Optimizing multiple-choice tests as tools for learning

Mem Cognit. 2015 Jan;43(1):14-26. doi: 10.3758/s13421-014-0452-8.

Abstract

Answering multiple-choice questions with competitive alternatives can enhance performance on a later test, not only on questions about the information previously tested, but also on questions about related information not previously tested-in particular, on questions about information pertaining to the previously incorrect alternatives. In the present research, we assessed a possible explanation for this pattern: When multiple-choice questions contain competitive incorrect alternatives, test-takers are led to retrieve previously studied information pertaining to all of the alternatives in order to discriminate among them and select an answer, with such processing strengthening later access to information associated with both the correct and incorrect alternatives. Supporting this hypothesis, we found enhanced performance on a later cued-recall test for previously nontested questions when their answers had previously appeared as competitive incorrect alternatives in the initial multiple-choice test, but not when they had previously appeared as noncompetitive alternatives. Importantly, however, competitive alternatives were not more likely than noncompetitive alternatives to be intruded as incorrect responses, indicating that a general increased accessibility for previously presented incorrect alternatives could not be the explanation for these results. The present findings, replicated across two experiments (one in which corrective feedback was provided during the initial multiple-choice testing, and one in which it was not), thus strongly suggest that competitive multiple-choice questions can trigger beneficial retrieval processes for both tested and related information, and the results have implications for the effective use of multiple-choice tests as tools for learning.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Choice Behavior / physiology*
  • Cues
  • Educational Measurement / standards*
  • Female
  • Humans
  • Learning / physiology*
  • Male
  • Mental Recall / physiology*
  • Neuropsychological Tests / standards*
  • Young Adult