eye velocity divided by CG hand velocity after excluding saccades in velocity traces. (e) Shaded areas represent standard deviation. (f) Shows the average pursuit gain in the range of 0 to 1000 ms. (g, h) Proportion of saccadic trials in the range of − 200 to 1600 ms from CG hand onset. Results are the mean ± standard deviation in Figs. 2d–h (n = 20).

Figure 3

Correlation between body awareness and eye–hand coordination performance in experiment 1 (n = 20). (a–c) Agency. (d–f) Ownership. Change in eye–hand coordination performance [i.e., eye latency (a, d), pursuit gain (b, e), and proportion of saccadic trials (c, f)] was defined as the change in performance when the sense of agency or the sense of body ownership became stronger in the active-synchronous condition compared with the active-asynchronous condition. One participant was excluded from Spearman’s rank correlation analysis for both eye latency and pursuit gain (see Methods for details).

Figure 4

Results of experiment 2. (a, b) Strength of the rated ownership and agency of the computer graphics (CG) hand shown as box-and-whisker plots for the four conditions. The ownership and agency indexes were defined as the difference in ratings between the illusion and control questions. (a) The standard score (z-score) was calculated for each of the ownership (blue) and agency (pink) indexes, which were defined as the standardized index. Horizontal black bars represent medians, and the boxes denote the interquartile ranges. Whiskers extend to the farthest data points within 1.5 times the interquartile range from the top and bottom edges of the boxes. Individual open circles represent outliers. (b) The standardized ownership index was subtracted from the standardized agency index. Colored bars represent the medians, and the error bars represent the interquartile ranges. (c, d) Eye latency, as defined by eye onset time relative to CG hand onset time. Orange and purple symbols represent the active and passive hand movement conditions, respectively, for individual participants. Square symbols represent the mean ± standard deviation. (e, f) Pursuit gain, defined as eye velocity divided by CG hand velocity after excluding saccades in velocity traces. (e) Shaded areas represent standard deviation. (f) Shows the average pursuit gain in the range of 0 to 1000 ms. (g, h) Proportion of saccadic trials in the range of − 200 to 1600 ms from CG hand onset. Results are the mean ± standard deviation in (d–h) (n = 29).

Figure 5

Correlation between body awareness and eye–hand coordination performance in experiment 2 (n = 29). (a–c) Agency. (d–f) Ownership. Change in eye–hand coordination performance [i.e., eye latency (a, d), pursuit gain (b, e), and proportion of saccadic trials (c, f)] was defined as the change in performance when the sense of agency or the sense of body ownership became stronger in the active-synchronous condition compared with the active-asynchronous condition.