By Aaron S. Richmond, Ph. D., Metropolitan State University of Denver
When giving guidance to students on how to take tests in your class, do you tell your students to always go with their first answer (go with their gut), or to always revise their answers, or that it depends on the question? Because many of you are fans of metacognition, likely you are wise and you choose the latter—it depends—and you would be correct. However, most students andmany teachers would choose “go with your gut instinct”, otherwise known as the First Instinct Fallacy (Kruger, Wirtz, & Miller, 2005). In this well-known article by Kruger and colleagues, they found (in 4 separate experiments) that when students change their answers, they typically change from incorrect to correct answers, they underestimate the number of changes from incorrect to correct answers, and overestimate the number of changes from incorrect to correct. Ironically, but not surprisingly, because students like to “go-with-their-gut”, they also tend to be very hesitant to switch their answers and regret doing so, even though they get the correct answer. However, what Kruger and colleagues did not investigate was the role that metacognition may play in the First Instinct Fallacy.
The [First] Instinct Fallacy: The Metacognition of Answering and Revising During College Exams
In two recent studies by Couchman et al. (2016), they investigated the mediating effects that metacognition may have on the First Instinct Fallacy. The procedure of both studies required students to complete a normal multiple-choice exam, indicate their confidence in their answers(whether they knew it or guessed the answer), and to indicate whether or not they changed their initial answer. Consistent with Kruger et al. (2005) results, Couchman and colleagues found that students more often change their initial response from incorrect to correct answers than the reverse. What was interesting is that when students thought they knew the answer and didn’t change their answer, they were significantly more likely to get the answer correct (indicating higher metacognition). When students guessed, and didn’t change their answer, they were significantly more likely to get the answer incorrect (indicating low metacognition). Moreover, when compared to questions students thought they knew, when students revised guessed questions, they choose the correct answer significantly more often than when they didn’t change their answer. In other words, students did better on questions when they guessed and changed their answer to when they thought they knew the answer and changed their answer. These results suggested that students were using the metacognitive construct of cognitive monitoring to deliberately choose when to revise their answers or when to stick with their gut on a question-by-question basis.
Moral of the Story: Real-Time Metacognitive Monitoring is Key to Falling Prey to the First-Instinct Fallacy
As demonstrated in Couchman and colleagues’ results, when student metacognitively monitor their knowledge and performance on a question-by-question basis, they will perform better. Metcalfe (2002) called this adaptive control—focusing on process that you can control in order to improve performance. Koriat et al. (2004) suggests that instead of reflective thinking in general on performance, in-the-moment and item-by-item assessment of performance may be more productive and effective.
So, you were correct in telling your students that “it depends”, but as a practitioner, what do you do to facilitate students’ ability to increase the metacognitive skills of adaptive control and monitoring? Couchman and colleagues suggested that teachers instruct their students to simplyindicate a judgment of confidence for each question on the test (either use a categorical judgment such as low vs. medium vs. high confidence or use a 0-100 confidence scale). Then, if students are low in their confidence, instructors should encourage them to change or revise their answer. However, if student confidence is high, they should consider not changing or revising their answer. Interestingly enough, this must be done in real-time, because if students make this confidence judgment at post-assessment (i.e., at a later time), they tend to be overconfident and inaccurate in their confidence ratings. Thus, the answer to the First Instinct Fallacy is—like most things—complicated. However, don’t just respond with a simple “it depends”—even though you are correct in this advice. Go the step further and explain and demonstrate how to improve adaptive control and cognitive monitoring.
Couchman, J. J., Miller, N. E., Zmuda, S. J., Feather, K., & Schwartzmeyer, T. (2016). The instinct fallacy: The metacognition of answering and revising during college exams. Metacognition and Learning, 11(2), 171-185. doi:10.1007/s11409-015-9140-8
Kruger, J., Wirtz, D., & Miller, D. T. (2005). Counterfactual thinking and the first instinct fallacy. Journal of Personality and Social Psychology, 88(5), 725–35.
Koriat, A., Bjork, R. A., Sheffer, L., & Bar, S. K. (2004). Predicting one’s own forgetting: the role of experience based and theory-based processes. Journal of Experimental Psychology: General, 133, 643–656.
Metcalfe, J. (2002). Is study time allocated selectively to a region of proximal learning? Journal of Experimental Psychology: General, 131, 349–363.