58 CHAPTER 3 modulated by the biomechanical complexity of the imaginedmovement. More specifically, participants responded faster for stimuli where the hand was rotated to a medial versus a lateral orientation, with respect to the body axis. This became apparent in a significant LATERALITY x ROTATION interaction (F(2.8, 109.4) = 30.9, p < 0.001, part. η2 = 0.44), where stimuli that were rotated clockwise (i.e. 45°, 75°, 105°, 135°) were easier (medial orientation) for left hands, but more difficult (lateral orientation) for right hands, and vice versa for stimuli that were rotated counter-clockwise (i.e.-45°,-75°,-105°,-135°). Post-hoc testing indeed revealed that participants were significantly slower for laterally oriented stimuli than for medially orientated stimuli (F(1,39) = 55.2, p < 0.001, part. η2 = 0.59). This previously documented biomechanical complexity effect shows that participants were sensitive to biomechanical constraints associated with task-related upper limb movements. This confirms that participants did not employ a purely visual imagery strategy to solve the task. Finally, across both groups, participants responded faster to congruent stimuli that matched the current position of their own hand, as compared to incongruent stimuli (significant main effect of POSTURE: F(1, 39) = 18.0, p < 0.001, part. η2 = 0.32; no significant interaction effects involving GROUP: F(1, 39) ≤ 0.2, p ≥ 0.25, part. η2 ≤ 0.04). Moreover, participants responded faster when the posture of their own two hands was symmetric, as compared to trials were the posture of their own two hands was asymmetric, and this result did not influence the main findings reported in this paper (see supplementary materials). The significant effect of POSTURE is an important piece of evidence to indicate that subjects incorporated their own body posture and thus relied on sensorimotor representations, using a first-person rather than a third-person perspective. Correlation with clinical measures The normalized ER for right hands did not correlate with functional capability (SRQ-DLV scores (r =- 0.01, p = 0.96), general DASH or DASH modules (r ≤ ± 0.14, p ≥ 0.56)) in NA patients. The time since the last attack in NA patients was also not correlated with the right normalized ER (ρ = 0.09, p = 0.69). Discussion We tested whether patients with a strictly lateralized peripheral nervous system disorder (neuralgic amyotrophy of the right upper limb) showed altered central sensorimotor representations of their affected right limb. Thus, we compared behavioral performance during the validated hand laterality judgment task, between 21 NA patients and 20 healthy controls. In line with our hypothesis, we found that NA patients showed a specific and significant deficit (increased error rate) when recognizing their affected right limb, as compared to healthy controls. Importantly, our results confirm that both groups used a first-person perspective, i.e. that they used their own body as a reference during imagery: RTs were sensitive to limbspecific biomechanical constraints (lateral or medial rotation with respect to the body axis) and to the position of the subjects’ own arm posture (congruent or incongruent with respect to the stimulus on the screen), in line with previous studies. 58, 69, 70, 83 These findings, especially the posture congruency effect, provide empirical evidence that NA patients did not employ alternative (visual) strategies to solve this task. 84, 85 There is some debate as to which processes underly the hand laterality judgment task. Since
RkJQdWJsaXNoZXIy MjY0ODMw