Retrospective Confidence and Performance in Collaborative Gameplay Using Physiological Arousal: A Multilevel Analysis
This study of 87 university students in collaborative board games reveals that higher physiological arousal (NS-SCRs) predicts lower retrospective confidence and that while confidence aligns with perceived individual performance, it does not correlate with actual team outcomes, suggesting the need for targeted support during high-uncertainty phases.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer
Confidence is the quiet voice inside our heads that tells us we are on the right track. In learning and decision-making, this feeling guides us: it helps us decide whether to keep trying, to ask for help, or to stick with a plan. When that feeling is accurate, it helps us adapt; when it is wrong, it can lead us to stubbornly follow a bad path or give up on a good one. Scientists have long wondered where this feeling comes from. Does it come from a direct readout of how well we actually did, or is it a guess we make based on clues we gather while working? One such clue is our own body's reaction. When we face something difficult or uncertain, our nervous system often wakes up, causing our skin to conduct electricity slightly better as we sweat. This physical reaction, known as physiological arousal, is a measurable signal of effort and stress. The big question is whether this physical buzz is something we use to judge our own success, and if so, how it plays out when we are working in a group.
A team of researchers at Tampere University in Finland set out to explore this connection in a setting that felt like real life: a collaborative board game about entrepreneurship. They brought together 87 university students and grouped them into teams of three. Each team played a game called ESHIP, which simulated the uncertainty of starting a business. The players had to work together to make decisions, manage risk, and create a business idea, all while trying to keep their "uncertainty score" as low as possible. The game was broken into three distinct parts: a tutorial where they learned the rules, the main gameplay where they made decisions, and a final pitch where they presented their business idea to a camera. While the students played, the researchers attached small, unobtrusive sensors to their hands to measure their skin conductance, counting how often their bodies reacted with a spike of arousal. After the game, the students were asked to rate how confident they felt that their team's solution would succeed in the real world, and how well they thought they had performed individually.
The researchers were looking for a pattern between the physical spikes in arousal and the students' reported confidence. They found a clear, but specific, relationship. When a student felt more physically aroused than their own teammates during the game, they tended to report lower confidence in their team's solution. It was not about how aroused the whole team was compared to other teams; it was about how a single person felt relative to the people sitting right next to them. If a student's skin conductance spiked more often than their partners', they interpreted that extra physical energy as a sign that the task was harder for them, leading them to feel less sure of their success. This link was strongest during the tutorial and the main gameplay phases. During the tutorial, higher arousal relative to the group suggested the student was struggling to grasp the rules. During the gameplay, it suggested the decisions were difficult or the outcomes were uncertain. However, this pattern disappeared during the final pitch. The researchers found that the physical stress of standing up and presenting to a camera did not translate into lower confidence about the work itself. This suggests that students could tell the difference between the stress of being judged and the stress of actually doing the work.
Perhaps the most surprising finding was that this confidence did not match up with how well the teams actually did. The students who felt more confident did not necessarily have better scores in the game, and the teams that felt good about their work did not necessarily win. The researchers concluded that confidence in this setting was not a direct measure of success. Instead, it was a story the students told themselves based on how the process felt. They used their own physical sensations—how hard they were working compared to their friends—to build a sense of certainty. Because the game involved chance and complex rules that were not fully explained by the score alone, the students had no way to check if their feelings matched reality. They were left to rely on their internal cues, which turned out to be a poor predictor of the final outcome.
The study suggests that in collaborative learning, what we feel in our bodies matters more for our confidence than the final result. When we are working with others, we constantly compare our own effort to the people around us. If we feel more stressed or exert more effort than our teammates, we assume we are doing worse, even if the group is succeeding. This insight changes how we might think about teaching and teamwork. It implies that simply feeling good about a project does not mean the project is good, and feeling stressed does not mean we are failing. The researchers suggest that to build true confidence, learners need more than just the experience of doing the work; they need structured moments to reflect on what they did and to understand that their physical reactions might be about the difficulty of the moment, not their ability to succeed. By helping students separate the stress of the situation from the quality of their work, educators might help them develop a more accurate sense of their own capabilities.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.