The association between traffic-related air pollution and adverse cardiovascular effects has been well documented; however, little is known about whether different commuting modes can modify the effects of air pollution on the cardiovascular system in human subjects in urban areas with heavy traffic. We recruited 120 young, healthy subjects in Taipei, Taiwan. Each participant was classified with different commuting modes according to his/her own commuting style. Three repeated measurements of heart rate variability (HRV) indices standard deviation of NN intervals (SDNN) and the square root of the mean of the sum of the squares of differences between adjacent NN intervals (r-MSSD), particulate matter with an aerodynamic diameter </= 2.5 mum (PM2.5), temperature, humidity and noise level were conducted for each subject during 1-h morning commutes (0900-1000 h) in four different commuting modes, including an electrically powered subway, a gas-powered bus, a gasoline-powered car, and walking. Linear mixed-effects models were used to investigate the association of PM2.5 with HRV indices. The results showed that decreases in the HRV indices were associated with increased levels of PM2.5. The personal exposure levels to PM2.5 were the highest in the walking mode. The effects of PM2.5 on cardiovascular endpoints were the lowest in the subway mode compared to the effects in the walking mode. The participants in the car and bus modes had reduced effects on their cardiovascular endpoints compared to the participants in the walking mode. We concluded that traffic-related PM2.5 is associated with autonomic alteration. Commuting modes can modify the effects of PM2.5 on HRV indices among young, healthy subjects.