The critical power (CP) concept has been used to determine the appropriate rest interval during intermittent exercise through the investigation of critical rest interval (CRI). Repeated sprint ability (RSA) testing has been developed to define the athlete’s ability to recover and maintain maximal effort during successive bouts. The CP model has been used to understand the physiological responses involved with intermittent exercise delineating between severe and heavy exercise intensity domains. The primary purpose of this study was to determine the CRI from the work-time relationship given by RSA testing using varying work to rest ratios. The secondary purpose was to determine the validity of CRI by evaluation of physiological responses above and below estimated rest interval values during intermittent cycling. Twelve recreationally trained males (mean ± SD; age 24.1 ± 3.6yr; height 175.8 ± 7.0cm; weight 77.6 ± 12.8kg; V̇ O2peak 43.3 ± 5.6ml/kg/min; Body Fat (%) 24.5 ± 4.4) were recruited for this study. Participants performed a graded exercise test to determine V̇ O2peak and peak heart rate. Eight participants completed the same three RSA protocols with 6s maximal sprints and varying rest intervals (12-24s) on a cycle ergometer. Intermittent critical power (ICP) was calculated through the linear total work (TW) and time-to-exhaustion (TTE) relationship, whereas CRI was estimated using the average work per sprint and ICP. Seven subjects completed trials above and below estimated CRI to evaluate the validity of this estimate through the examination of the physiological responses. Breath-by-breath oxygen consumption (V̇ O2) and heart rate (HR) values were recorded during the validation trials. One-way repeated measures analysis of variance (ANOVA) was used to analyze the variables from the RSA trials. Paired samples t-tests were performed to compare performance and physiological variables above or below CRI during the validation trials. Two-way repeated measures ANOVA was used to examined the changes in oxygen consumption (V̇ O2), HR, mean power (MP), and TW throughout the validation trials. Significant differences (p < 0.1) were found for the number of intervals completed, TTE, average work per sprint, peak and mean V̇ O2 between RSA protocols. Linearity between TW and TTE was r2 = 0.952 ± 0.081. During the validation trials, TTE was significantly greater in the above versus the below CRI trial (2270.43 ± 941.15s vs. 1511.00 ± 811.0s). Furthermore, blood lactate concentration (8.94 ± 4.89mmol/L vs. 6.56 ± 3.45mmol/L), AveV̇ O2 (2.05 ± 0.36L/min vs. 1.78 ± 0.26L/min), V̇ O2peak (2.84 ± 0.48L/min vs. 2.61 ± 0.43L/min), and AveHR (151.14 ± 18.46bpm vs. 138.14 ± 17.51L/min) were significantly greater in the below CRI trial when compared to the above CRI trial. Significant interactions were found between above and below trials within minimal V̇ O2 response (F = 6.886, p = 0.024, η2 = 0.534) to the recovery intervals and maximal HR (F = 4.51, p = 0.016, η2 = 0.429) response to the work intervals. During the above CRI trial, minimal V̇ O2 response decreased over time (51-43%V̇ O2 peak) while maximal HR response achieved a steady state level (81-84%HRpeak). Conversely, minimal V̇ O2 response during the below CRI trial achieved a steady state level (54-58%V̇ O2 peak), whereas maximal HR response increased over time (84-90%HRpeak). The relationship between TW and TTE is appropriate for use with RSA testing with varying rest intervals. The differing physiological response during the validation trials may reflect changes in energy system contribution. In conclusion, CRI distinguished between physiological responses related to exercise intensity domains in a manner similar to CP estimates determined from other testing and exercise modalities.