Reaction time measures of speed of processing: Speed of response selection increases with age but speed of stimulus categorization does not

We report an experiment that investigates the relationship between speed of information processing and development. The goal of the experiment was to compare developmental changes at two stages of processing--response selection and stimulus categorization. The experiment compared developmental change on three kinds of reaction time task. The first was a standard Jensen (1982) procedure which separates the decision and motor components of reaction time, the second was a modification of this procedure suggested by Smith & Carew (1987) to control for possible anticipatory strategies, and the third was a more traditional reaction time task where each stimulus is paired with a unique response (lights-to-keys). Age changes in speed of information processing were found for the lights-to-keys task but not for either of the Jensen tasks. We argue that this is because the lights-to-keys task taps response selection factors that change with age and that are orthogonal to differences in speed of stimulus categorization indexed by the standard Jensen task. If the latter is taken as the purer index of speed of processing then we conclude that speed of processing does not change with age.