Suppose that a student conducts a study in which the manipulated independent variable is level of white noise (60 decibels versus 65 decibels). Ten participants are assigned to each level of noise; these participants vary widely in age, hearing acuity, and study habits. The outcome variable is performance on a verbal learning task (how many words on a list of 25 words does each participant remember). The t test obtained is not statistically significant. What advice would you give to this student about ways to redesign this study that might improve the chances of detecting an effect of noise on verbal learning recall?