The result: measurable gains in inhibitory control and working memory, in some cases exceeding what actually performing the exercise produced. For anyone whose training load competes with coursework — or whose cognitive bandwidth is squeezed by fatigue and stress — the finding offers a low-intensity lever with documented effect sizes.
The protocol
Participants were randomly assigned to one of three groups: a squat imagery condition with 30 minutes of seated motor rehearsal (n = 36), a squat execution condition consisting of five sets of one-minute squats separated by 10-second rests (n = 32), or a 30-minute seated rest control (n = 30). Cognitive performance was measured before and after each condition using the Stroop, Flanker, Sternberg, and Visual Search tasks — a battery covering inhibitory control, working memory, and visual attention. Two-way repeated-measures ANOVA (group × time) was used, with Benjamini–Hochberg false discovery rate correction applied across 18 primary interaction tests.
Where imagery edged out execution
Both active conditions outperformed rest, but follow-up analyses attributed the gains primarily to the imagery group. Group × time interactions reached significance for Stroop simple and complex response times and Sternberg one- and three-item conditions (all p < 0.01, partial η² = 0.10–0.14). Against physical execution, imagery produced greater reductions in response time on the Stroop simple and Sternberg three-item tasks, with comparable effects on Stroop complex and Sternberg one-item. On Visual Search congruent trials, imagery outperformed both execution and control. One detail worth noting: accuracy on Visual Search incongruent trials dropped after real squats but held stable after imagery and rest. The physical group traded precision for speed; the imagery group did not pay that cost.
A defensible low-load protocol
The takeaway is not "imagery beats exercise." It is that mental rehearsal produced measurable cognitive gains without the metabolic load or the accuracy decrement seen in the physical execution condition. For contexts where fatigue, scheduling, or injury limits physical practice — pre-competition routines, rehab windows, dense academic blocks — motor imagery activates overlapping sensorimotor circuits at near-zero physiological cost. The sample is modest (n = 98) and the population narrow, but effect sizes up to partial η² = 0.14 cross the threshold worth tracking. If you prototype a protocol, the parameters are explicit: 30 minutes seated, structured motor imagery of a specific compound movement, with pre- and post-assessment via Stroop or Sternberg to verify cognitive gain.