The critical role of retrieval processes in release from proactive interference

Proactive interference (PI) refers to the finding that memory for recently studied (target) information can be vastly impaired by the previous study of other (nontarget) information. PI can be reduced in a number of ways, for instance, by directed forgetting of the prior nontarget information, the testing of the prior nontarget information, or an internal context change before study of the target information. Here we report the results of four experiments, in which we demonstrate that all three forms of release from PI are accompanied by a decrease in participants’ response latencies. Because response latency is a sensitive index of the size of participants’ mental search set, the results suggest that release from PI can reflect more focused memory search, with the previously studied nontarget items being largely eliminated from the search process. Our results thus provide direct evidence for a critical role of retrieval processes in PI release. 2012 Elsevier Inc. All rights reserved. Introduction buildup of PI is caused by a failure to distinguish items Proactive interference (PI) refers to the finding that memory for recently studied information can be vastly impaired by the previous study of further information (e.g., Underwood, 1957). In a typical PI experiment, participants study a (target) list of items and are later tested on it. In the PI condition, participants study further (nontarget) lists that precede encoding of the target information, whereas in the no-PI condition participants engage in an unrelated distractor task. Typically, recall of the target list is worse in the PI condition than the no-PI condition, which reflects the PI finding. PI has been extensively studied in the past century, has proven to be a very robust finding, and has been suggested to be one of the major causes of forgetting in everyday life (e.g., Underwood, 1957; for reviews, see Anderson & Neely, 1996; Crowder, 1976). Over the years, a number of theories have been put forward to account for PI, most of them suggesting a critical role of retrieval processes in this form of forgetting. For instance, temporal discrimination theory suggests that . All rights reserved. ie.uni-regensburg.de from the most recent target list from items that appeared on the earlier nontarget lists. Specifically, the theory assumes that at test participants are unable to restrict their memory search to the target list and instead search the entire set of items that have previously been exposed (Baddeley, 1990; Crowder, 1976; Wixted & Rohrer, 1993). Another retrieval account attributes PI to a generation failure. Here, reduced recall levels of the target items are thought to be due to the impaired ability to access the material’s correct memory representation (Dillon & Thomas, 1975). In contrast to these retrieval explanations of PI, some theories also suggested a role of encoding factors in PI, assuming that the prior study of other lists impairs subsequent encoding of the target list. For instance, attentional resources may deteriorate across item lists and cause the target material to be less well processed in the presence than the absence of the preceding lists (e.g., Crowder, 1976).