The five most used cold pressor test heart rate protocols were compared to the true R-R interval responses recorded during a cold hand test. Forty-one nonsmoking male volunteers between 18 and 55 yr were given a cold hand pressor test. Heart rate was evaluated by averaging the R-R interval within 2-, 5-, 6-, 10-, and 30-s time intervals and by the true R-R interval measurement. No significant physiological differences existed at rest; however, during stress the maximum values obtained were successively diminished the greater the time utilized to average the heart rate pressure (P less than 0.05). During recovery an underestimation of the true response was observed that increased as the averaging time increased (P less than 0.05). Comparison between the R-R interval measure and the 6-s average data indicated a 18% decrease in the time of peak response. Therefore, the responses recorded with the 6-s averaged data were less sensitive and obscured the rapidity of autonomic changes. Hence, when sensitivity and absolute response are a primary concern in determining autonomic function, measurements should include R-R interval measures of heart rate or the average of the number of R-R intervals in 2-s time blocks.