Table 1.
Participant main characteristics and performance details (n = 11).
Fig 1.
Flowchart for the randomized crossover trial.
Eleven participants were assessed for eligibility, enrolled, and randomized to complete three experimental conditions (CHO-MR, MUS, and PLA) in a counterbalanced crossover design. All participants completed all conditions, and all were included in the final analysis. CHO-MR, carbohydrate mouth rinse; MUS, music listening; PLA, placebo mouth rinse.
Fig 2.
Schematic overview of the experimental protocol.
Resting-state fNIRS recordings and Stroop tasks (3 min each) were performed to assess DLPFC oxygenation before intervention, immediately after intervention (CHO-MR, MUS, or PLA), and following completion of the 4-km cycling time trial. Blood lactate concentration, heart rate, and perceived exertion were recorded at predefined intervals. Abbreviations: CHO-MR, carbohydrate mouth rinsing; DLPFC, dorsolateral prefrontal cortex; fNIRS, functional near-infrared spectroscopy; HR, heart rate; MUS, music listening; PLA, placebo mouth rinse; [La⁻], blood lactate concentration; RPE, rating of perceived exertion; TT, time trial.
Fig 3.
Region-of-interest (ROI) configuration and fNIRS processing workflow.
(A) Channel layout illustrating bilateral DLPFC regions analyzed in the primary outcome. The right DLPFC comprised channels 1, 2, 3, 5, 6, 11, 17, and 18, while the left DLPFC included channels 19, 20, 33, 34, 35, 38, 39, and 43. (B) Preprocessing workflow. Only channels with 30-mm source–detector separation were included. Signals were band-pass filtered (0.01–0.10 Hz), low-quality channels (signal-to-noise ratio < 30 dB) were excluded and accHbO₂ concentration changes were computed using the modified Beer–Lambert law. accHbO₂, accumulated oxygenated hemoglobin; DLPFC, dorsolateral prefrontal cortex; FPC, frontopolar cortex; fNIRS, functional near-infrared spectroscopy; HbO, oxygenated hemoglobin; HbR, deoxygenated hemoglobin; OFC, orbitofrontal cortex; ROI, region of interest; VLPFC, ventrolateral prefrontal cortex.
Fig 4.
Topographic 2D maps of prefrontal cortex oxygenated hemoglobin concentration changes (ΔHbO) across experimental conditions and time points.
Fig 5.
Changes in oxygenated hemoglobin (HbO) within the DLPFC across experimental conditions.
(A) Right DLPFC and (B) left HbO concentration changes across rest, baseline (task), post-intervention, and post-TT under CHO-MR, MUS, and PLA conditions. Global model results are shown for stage and condition × stage interaction. Values are presented as mean ± SD. Between-condition comparisons within stages were conducted using Holm-adjusted EMM. Symbols denote significant differences between CHO-MR and PLA (* p < 0.05, ** p < 0.01, *** p < 0.001) and between CHO-MR and MUS (p < 0.05, ## p < 0.01, ### p < 0.001). HbO, oxygenated hemoglobin; CHO-MR, carbohydrate mouth rinsing; DLPFC, dorsolateral prefrontal cortex; MUS, music listening; PLA, placebo mouth rinse; TT, 4-km cycling time trial.
Fig 6.
Stroop performance and associations with DLPFC oxygenation.
(A) Stroop performance across baseline, intervention, and post-TT under PLA, CHO-MR, and MUS conditions. (B–D) GEE-based associations using pooled change scores (Δ = post-TT − baseline; n = 33): (B) right versus left DLPFC ΔaccHbO₂, (C) ΔStroop versus right DLPFC ΔaccHbO₂, and (D) ΔStroop versus left DLPFC ΔaccHbO₂. Values are mean ± SD. In panel (A), symbols indicate Holm-adjusted pairwise differences at post-TT (* p < 0.05, ** p < 0.01, *** p < 0.001 for CHO-MR vs. PLA; # p < 0.05, ## p < 0.01, ### p < 0.001 for CHO-MR vs. MUS). ΔaccHbO₂, accumulated oxygenated hemoglobin change; CHO-MR, carbohydrate mouth rinsing; DLPFC, dorsolateral prefrontal cortex; GEE, generalized estimating equations; HbO, oxygenated hemoglobin; MUS, music listening; PLA, placebo mouth rinse; TT, time trial; Δ, change from baseline to post-TT.
Fig 7.
Effects of CHO-MR, MUS, and PLA on performance outcomes during the 4-km cycling time trial.
(A) Completion time (s). (B) Peak power (W). (C) Mean power output (W). (D) Mean speed (km/h). Analyses were conducted using GEE adjusted for period, sequence, and first-order carryover. Symbols denote Holm-adjusted pairwise differences (* p < 0.05, ** p < 0.01, *** p < 0.001 for CHO-MR vs. PLA; # p < 0.05, ## p < 0.01, ### p < 0.001 for CHO-MR vs. MUS). Values are presented as mean ± SD with individual data points overlaid. Abbreviations: CHO-MR, carbohydrate mouth rinsing; GEE, generalized estimating equations; MUS, music listening; PLA, placebo mouth rinse.
Fig 8.
Ratings of perceived exertion, blood lactate concentration, and heart rate during the 4-km cycling time trial under CHO-MR, MUS, and PLA conditions.
(A) RPE at 500-m intervals. Significant main effects of condition (p < 0.01) and distance (p < 0.001) and a condition × distance interaction (p < 0.001) were observed. (B) Blood lactate concentration at rest/pre-exercise and post-exercise. A significant main effect of stage was observed (p < 0.001) without between-condition differences (p ≥ 0.05). (C) HR at 500-m intervals. A significant main effect of distance (p < 0.001) and condition × distance interaction (p < 0.001) were observed, with no between-condition differences at individual distances (p ≥ 0.05). Values are presented as mean ± SD. In panel (A), symbols denote Holm-adjusted pairwise differences between CHO-MR and PLA (* p < 0.05, ** p < 0.01, *** p < 0.001) and between CHO-MR and MUS (# p < 0.05, ## p < 0.01, ### p < 0.001). CHO-MR, carbohydrate mouth rinsing; HR, heart rate; MUS, music listening; PLA, placebo mouth rinse; RPE, rating of perceived exertion.