Neural and Behavioral Outcomes Differ Following Equivalent Bouts of Motor Imagery or Physical Practice

Abstract

Despite its reported effectiveness for the acquisition of motor skills, we know little about how motor imagery (MI)-based brain activation and performance evolves when MI (the imagined performance of a motor task) is used to learn a complex motor skill compared to physical practice (PP). The current study examined changes in MI-related brain activity and performance driven by an equivalent bout of MI- or PP-based training. Participants engaged in 5 days of either MI or PP of a dart-throwing task. Brain activity (via fMRI) and performance-related outcomes were obtained using a pre/post/retention design. Relative to PP, MI-based training did not drive robust changes in brain activation and was inferior for realizing improvements in performance: Greater activation in regions critical to refining the motor program was observed in the PP versus MI group posttraining, and relative to those driven via PP, MI led only to marginal improvements in performance. Findings indicate that the modality of practice (i.e., MI vs. PP) used to learn a complex motor skill manifests as differences in both resultant patterns of brain activity and performance. Ultimately, by directly comparing brain activity and behavioral outcomes after equivalent training through MI versus PP, this work provides unique knowledge regarding the neural mechanisms underlying learning through MI.

Leveraging the effector independent nature of motor imagery when it is paired with physical practice

“…Changes related to RE and the ACF1 reflect the onset of effector dependent encoding, whereby the changes observed from trial-totrial represent the well-established error detection/correction mechanism that facilitates the ideal kinematics of www.nature.com/scientificreports/ the end-effector and allow for the task goal to be achieved (i.e., the final position of the dart) 50 . Further, greater activation localized to the SMA, cerebellum, and anterior cingulate cortex observed following an equivalent bout of motor imagery-versus physical practice-based training (i.e., at the mid-training scan 40 ) supports the notion that physical practice is required to facilitate effector dependent encoding. Specifically, these regions are implicated in both the 'kinematics-to-dynamics' transformation (i.e., reflecting the dynamics of the upcoming movement and commands sent to the effectors) 22,51,52 , and error detection/correction mechanisms that rely on sensory reafference to disengage in the incorrect response 53,54 and update the motor program [55][56][57][58][59] .…”

“…Improvements in performance are noted by negative effect sizes reflecting a decrease in error (mean RE and BVE) and global kinematic variability, and positive effect sizes reflecting an increase in trial-by-trial correction factor (ACF1) and angular velocity. Data for test sessions on day 1 and 5 is previously reported in Kraeutner et al 40 ). for both RE and BVE were obtained between day 10 vs 6 indicating a worsening in performance in the second half of training (i.e., through motor imagery) for this group.…”

“…Interestingly, a system comparing visual feedback to simulated visual consequences is critical to planning motor actions (during physical practice) prior to effector selection (i.e., mapping motor representations to an effector), separate from that which generates the efference copy and intended consequences of the movement 66 . In showing that activation in regions previously thought to reflect the reliance on a visual image during motor imagery of a task 67 negatively correlated with improvements in performance (i.e., bilateral parietal cortices and lingual gyri at the mid-training scan; and occipital regions as well as left superior and bilateral middle frontal gyri at the posttraining) 40 , our findings may suggest that previously theorized error detection/correction mechanisms active during motor imagery 18,68 only supports refinement of effector independent representations, relying on these visual comparisons. This argument is further substantiated by our finding of increased activation localized to occipital-temporal areas observed following training regardless of the order that motor imagery was applied, in line with work examining activation during (imagery-based) pre-shot routines in expert versus novice archers 69 and golfers 70 .…”

“…Within-group comparisons (pre vs. mid; mid vs. post) indicated that training-related changes in brain activation that achieved statistical significance were only driven via physical practice, and only during the initial stage of training. This analysis is included as Supplementary Material (see Supplementary Table 4)-for an in-depth discussion of the impact of training-modality on resultant patterns of motor imagery-based brain activity, see 40 . As all groups demonstrated similar improvements in performance over the course of the study, we conducted a comparison of brain activation from pre-to post-training (i.e., post-> pre-training scan) across all subjects, revealing greater activation localized to regions including bilateral precuneus, fusiform gyri, middle temporal gyri, and occipital regions at the post-training scan (post > pre; Fig.…”

The Effects of Computer-Based and Motor-Imagery Training on Scoring Ability in Lacrosse

“…Motor-related activations in the brain were more strongly induced by experiencing kinesthetic imagery than visual imagery ( Neuper et al, 2005 ; Guillot et al, 2009 ). Therefore, motor-imagery training could be an effective tool to improve athletic performance; however, actual training is generally considered to be a more effective means of developing physical ability ( Driskell et al, 1994 ; Gentili et al, 2010 ; Kraeutner et al, 2020 ). In the current study, we aimed to develop an effective and convenient training by combining the elements that were proven to be important for the imagery.…”