An examination of orienting task relationships in a proactive interference paradigm

Three experiments investigating release from proactive interference were conducted, in which orienting tasks were employed to bias encoding. Following earlier experiments by Bird (1976, 1977), it was expected that release would be observed when tasks were changed after several trials, but only to the extent that the tasks required different processing. Two obviously related nonsemantic tasks were compared in Experiment 1, and no release was obtained. Experiment 2 was a comparison of part-of-speech classification, considered by some to be a nonsemantic task, and of judgments of word pleasantness. The release obtained was sufficiently low to suggest that part-of-speech decisions involve substantial semantic processing. Finally, Experiment 3 employed four tasks, in order to address various questions about task relationships raised in the earlier experiments. Based on the levels of release observed across experiments and the finding that some tasks led to less proactive interference than others, a tentative categorization of tasks was proposed.