Investigating premotor reaching biases after prism adaptation

ABSTRACT

Prism adaptation (PA) is both a visuomotor learning task and potential treatment for spatial neglect after stroke. While PA’s aftereffects can improve neglect symptoms, therapeutic benefits vary across individuals, possibly due to differences in neglect subtypes. Neglect symptoms can be described along an information processing pathway, yielding perceptual (input) and premotor (output) neglect subtypes. There is some evidence that PA mainly benefits persons with premotor neglect. We investigated whether PA modulates the premotor stage of information processing by examining whether PA could induce a premotor bias in healthy adults. We measured perceptual and premotor biases using a speeded reach task that compares the initiation time of leftward and rightward reaches to lateralized targets from different hand start positions. Using a randomized mixed experimental design, 30 right-handed healthy adults completed this speeded reach task before and after either left-shifting (n = 15) or right-shifting (n = 15) PA. As hypothesized, left-shifting PA speeded initiation time specifically for reaches in the rightward direction, regardless of target location (p = .02, η_p²= .18), suggesting that PA induced a premotor bias in the direction of the prism aftereffect. These findings have implications for PA’s underlying mechanisms, which can inform visuomotor learning theories and PA’s use as a treatment for spatial neglect.

KEYWORDS:

• Spatial neglect
• Prism adaptation
• Perceptual
• Premotor
• Subtypes

Acknowledgements

This work was supported by a Student Research Award awarded to JRA from the Canadian Psychological Association Neuropsychology Section. JRA was supported by scholarships from the Killam Trust, Dalhousie University, and the Canadian Institutes of Health Research (Vascular Cognitive Impairment Training Platform, VAST). We would like to thank: Dr. Christopher Striemer, for providing a reference copy of the experimental program used in Striemer and Borza (2017); Dr. Heather Neyedli, for supplying the lab space and PLATO goggles; Dr. Tracy Taylor-Helmick, for providing access to the experimental software; Samantha Good and Dr. Sarah Kraeutner, for their preliminary work on speeded reach task data analysis pipelines; Dr. Hisham Abboud, for his kind assistance troubleshooting the experimental program; Christopher Wright (Dalhousie psychology shop technician), for building the USB-to-serial port converter and start key for the speeded reach task and troubleshooting assistance; Ashton Sheaves, for assistance with documentation of the experimental setup; and Drs. Raymond Klein and David Westwood for serving on JRA’s thesis committee and providing advice and guidance on the project conceptualization and interpretation.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

All anonymized data and R code required to replicate the analyses in this paper are available on the Open Science Framework (private read-only link for peer review: https://osf.io/q3wcp/?view_only=ed6382ba6cc2476a90692272f9381cd4).

Notes

1 The EHI-SF is a four-item version of the original EHI, which measures handedness by asking how often the respondent uses their left or right hand for various manual activities (e.g., writing, throwing). The EHI-SF has very good internal consistency (α = .93) and a strong correlation with scores on the widely used original 10-item scale (r² = .94; Oldfield, 1971; Veale, 2014). Handedness laterality quotients were calculated according to Veale (2014), with a positive quotient indicating a right-hand preference. Mean laterality quotient for overall sample was 83.75 (SD = 21.81).

2 Although Figure 2 displays perceptual and premotor biases separately, we do not intend this to suggest that these biases are mutually exclusive; rather, a PwN could display both biases to varying degrees.

3 0.2% of total PSA data were removed due to behavioural artifacts.

4 We explored whether laterality quotient on the EHI-SF correlated with the shift in premotor bias post-LPA, and the correlation was not statistically significant (p > .5). We noted that the restricted range of EHI-SF scores in our sample (i.e., 16/30 participants with laterality quotient of +100) may make it more difficult to detect significant correlations using this variable. We plan to also run this correlation in left-handers, who may have a greater range of laterality quotients than right-handers.

Additional information

Funding

This work was supported by a Student Research Award awarded to JRA from the Canadian Psychological Association Neuropsychology Section.