Test expectancy and memory for important information

Abstract

Prior research suggests that learners study and remember information differently depending upon the type of test they expect to later receive. The current experiments investigate how testing expectations impact the study of and memory for valuable information. Participants studied lists of words ranging in value from 1 to 10 points with the goal being to maximize their score on a later memory test. Half of the participants were told to expect a recognition test after each list, whereas the other half were told to expect a recall test. After several lists of receiving tests congruent with expectations, participants studying for a recognition test instead received an unexpected recall test. In Experiment 1, participants who had studied for a recognition test recalled less of the valuable information than participants anticipating the recall format. These participants continued to attend less to item value on future (expected) recall tests than participants who had only ever experienced recall testing. When the recognition tests were made more demanding in Experiment 2, value-based recall improved relative to Experiment 1: though memory for the valuable information remained superior when participants studied with the expectation of having to recall the information, there were no longer significant differences after accounting for recall testing experience. Thus, recall-based testing encouraged strategic, value-based encoding and enhanced retrieval of important information, whereas recognition testing in some cases limited value-based study and memory. These results extend prior work concerning the impact of testing expectations on memory, offering further insight into how people study important information. (PsycINFO Database Record

Figure 1

Recall probability in Experiment 1 as a function of condition and item value in (a) List 5, (b) Lists 6–8, and (c) the first three expected recall tests, averaged across lists. Error bars represent standard error.

Figure 1

Recall probability in Experiment 1 as a function of condition and item value in (a) List 5, (b) Lists 6–8, and (c) the first three expected recall tests, averaged across lists. Error bars represent standard error.

Figure 2

Recall probability in Experiment 2 as a function of condition and item value in (a) List 5, (b) Lists 6–8, and (c) the first three expected recall tests, averaged across lists. Error bars represent standard error.

Figure 2

Recall probability in Experiment 2 as a function of condition and item value in (a) List 5, (b) Lists 6–8, and (c) the first three expected recall tests, averaged across lists. Error bars represent standard error.