Cognitive Processing Efficiency (Throughput) Improves with Aerobic Exercise and Is Independent of the Environmental Oxygenation Level: A Randomized Crossover Trial

Aerobic exercise with eicosapentaenoic acid (EPA) may enhance cognition via cerebrovascular pathways. We tested whether mild hyperbaric oxygen (HBO; 1.41 atmospheres absolute [ATA], approximately 30% O₂) adds to gains in cognitive processing capacity (throughput) versus normobaric normoxia (1.0 ATA, approximately 21% [20.9%] O₂). Healthy young adults (<i>n</i> = 16) performed cycling exercise at 60-70% VO_2peak for 60 min, twice weekly, for 4 weeks per environment with a 1-week washout; EPA (2170 mg·day^-1) was taken during each 4-week training phase (total 8 weeks) and was paused during the washout. An EPA-only control (<i>n</i> = 8) was included for supplementary analysis. The primary outcome was throughput (correct·min^-1; T1-T4); secondary outcomes were interference indices (I1: stroop interference, I2: reverse-stroop interference). Effects were estimated using linear mixed models [environment, time, environment × time; AR(1), REML] and Hedges' <i>g</i>_av; accuracy used generalized estimating equations. Throughput improved mainly with time (T1-T2 <i>p</i> < 0.001; T4 <i>p</i> = 0.017; T3 <i>p</i> = 0.055), with no environment or interaction effects. I1/I2 showed no significant change, and one task exhibited an accuracy ceiling. Under safe, feasible conditions (≤1.41 ATA), aerobic exercise improved processing capacity (throughput) independently of environmental oxygenation level. The absence of detectable additive effects should be interpreted cautiously under conservative settings.