Agency judgments in post-stroke patients with sensorimotor deficits

Abstract

Sense of agency refers to the feeling of being in control of one's actions. Previous research has demonstrated that sense of agency is produced through the sensorimotor system, which is involved in comparing internal predictions with sensory feedback in motor control. Therefore, sensorimotor deficits might impair agency through a sensorimotor system malfunction. The present study examined this hypothesis by investigating post-stroke patients who had suffered a subcortical stroke that damaged regions associated with sensorimotor function. To examine agency judgments with respect to motor control, we adopted a selfother attribution task and applied it to post-stroke patients. Participants traced a horizontal straight line and received visual feedback through a cursor on a monitor. The cursor movement reflected either the participants' actual movement or the movement of an "other" that had been previously recorded. Participants judged whether the cursor movement reflected their own movement (self) or an other's movement while they engaged in four cycles of the horizontal tracing movement. After each trial, participants reported their self-other judgment on a nine-point scale. Post-stroke patients completed the experiment with their paretic as well as their non-paralyzed upper limbs. Compared to healthy controls, patients made significantly more self-attributions of others' movements. Interestingly, such misattributions were observed in the patients' performance using both paretic and non-paralyzed upper limbs. These results suggest that post-stroke patients with sensorimotor deficits form misattributions that cannot be explained solely by the sensorimotor system's role in motor control. We discuss these misattributions in post-stroke patients in terms of cue integration theory.

Dynamic Relationship between Sense of Agency and Post-Stroke Sensorimotor Deficits: A Longitudinal Case Study

“…Regarding the results of motor performance, the difference in movement errors between the SELF and OTHER conditions was observed in each experiment week. This result replicates that of the previous studies [15,16,32], suggesting that PA and PB correctly completed the movement tasks, just like the participants in their previous studies did. Although movement errors of PA were greater in the OTHER condition than those of PB, there were no obvious differences in movement errors across the second, fourth, and eighth weeks in the performances of PA and PB.…”

“…This modification was done because we focused on the self-other attributions in a nonparalyzed upper limb use, and we needed to reduce the number of experiments to ease the burden on the patients as much as possible. The previous study confirmed that there were no significant differences in self-other judgments between the right and left upper limb uses [15].…”

“…The procedure performed in this study was similar to that from our previous study [15] (i.e., modified Asai's [16] self-other attribution task) with two modifications. One of the modifications was that the experiments were conducted at the second, fourth, and eighth weeks after the stroke.…”

“…To confirm whether PA and PB completed the movement task correctly, we compared movement errors between the SELF and OTHER conditions. In this task, the movement errors in the SELF condition should be smaller than those in the OTHER condition [15,16,32]. Since differences in movement errors across cycles were not of interest, the average movement error for each condition across cycles was calculated (Figure 3).…”

“…To examine these possibilities, we conducted a longitudinal case study using our previous study paradigm [15] (i.e., modified Asai's [16] self-other attribution task). The agency dynamics in two post-stroke patients were examined: Patient A (PA) and Patient B (PB); PA with sensorimotor deficits rarely used her paretic upper limb in her daily life; and PB, with very mild sensorimotor deficits, almost used his paretic upper limb normally.…”