Women show enhanced proprioceptive target estimation through visual-proprioceptive conflict resolution

Abstract

To form a unified and coherent perception of the organism's state and its relationship with the surrounding environment, the nervous system combines information from various sensory modalities through multisensory integration processes. Occasionally, data from two or more sensory channels may provide conflicting information. This is particularly evident in experiments using the mirror-guided drawing task and the mirror-box illusion, where there is conflict between positional estimates guided by vision and proprioception. This study combined two experimental protocols (the mirror-box and the mirror-guided drawing tasks) to examine whether the learned resolution of visuo-proprioceptive conflicts in the mirror-guided drawing task would improve proprioceptive target estimation of men and women during the mirror-box test. Our results confirm previous findings of visual reaching bias produced by the mirror-box illusion and show that this effect is progressively reduced by improvement in the mirror drawing task performance. However, this was only observed in women. We discuss these findings in the context of possible gender differences in multisensory integration processes as well as in embodiment.

Keywords: embodiment; mirror box illusion; mirror drawing; multisensory integration; star-tracing; visuo-proprioceptive conflict; visuo-proprioceptive recalibration; visuomotor adaptation.

Figure 1

Schematic of target-reaching task, from the researcher’s perspective. (A) Initial task setup at the initial position of 26 cm. (B) Measurement of reaching error, after the participant’s reaching attempt.

Figure 2

Schematic of the star-tracing task, from the researcher’s perspective.

Figure 3

Visual capture effect. Bar graph depicting the mean target estimation errors made by participants in the reaching task for the mirror and no-mirror conditions. The x-axis shows the initial position of the hidden left hand in centimeters, and the y-axis shows the mean reaching errors in cm. Bars are means ± SEM. *p < 0.05; ***p < 0.001.

Figure 4

Evolution of reaching errors in the ‘mirror’ condition. Upper and lower panels show the errors for ‘initial position’ of 5 and 26 cm, respectively. Left and right panels show the errors for ‘reaching phase’ before and after the mirror drawing task, respectively. (A) Mean reaching errors for ‘reaching phase’ before the mirror drawing task, at the ‘initial position’ 5 cm. (B) Mean reaching errors for ‘reaching phase’ after the mirror drawing task, at the ‘initial position’ 5 cm. (C) Mean reaching errors for ‘reaching phase’ before the mirror drawing task, at the ‘initial position’ 26 cm. (D) Mean reaching errors for ‘reaching phase’ after the mirror drawing task, at the ‘initial position’ 26 cm. Symbols are means ± SEM. *p < 0.05, **p < 0.01, and ***p < 0.001.

Figure 5

Evolution of accuracy and speed indices for the mirror drawing task. (A) Mean accuracy indices for men and women in the mirror drawing task across sessions. (B) Mean speed indices for men and women in the mirror drawing task across sessions. Symbols are means ± SEM. **p < 0.01, and ***p < 0.001.

Figure 6

Correlations between reaching errors in ‘mirror’ condition and mirror drawing indices. Upper and lower panels show scatter plots and tendency lines for ‘initial position’ of 5 and 26 cm, respectively. Left and right panels show the scatter plot and tendency lines for mirror drawing accuracy and speed indices, respectively. (A) Correlation between reaching errors at 5 cm ‘initial position’ and Accuracy Indices. Women: r = −0.456, p < 0.05; Men: No significant correlation. (B) Correlation between reaching errors at 5 cm ‘initial position’ and Speed Indices. No significant correlations found for women and men. (C) Correlation between reaching errors at 26 cm ‘initial position’ and Accuracy Indices. Women: r = −0.462, p < 0.05; Men: No significant correlation. (D) Correlation between reaching errors at 26 cm ‘initial position’ and Speed Indices. Women: No significant correlation. Men: r = −0.504, p < 0.05. Solid and traced lines represent linear fits. Accuracy and speed indices showed a positive correlation of 0.579 (p < 0.001).