Edited by:
Reviewed by:
*Correspondence:
This article was submitted to Educational Psychology, a section of the journal Frontiers in Psychology
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Based on numerous studies showing that testing studied material can improve long-term retention more than restudying the same material, it is often suggested that the number of tests in education should be increased to enhance knowledge acquisition. However, testing in real-life educational settings often entails a high degree of extrinsic motivation of learners due to the common practice of placing important consequences on the outcome of a test. Such an effect on the motivation of learners may undermine the beneficial effects of testing on long-term memory because it has been shown that extrinsic motivation can reduce the quality of learning. To examine this issue, participants learned foreign language vocabulary words, followed by an immediate test in which one-third of the words were tested and one-third restudied. To manipulate extrinsic motivation during immediate testing, participants received either monetary reward contingent on test performance or no reward. After 1 week, memory for all words was tested. In the immediate test, reward reduced correct recall and increased commission errors, indicating that reward reduced the number of items that can benefit from successful retrieval. The results in the delayed test revealed that reward additionally reduced the gain received from successful retrieval because memory for initially successfully retrieved words was lower in the reward condition. However, testing was still more effective than restudying under reward conditions because reward undermined long-term memory for concurrently restudied material as well. These findings indicate that providing performance–contingent reward in a test can undermine long-term knowledge acquisition.
A central question of both experimental research and educational practice is how learning and retention can be promoted. A very powerful technique to improve long-term memory seems to be retrieving previously learned materials while taking a test (e.g.,
However, in real-life educational settings, test-taking may have additional effects on the emotions and motivations of learners, factors that have been largely neglected in previous research on the effect of testing. This neglect is particularly interesting because there is reason to assume that such effects may undermine the effectiveness of testing in enhancing long-term memory. For instance, regarding emotions, if a test induces a high degree of performance-related anxiety, the reduction in cognitive resources due to distraction by task-irrelevant emotion-induced thoughts (e.g.,
At the motivational level, a typical effect of testing in real-life educational settings is that the motivation of learners is shifted toward an extrinsically motivated state due to the common practice of placing important consequences on the outcome of a test (for a review, see
Basically, there are two possibilities why a test that induces a high degree of extrinsic motivation may undermine the effectiveness of testing in enhancing long-term knowledge acquisition. First, by providing gains contingent on performance, such a test induces a strong desire to perform as well as possible. Such a desire may impair the quality of retrieval of actually stored knowledge. On the one hand, the rate of successfully retrieved information may be decreased because it has been shown that people often perform below actual abilities when trying to perform as well as possible, an observation that is commonly attributed to the experience of performance pressure. Such performance pressure often leads to the occupation of attention by task-irrelevant thoughts, such as ruminations about one’s performance and its consequences (e.g.,
Second, a test that induces extrinsic motivation may even reduce the benefit received from successful retrieval. Most theoretical accounts proposed to explain the high effectiveness of testing assume that retrieval of information from memory represents a new learning event (i.e., reconsolidation; e.g.,
The aim of the present study was to examine the effect of extrinsic motivation on the long-term memory effects of testing. To examine the issue, we employed a standard testing-effect paradigm and manipulated the degree of extrinsic motivation during immediate testing. Participants first studied Swahili–German vocabulary pairs (e.g., Mashua–Boat) without mentioning that they may be rewarded for their later test performance. In a subsequent immediate memory test, one-third of the vocabulary pairs were tested, one-third were presented for restudy, and the remaining third did not appear in the test and served as control pairs. In order to manipulate the degree of extrinsic motivation during immediate testing, participants received either performance-contingent monetary reward for test performance (high extrinsic-motivation condition), or not (low extrinsic-motivation condition; e.g.,
In the immediate test, we expected to replicate the detrimental effects of providing extrinsic reward contingent on test performance on the quality of retrieval of learned knowledge (e.g.,
Sixty undergraduate students (49 females;
The study list consisted of 30 Swahili–German vocabulary pairs drawn from
Half of the participants were randomly assigned to a low extrinsic-motivation condition, and the other half to a high extrinsic-motivation condition. In each condition, the experiment consisted of three main phases: a study phase, a (combined) test/restudy phase, and a 1-week delayed final test phase. In the study phase, participants were presented 30 vocabulary pairs (e.g.,
Memory performance in the immediate test as a function of motivational condition is shown in
A 2 × 3 ANOVA with the within-participants factor of vocabulary type (tested, restudied, and control) and the between-participants factor of extrinsic motivation (high and low) on the probability of commission errors (intra-list intrusions) revealed neither a main effect of vocabulary type nor a main effect of extrinsic motivation,
Finally, we examined the effect of reward on memory for vocabulary pairs which were initially successfully retrieved. To control for potential item-selection artifacts (i.e., artifacts due to unbalanced distribution of vocabulary pairs across conditions because of differential recall in the immediate memory test), we determined for each vocabulary pair the conditional probability of correct recall in the delayed test given successful recall in the immediate test, collapsing data across participants. As shown in
Previous research has shown that retrieving previously learned contents in a test can improve long-term memory for tested contents, suggesting that the number of tests in education should be increased to enhance knowledge acquisition (see
More detailed analyses showed that the detrimental effects of reward were attributable to two factors. First, the provision of monetary reward contingent on performance reduced correct recall and increased commission errors in the immediate test, a pattern that typically occurs in situations where people try to perform as well as possible to maximize promised extrinsic gains (e.g.,
Second, even for vocabulary pairs that were initially successfully retrieved, long-term memory was reduced when reward was initially provided. Thus, extrinsic motivation seems to undermine even the benefit gained from successfully retrieving stored information in a test. Such an effect is consistent with findings showing that the quality of learning differs depending on motivational state. Compared to intrinsic motivation, extrinsically motivated learners show a less elaborative learning style characterized by more passive and less effortful learning that vanishes beyond the point of being rewarded or punished (e.g.,
The present results further show that high extrinsic motivation can even have detrimental effects on long-term memory for material that is restudied. Compared to the no-reward condition, memory in the delayed test for vocabulary pairs that were initially restudied was reduced in the reward condition as well. As participants were forced to write down each of the to-be-restudied vocabulary pairs during restudy, such a finding cannot easily be explained by the simple assumption that rewarding participants only for some vocabulary pairs led them to abandon processing of not rewarded vocabulary pairs. However, in such a situation, the induced extrinsic motivation seems to bring about a less effortful restudying of not rewarded information.
In the present study, we examined the effect of providing performance-contingent reward in an immediate memory test on performance in a delayed long-term memory test where no reward was provided. This situation mimics the typical educational scenario in which the objective is to provide learners with knowledge to prepare them for later life and work, where knowledge retrieval is not necessarily driven by extrinsic forces. Doing so, we found that providing extrinsic reward for test performance can undermine long-term knowledge acquisition of the assessed contents. The situation may be different however, when extrinsic motivation is increased during immediate test-taking because learners are aware that they are preparing for a delayed test for which they will be rewarded based on their performance. In such a situation, additional motivational factors may play an important role during immediate test taking, such as the motivation to learn the material as well as possible for the delayed test (see
The present study also raises several questions that should be addressed in future research. First, our sample consisted mainly of female undergraduate students. Thus, future research should examine whether the results of the present study generalize across gender and different levels of education. Second, in order to be able to relate our results to prior findings, the study material consisted of foreign language word pairs that have been frequently used in research on the effects of testing (e.g.,
Finally, the present results may have important implications for applied settings, such as educational practice. Based on the finding that test taking can enhance later memory, it has been argued that increasing the number of tests in education is a promising technique to boost educational achievement (e.g.,
CK developed the study concept. All authors contributed to the study design and all authors analyzed and interpreted the data. CK prepared the draft manuscript, and AA, KE, and KM provided critical revisions. All authors approved the final version of the manuscript for submission.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The reviewer Michael S. Dempsey and handling Editor declared their shared affiliation, and the handling Editor states that the process nevertheless met the standards of a fair and objective review.