Supplementary Material for: The Number of Items on Each Stroop Test Card Is Unrelated to Its Sensitivity

Objects: Different variations of the Stroop Color-Word Test (SCWT) exist and the tests differ in the number of items on each test card. The optimal number of items per card of the SCWT is unclear. Methods: A comprehensive neuropsychological battery including the SCWT was administered to 237 cognitively normal controls (NC), 221 patients with mild cognitive impairment (MCI), and 160 patients with Alzheimer’s disease (AD). We compared the accuracy for different items of card C (SCWT-C10accuracy, SCWT-C24accuracy, and SCWT-C50accuracy) in distinguishing patients with MCI and AD from NC. Results: The correct responses for different items of card C were ranked in the order: NC > MCI > AD group and mild AD > moderate AD, except for SCWT-C24accuracy. The validity of SCWT-C24accuracy was similar to that of SCWT-C50accuracy for discriminating MCI and AD from NC, as well as different levels of AD. Conclusion: We concluded that the number of items on the accuracy of the response was unrelated to its sensitivity to deficits and we preferred the 24-item version in daily clinical use.