Fig. 2

Subjective ratings of action observation and motor imagery. Participants were asked to provide a confidence rating for how accurately they could perform the dance if they were required to for the AO task, and to rate how clearly and vividly they imagined the instructed behaviour for the KMI and VMIQ-2 tasks on a scale from 1 (Perfectly accurate or clear) to 5 (Not at all, or no image at all). Non-dancers provided significantly higher ratings (i.e., poorer perceived ability) during the AO task relative to ballet dancers (p = 0.004) and non-ballet dancers (p = 0.044). All groups provided significantly lower ratings, indicative of clearer imagery, when imagining non-expressive movements during the VMIQ-2 task relative to KMI of the newly-learned dance (ballet: p = 0.005, non-ballet and non-dance: p = 0.000). Error bars represent SEM