Differcntial feature analysis in the recognition memory of reflective and impulsive children.

Seventy-two children. 18 reflective, and 18 impulsive first and fifth graders were tested in a forced-choice recognition memory task. Experimental conditions systematically varied the possibility that correct responses could be made on the basis of verbal labels, visual feature analyses, or both. Although reflective children made more correct recognition choices than did impulsive children under all experimental conditions, this difference was, as predicted, significant only in a condition in which the sole basis for correct response depended on a purely visual feature analysis. Order of condition difficulty indicated that visual feature analysis independent of verbal processes was responsible for successful recognition performance in these Ss. Results were discussed with reference to the feature-testing model for recognition proposed by Selfridge and Neisser.