Effects of instructional manipulations and spacing of repetitions in cued-memory tests

Bradford Henry Challis, Purdue University

Abstract

A recent two-factor account of spacing effects (the beneficial effect of spacing repetitions on retention) assumes that a voluntary rehearsal process underlies such effects in cued-memory tests (Greene, 1989). This account predicts that spacing effects will be eliminated in cued tests with incidentally-learned material, if certain conditions are met. The reported experiments examined this claim. In the experiments, processing was controlled with semantic (judging each word for pleasantness or imagery), graphemic (counting the number of letters with certain attributes) or rote rehearsal (repeating each word covertly) orienting tasks. Words (e.g., copper) were studied once or twice at lags of 0, 10 or 30. Explicit memory tests included frequency judgments, and recall to semantic (bronze), graphemic (chopper) or general knowledge (What makes up 10% of yellow gold?) cues. Implicit tests required the completion of word fragments $\rm (c~\underline{\ }~p~p~\underline{\ }~\underline{\ })$ or answering of general knowledge questions. The findings showed that spacing effects occur in cued tests with incidentally-learned material, depending on the processing promoted at study and type of cued test. In the first experiment, the magnitude of spacing effects was similar in intentional and incidental study conditions, if semantic processing was promoted in the incidental condition. Across six cued tests, the magnitude of spacing effects varied considerably. Subsequent experiments involved incidental learning conditions and selected tests. With low level (i.e., graphemic) processing, spacing effects were eliminated (Experiment 2). With semantic processing on the first occurrence and graphemic processing on the second occurrence of repeated items, spacing effects were present. If the level of processing on first and second occurrences was reversed, spacing effects were absent (Experiment 3). With semantic processing on the first and rote rehearsal on second occurrence, or vice versa, spacing effects were obtained in a frequency judgment test, but not in primed word fragment completion (Experiment 4). If processing was uncontrolled, spacing effects were present in a frequency judgment test (Experiments 5A and 5B). The results are problematic for Greene's account, because they contradict the prediction that spacing effects do not occur in cued tests with incidentally-learned material. A general account of spacing effects is offered.

Degree

Ph.D.

Advisors

Roediger, Purdue University.

Subject Area

Psychology|Experiments|Educational psychology

COinS