"Blips" of Knowledge Reduce Accuracy and Increase Confidence
[Cover image by Tung Nguyen on Pixabay]
by Cindy Nebel
What do you do when you get a random new physiological symptom? A new rash? A cough? That strange pain that comes and goes in your abdomen? If you’re like me, you do a quick internet search for possible causes or, more often, “When should I see a doctor for…?”
On a personal note, I tend to err on the side of seeing the doctor too often. These internet searchers usually make me feel better about my decision to see the doctor (because according to the internet, it’s probably something bad), but that’s not true for everyone. Often the decision to search online can cause someone to self-diagnose quickly and choose not to see a doctor or, perhaps, to see a doctor with a self-diagnosis in hand.
The study I’m reviewing today looks specifically at how short pieces of information impact our confidence in our knowledge and our accuracy (1). This is related to a popular phenomenon known as the Dunning-Kruger effect (2). In the Dunning-Kruger effect, individuals with little knowledge tend to be more confident that they know a lot than individuals with a lot of knowledge. Arguably, you have to know quite a bit to realize how much you have left to learn.
Image by Darko Stojanovic from Pixabay
In this study, participants were split into three groups: No knowledge, Short knowledge, and Long knowledge, referring to how much information participants were given about different diagnoses. In the No knowledge condition, participants read symptom descriptions only. In the Short knowledge condition, participants were given a sheet to study first that included a list of common illnesses and their typical symptoms. In the Long knowledge condition, they were given the same sheet, but allowed to review it throughout the study.
Unsurprisingly, individuals who were taking an open-book test (the Long knowledge condition) performed the highest, but more surprisingly the individuals who were given a brief review of common illnesses performed the lowest and showed the highest level of overconfidence. A little bit of knowledge made them feel like they could diagnose common diseases, but made them worse at it! When given the choice to learn more about the patient’s symptoms, that overconfidence meant that participants didn’t feel like they needed any additional information.
Implications for Medicine
The implications are clearest here for the field of medicine. If medical students, for example, have a shallow amount of knowledge, they might feel confident and not take the time to seek additional information from patients. The same can be true for physicians - overconfidence in diagnosing can lead to lower rates of correct diagnosis.
Image from Pixabay
Implications for Education
This article studies medical diagnostic information. The implications I’m about to share represent an assumption that this might translate to other domains.
In a world where information is not only available, but sometimes thrust upon us whether we like it or not, it’s important to be aware that these short “blips” of knowledge may be leading us to have high confidence in our understanding of an area, with low accuracy in our ability to apply it. A quick headline or social media post may give us a false belief that we understand a domain that we would have actually been more accurate in had we not received that information at all.
Let me say that again for the people in the back: Reading headlines without seeking additional information makes you more confident that you understand what’s going on while actually reducing your understanding.
In our classes, this may mean that our students stop studying when they have a cursory amount of information, feeling as though they know enough. They likely do not realize how much more there is to know. A little bit of knowledge may actually stunt their curiosity and interest to learn more.
What do we do about this? I would argue that this is where retrieval practice and elaborative interrogation lead to better learning. Retrieval practice has repeatedly been shown to improve metacognitive accuracy, reducing the overconfidence that comes from rereading. Elaborative interrogation asks how and why questions to deepen and organize the knowledge students have, leading to better understanding.
Final Thought
While this study has major implications in a world of rapid knowledge acquisition, it also signals that, as educators, we need to ensure that we are using the effective learning strategies that extend knowledge beyond the shallow level here that is so problematic. We need to be aware not just of students’ knowledge, but also their confidence in their knowledge and work to improve both to develop better self-directed learners.
Reference:
1) Hong, S. S., Son, L. K., & Kim, K. (2026). Metacognition and diagnostic decision-making: short" blips" of knowledge and the consequences of overconfidence. Cognitive Research: Principles and Implications, 11(1), 22.
2) Rahmani, M. (2020). Medical trainees and the Dunning–Kruger effect: when they don't know what they don't know. Journal of Graduate Medical Education, 12(5), 532-534.
