Words: 1761
Running head: PRIMING AND OVERCONFIDENCE Misuse of Priming Effects Overconfidence Abstract This experiment was designed to test whether priming participants through testing can lead to overconfidence on further testing when the information changes. 36 Participants were selected from the ——- to participate in this experiment. The participants were given reading comprehension tests that were amended from Pearson-Longman Education testing materials. 18 participants were given Test 1 followed by a Control Test, and 18 participants were given Test 1 followed by an Experiment Test.
Control Test participants did not exhibit a priming effect nor did Experiment Test participants exhibit overconfidence. It was also shown that statistically Test 1 did not have any correlation between either the Control Test or the Experiment Test. Other research indicated that participants in testing tend to exhibit underconfidence in subsequent testing, and I believe that this is correct; however, further research should be conducted to further understanding of the Priming Effect. Misuse of Priming Effects Overconfidence
Implicit forms of memory (unconscious, unconscious retrieval) have used priming to explain how the recollection of information can be hindered or helped. These tests typically show that participants improved performance on tasks for which they have been subconsciously prepared (Graf and Mandler, 1984). The overconfidence effect is a well-established bias in which someone’s subjective confidence in their judgments is reliably greater than their objective accuracy, especially when confidence is relatively high (Pallier, 2002).
The current research will use priming to affect confidence on subsequent tasks. The experiment of Allwood, Innes-Ker, Homgren and Fredin, (2008) showed the importance of question format when measuring confidence in people. This also is evident in several studies involving the priming effect. Stolz (2005) showed that test subjects increased their reliability to significant levels under conditions that lent themselves to help with target recognition. The priming affect with implicit memory was only evident in specific instances of the word recall test until the test was tailored to help exhibit priming.
Other research such as Jacoby and Dallas (1981) demonstrates that the priming effect is severely reduced by the change in modality from the studying of words to be recalled to the testing. With respect to overconfidence, illusions of competence are thought to occur when judgments of learning (JOLs) made in the presence of intact cue-target pairs during study create a foresight bias or an illusion of competence, such that JOLs are inflated by the apparent association between a cue and a target by a magnitude much greater than the true benefit the association has for recall performance (Koriat and Bjork, 2005).
Tiede and Leboe (2009) showed that overconfidence was found for homophones, synonyms, words spelled similarly, and unrelated items in their word pairs recall tests, whereas no overconfidence was found for word pairs with a relatively high forward-semantic association. Research from Finn and Melcalf (2007) has shown that judgments about upcoming test performance generally exhibit overconfidence, whereby they are higher on average than mean test performance. This finding is widely, but not always found.
The conclusion that people are usually overconfident is at odds with the findings of Koriat, Sheffer, and Ma’ayan (2002) showing that overconfidence is confined to the first trial in a multitrial learning situation. Judgment inaccuracy witnessed in the first trial as overconfidence, shifts toward underconfidence in Trial 2 and beyond. This phenomenon has been dubbed the Underconfidence With Practice effect. The primary goal of this study is to examine if the misuse of priming from one test can lead to overconfidence on the next.
Misuse can be illustrated by priming participants from the first test, but further testing information will not coincide with information from the first test. Inspired by the Koriat, Sheffer, and Ma’ayan (2002) study, I believe that with priming, overconfidence on performance with the following tests can be witnessed. I have found no research that connects the two topics of overconfidence and priming in any way, and hope to show a connection in this study. Method Participants The participants in this study were selected on the basis of where they work, as well as their willingness to aid in a psychological study.
All participants worked at —-. I requested their participation, however it remained optional. A total of 39 workers were willing to participate, however, only 36 participants were selected based on age restrictions of 18-65 years of age. This was the only limiting factor in their selection. Of the 36 selected, 27 were women and nine were men, and their ages varied from 19 years to 63 years of age. This test was administered at —-. Group assignment was based on availability of their time to complete the given tests during their shifts at the two locations.
The first testing period was on — where 21 participants were able to enter into testing, and the second testing session was on—- where the remaining 15 completed testing. Group assignments on these two days for testing was decided by alternating the distribution of test materials. Prior to testing, 40 identical tests were printed, 20 tests for the control group and 20 for the experimental group. The 20 tests for the control group and the 20 for the experimental group were ordered control test, experimental test, control test, etc.
During testing, each participant received identical tests first, and then an alternate test of control or experimental test. Instruments Actual distributed tests were amended from testing materials published by Pearson/ Longman Education for the testing of sixth through eighth grade students in reading comprehension. These testing materials were the only instruments used to compile the data. Procedure Testing was carried out on two subsequent days. During testing on day one 21 participants were present to complete testing.
Once all available participants were assembled, they were told about the experiment and how it was to be administered. Participants were told that this was a test in reading comprehension and confidence in their own comprehension. They were then told that they will receive two similar reading passages followed by a reading test for each passage. They were also warned that once they begin testing their passages will be collected and cannot be referred back to. After completion of the first test, they were asked to indicate whether they believed they would do better or worse on the second test by writing a whole umber score of how many questions they think they would get correct out of the 15 possible test questions. Once confidence of testing was established, participants were distributed either the control or experiment reading passage by an alternating method followed by the corresponding test. Testing for day two was conducted in the same manner with the exact same instructions for the remaining 15 participants. The testing materials were collected from each participant in order to analyze the data. . Results Participants’ mean scores correct for Test 1, Control Test and Experiment Test were 9. 3 (SD= 3. 3), 10. 06 (SD= 2. 94) and 10. 78 (SD= 3. 46) respectively. When comparing testing results from Test 1 and Control Test to determine whether a priming effect occurred, the results were not significant (t = -. 881, p >. 05) showing that there was not a significant priming affect, even though the mean correct scores were slightly higher for the Control Test than Test 1. Also when Test 1 and the Experiment Test are compared to determine whether Test 1 influenced a lower score for the Experiment Test, results were also not statistically significant (t= -2. 06, p > . 5) indicating Test 1 did not influence results for the Experiment Test. Testing for overconfidence indicated a slight under confidence for both the Control Test and the Experiment Test. The Control Test participants predicted mean confidence scores were 9. 94, while their actual mean correct scores were 10. 78. Experiment Test participants mean confidence scores were 9. 28, while their mean correct scores were 10. 06. Discussion When reviewing the results to determine if participants were primed from Test 1, both the Control and Experiment Tests were not influenced by Test 1.
This was contrary to the hypothesis in which it was predicted that Test 1 would prime participants to do better on the Control Test and worse on the Experiment Test. Test 1 did not statistically influence the results of the Control Test or the Experiment Test. Results for overconfidence testing did not support the hypothesis that overconfidence would be witnessed on the Experiment Test, but supported the findings of Koriat, Sheffer, and Ma’ayan (2002) that underconfidence would be witnessed after the first test, and in this case testing did result in slight underconfidence with participants in oth the Control and Experiment Tests. Both the priming effect and overconfidence are established conditions in psychology. However, in this experiment, neither were shown by participants. This experiment was meant to contradict the findings of Koriat, Sheffer, and Ma’ayan (2002) that underconfidence on testing would be witnessed after the first test, however, underconfidence was witnessed in this experiment.
After completing this study I believe that the findings of Koriat, Sheffer, and Ma’ayan are correct for underconfidence, but further research should be done to determine if priming can influence results on subsequent testing, especially in the case that information in testing is changed from the first test to subsequent tests. Possible problems for this experiment include limited design of the tests given to participants. Further testing should include established tests designed specifically to measure the priming affect. Additional testing should also include a larger sample size. References
Allwood, C. M. , Innes-Ker, A. H. , Holmgren, J. , & Fredin, G. (2008). Children’s and adults’ realism in their event-recall confidence in responses to free recall and focused questions. Psychology, Crime & Law, 14(6), 529-547. Finn, B. & Metcalfe, J. (2007). The role of memory for past test in the underconfidence with past test. Journal of Experimental Psychology: Learning, Memory, and Cognition, 33(1), 595-608. Graf, P. & Mandler, G. (1984). Activation makes words more accessible, but not necessarily more retrievable. Journal of Verbal Learning and Verbal Behavior, 23, 553-568. Jacoby, L.
L. & Dallas, M. (1981). On the relationship between autobiographical and perceptual learning. Journal of Experimental Psychology: General, 110, 306-340. Koriat, A. , & Bjork, R. A. (2006). Mending metacognitive illusions: A comparison of menomic-based and theory-based procedures. Journal of Experimental Psychology: Learning, Memory, and Cognition, 32, 1133–1145. Koriat, A. , Ma’ayan, H. , Sheffer, L. , & Bjork, R. A. (2006). Exploring a mnemonic debiasing account of the underconfidence with practice effect. Journal of Experimental Psychology: Learning, Memory, and Cognition, 32, 595-608.
Pallier, G. (2002). The role of individual differences in the accuracy of confidence judgments. The Journal of General Psychology, 129(3), 257-302. Stoltz, J. A. , Besner, D. , & Carr, T. H. (2005). Implications of measures of reliability for theories of priming: Activity in semantic memory is inherently noise and uncoordinated. Visual Cognition, 12(2), 284-336. Tiede, H. L. , Leboe, J. P. (2009). Illusions of competence for phonetically, orthographically, and semantically similar word pairs. Canadian Journal of Experimental Psychology, 63(4),