Does the Side of Stroke Matter? An fMRI Study on the Role of Stroke Laterality on the Action Observation Network

PURPOSE: The priorities of health care are shifting toward value-based care. Understanding how the brain responds to different interventions after stroke using brain imaging can inform occupational therapists’ decisions on treatments that maximize independence and quality of life. The action observation network (AON) is composed of premotor and parietal regions in the brain (Caspers, Zilles, Laird, & Eickhoff, 2010) that are active when doing an action and when observing someone else do an action (Rizzolatti & Craighero, 2004). Garrison, Aziz-Zadeh, Wong, Liew, and Winstein (2013) found that the AON was more engaged when observing hand actions that corresponded to the paretic right hand of individuals who had a left hemisphere stroke. This suggested a potential way of promoting activity on the damaged side of the brain. Yet, since all participants in the study were right-handed/left hemisphere dominant, it was unclear whether the lateralized activity in the AON was due to the brain’s preference of the lesioned side for recovery or the dominance of the side of stroke. The purpose of this study is to determine the role of stroke side in activating the AON. To do this, we compared AON activity in individuals with left hemisphere (LH) stroke, individuals with right hemisphere (RH) stroke, and age-matched healthy control participants.

DESIGN: Thirty-six individuals (12 LH stroke, 12 RH stroke, 12 healthy control) were included in the study. The data from the LH stroke and control groups were previously collected and reanalyzed here (Garrison et al., 2013). All participants were right-handed, had normal or corrected vision, and were safe for MRI scanning. Inclusion criteria were as follows: first stroke, chronic (>3 mo since stroke onset), and middle cerebral artery stroke resulting in upper-extremity motor impairment. We used a quasi-experimental design to examine how each stroke group compared with the control group.

METHOD: We used fMRI to scan participants while they observed videos of goal-directed hand actions (left, right) adapted from the Wolf Motor Function Test. Preprocessing and statistical analyses using a general linear model were carried out in FMRIB Software Library (Beckmann, Jenkinson, & Smith, 2003). The laterality index (LI) for a priori regions of interest in the AON was calculated as LI = (L–R)/(L+R), in which L and R represent surviving active voxels in the left and right hemispheres (Jansen et al., 2006). The average LI was calculated across several thresholds (Z = 1.0, 1.5, 2.3) and ranged from –1 (all right hemisphere activity) to +1 (all left hemisphere activity).

RESULTS: For the control group, all regions showed either bilateral (|LI| < 0.1) or right-dominant lateralized activity (LI < –0.1). Regions in the AON had similar LI values during right and left hand observation, except for the precentral gyrus, which was right dominant during left hand observation and bilateral during right hand observation. We found significant differences in AON activity between the control group and the stroke groups. When directly comparing the RH stroke and LH stroke groups, there were no differences in LI values. When observing right hand actions, all participants with stroke showed LH-dominant lateralized (LI > 0.1) activity for all regions in the AON. During left hand observation, activity in most regions in the AON were bilateral, except for the precentral gyrus (LI = –0.23 for LH stroke group; LI = –0.17 for RH stroke group), which was expected due to its contralateral mapping of the body.

CONCLUSION: For individuals with stroke, activity in the AON was consistent with the side of motor dominance (the same for both stroke groups). This suggests that the dominance of the lesioned hemisphere may affect neural patterns during AON-based therapies. Our findings provide an appreciation for how the dominance of the stroke side could impact the brain’s recovery processes in a way that may affect how people react to different interventions.

References

Beckmann, C. F., Jenkinson, M., & Smith, S. M. (2003). General multilevel linear modeling for group analysis in FMRI. NeuroImage, 20, 1052–1063. https://doi.org/10.1016/S1053-8119(03)00435-X

Caspers, S., Zilles, K., Laird, A. R., & Eickhoff, S. B. (2010). ALE meta-analysis of action observation and imitation in the human brain. NeuroImage, 50, 1148–1167. https://doi.org/10.1016/j.neuroimage.2009.12.112

Garrison, K. A., Aziz-Zadeh, L., Wong, S. W., Liew, S. L., & Winstein, C. J. (2013). Modulating the motor system by action observation after stroke. Stroke, 44, 2247–2253. https://doi.org/10.1161/STROKEAHA.113.001105

Jansen, A., Menke, R., Sommer, J., Förster, A. F., Bruchmann, S., Hempleman, J., . . . Knecht, S. (2006). The assessment of hemispheric lateralization in functional MRI—Robustness and reproducibility. NeuroImage, 33, 204–217. https://doi.org/10.1016/j.neuroimage.2006.06.019

Rizzolatti, G., & Craighero, L. (2004). The mirror-neuron system. Annual Review of Neuroscience, 27, 169–192. https://doi.org/10.1146/annurev.neuro.27.070203.144230