Learning and Awareness in the Hebb Digits Task

Abstract

Subjects (N = 163) were tested on the Hebb digits task, for which they listened to 24 lists of nine digits read at the rate of about one digit per second and, following each list, attempted to recall the digits in the correct order. They were not informed, however, that the series on Trial 3 recurred on every subsequent third trial. On the basis of a postexperimental questionnaire, subjects were categorized as unaware (n = 21), late aware (n = 29), middle aware (n = 31), and early aware (n = 29) with respect to the point where they reported noticing list repetition. For both aware and unaware subjects, the Hebb repetition effect was replicated in that the recall scores on the repeating series improved over trials relative to those on the novel ones. Moreover, the rate of learning of the critical series was similar in all four groups. These results are inconsistent with Bower and Winzenz's (1969) reallocation hypothesis, in which it is postulated that learning is mediated by recognition of identity, and support Hebb's contention that repetition of a set of digits can set up some form of direct storage in long-term memory.