SERIAL ACQUISITION AS A FUNCTION OF STAGE OF LEARNING

A 12-unit serial list was learned in each condition of a 3 * 5 factorial design. Ss were l80 undergraduates. After reaching a criterion of 0/12 (1 trial), 4/12, 8/12, 12/12, or 12/12 + 10 trials, S was interrupted and given 1 of 3 tasks in order to determine how many items could be recalled, how many items could be stated correctly according to absolute serial position, and how many adjacent associations could be stated correctly. Data indicate that Ss are able to recall more items than they anticipate correctly on the trial which precedes or follows the interruption. Data also indicate that Ss generally were able to make correct responses on the trial immediately preceding the test without being able to state the item as an associate or as related to absolute serial position. Similarly, when an item was stated correctly as an associate or in reference to its absolute serial position, S was not necessarily able to anticipate the item on the next trial. Data were taken as evidence against both chaining and the absolute positional hypotheses and instead were interpreted in terms of a stage analysis involving item learning and placement of the items along a perceptual dimension. (PsycINFO Database Record (c) 2006 APA, all rights reserved). © 1969 American Psychological Association.