Background/AimsMemory and selective attention are important skills for academic and professional performance. Techniques to improve these skills are not taught either in education or company training courses. Any system which can systematically improve these skills will be of value in schools, universities, and workplaces. Aims:To investigate possible improvements in memory and selective attention, as measured by the Digit–Letter Substitution Task (DLST), due to practice of Cyclic Meditation (CM), a yoga relaxation technique, as compared to Supine Rest (SR). Materials and Methods Subjects consisted of 253 school students, 156 boys, 97 girls, in the age range 13–16 years, who were attending a 10-day yoga training course during summer vacation. The selected subjects had English as their medium of instruction in school and they acted as their own controls. They were allocated to two groups, and tested on the DLST, immediately before and after 22.5 minutes practice of CM on one day, and immediately before and after an equal period of SR on the other day. The first group performed CM on day 9 and SR on day 10. For the second group, the order was reversed. Results Within each group pre-post test differences were significant for both the relaxation techniques. The magnitude of net score improvement was greater after SR (7.85%) compared to CM (3.95%). Significance levels were P < 0.4 × 10-9for SR and P < 0.1 × 10-3 for CM. The number of wrong attempts also increased significantly on both interventions, even after removing two outlier data points on day 1 in the SR group. Conclusions: Both CM and SR lead to improvement in performance on the DLST. However, these relaxation techniques lead to more wrong cancellation errors.