An important aspect of the enhanced pre- and postnatal developmental (ePPND) toxicity study in nonhuman primates (NHP) is that it combines in utero and postnatal assessments in a single study. However, it is unclear if NHP ePPND studies are suitable to perform all of the evaluations incorporated into rodent PPND studies. To understand the value of including cognitive assessment in a NHP ePPND toxicity study, we performed a power analysis of object discrimination reversal task data using a modified Wisconsin General Testing Apparatus (ODR-WGTA) from two NHP ePPND studies. ODR-WGTA endpoints evaluated were days to learning and to first reversal, and number of reversals. With α = 0.05 and a one-sided t-test, a sample of seven provided 80% power to predict a 100% increase in all three of the ODR-WGTA endpoints; a sample of 25 provided 80% power to predict a 50% increase. Similar power analyses were performed with data from the Cincinnati Water Maze (CWM) and passive avoidance tests from three rat PPND toxicity studies. Groups of 5 and 15 in the CWM and passive avoidance test, respectively, provided 80% power to detect a 100% change. While the power of the CWM is not far superior to the NHP ODR-WGTA, a clear advantage is the routine use of larger sample size, with a group of 20 rats the CWM provides ~90% power to detect a 50% change. Due to the limitations on the number of animals, the ODR-WGTA may not be suitable for assessing cognitive impairment in NHP ePPND studies.