The impact of gamification on students’ learning, engagement and behavior based on their personality traits

Gamification doesn't work the same way for everyone and a team of Brazilian researchers has the data to prove it. Forty students, two versions of the same app, four months of logging in and writing code, and a personality questionnaire taken before anyone touched a line of Java. What Smiderle and colleagues found wasn't that gamification works or that it doesn't. It was something more precise: the effect depends entirely on who you're gamifying for. That finding lands harder when you understand how contested this space already is. Gamification in education means what you'd expect — taking the mechanics of games and dropping them into learning environments. Points for correct answers, badges for streaks and achievements, and leaderboards showing how you rank against classmates. The case for it sounds intuitive: games produce extraordinary levels of engagement, so borrow their tools. Researchers like Knutas and colleagues and Borges and colleagues argued it could enhance skills, support behavior change, even socialize learners. And plenty of studies backed that up. Hakulinen and Auvinen found gains in engagement and retention, while Tvarozek and Brza reported similar results. But the record isn't clean. Hanus and Fox found gamification actually decreased pleasure and motivation. Haaranen and colleagues documented adverse emotional reactions to badges specifically. Christy and Fox showed that rankings could affect women in complex, sometimes opposite ways. Mekler and colleagues got more activity but no improvement in quality. What the field was left with was a pile of contradictory findings and no satisfying explanation for why the same set of tools could help in one study and hurt in another. Smiderle and colleagues thought they knew where to look: personality. If different people respond differently to the same game mechanics, then measuring who those people are before the experiment starts might be the variable that makes sense of the mess. To do that, they used an instrument called the IGFP-5, a validated 44-item questionnaire measuring the Big Five personality traits: Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism. Think of them as five dials. Openness captures imagination and curiosity. Conscientiousness maps to organization and reliability. Extraversion is sociability and energy. Agreeableness is cooperativeness and warmth. Neuroticism tracks anxiety and emotional sensitivity. The IGFP-5 had been validated on a Brazilian sample of over 5,000 respondents, making it appropriate for this population. Each scale runs from 5 to 25, and with 40 participants, the researchers used the sample median to split students into high and low groups on each trait — a practical choice they later flag as a limitation. The experiment itself was carefully constructed. All 40 students used a web-based learning platform called Feeper alongside the BlueJ Java coding environment. The students were first-year undergraduates from two computing classes, ranging in age from 17 to 34, mostly male. They were randomly assigned: 21 to the non-gamified control group and 19 to the experimental group. Here's the key design move: everyone started on the plain version of Feeper. After the midterm exam, the experimental group switched to the gamified version, which visibly displayed points, badges, and rankings. The control group kept using the non-gamified interface, which still tracked scores internally but never showed them. The gamification was specific. Points declined by five for each incorrect submission, with a cap of 70 points lost per task. Badges came in nine types, each awarded at gold, silver, and bronze levels — a total of 27 possible badges — earned for things like logging in daily, submitting correct solutions, or topping the leaderboard. Rankings were displayed both at the class level and across the whole platform. Students met once a week for just under three hours, and Feeper was active in 15 of those sessions. Learning was measured through three exams — a midterm called Grade A, a final called Grade B, and a remediation exam called Grade C. Engagement was pulled from system logs: logins, points collected, badges earned, and how often students actually looked at the ranking and badge displays. That last measure, how often students looked at the leaderboard, turns out to be where the personality story gets most vivid. Smiderle and colleagues used Shapiro-Wilk tests to check data normality, then applied t-tests or Wilcoxon rank-sum tests as appropriate, reporting Cohen's d and nonparametric effect sizes alongside p-values. The clearest single finding: extroverted students in the gamified condition looked at the ranking far less than introverted students in the same condition — a strong negative correlation of negative 0.79 between extroversion and ranking views, with a p-value below 0.01. That's not a subtle effect. Extroverts weren't just slightly less interested in the leaderboard; they were largely ignoring it. And it wasn't just rankings. The correlation between extroversion and total points in the gamified group was negative 0.52. Between extroversion and accuracy — the proportion of submissions that were correct on the first try — it was negative 0.57. Between extroversion and badges collected, negative 0.52. All of these were significant. Introverted students, by contrast, accumulated more points and badges and showed gains in accuracy under gamification. The leaderboard, which competition-minded designers probably imagined would fire up the socially energetic extroverts, was doing essentially the opposite. It was the quieter students who engaged with it most. Other traits added texture to this picture. Students scoring low on openness and low on agreeableness in the gamified group showed significant accuracy improvement in the second half of the course, after the midpoint when gamification kicked in. High-conscientious students were accurate in both conditions, which is what you'd expect from organized, reliable people. But the finding for low-conscientious students is more interesting: those students in the non-gamified group experienced a significant accuracy drop from the midterm to the final. That same drop did not happen in the gamified group. Gamification appeared to act as a kind of scaffold for students who might otherwise have lost their footing as the course progressed. Neuroticism produced no meaningful differences between the two conditions, though neurotic participants did log in more frequently across the board. Step back and the pattern becomes clear. Gamification did produce a statistically significant improvement in solution quality for the experimental group overall. Accuracy improved from Grade A to Grade B in the gamified condition. But that aggregate result hides a disaggregated reality. The gains were concentrated in specific personality subgroups: introverts, students low on agreeableness, students low on openness, and students low on conscientiousness who needed the structure. For extroverts, the leaderboard and badge system didn't just fail to motivate — based on the correlations, it may have actively distanced them from the material. This is where the study's limitations matter, and the authors name them directly. Forty participants is a small sample. The cohort came from a single Brazilian university and skewed young and male. The domain was programming only. The median-split method for classifying personality is a rough tool. With a larger sample, you could focus on trait extremes rather than dividing everyone at the midpoint. Smiderle and colleagues call for replication with larger, more diverse groups, extension to other disciplines like mathematics and physics, and longer studies that can test whether gamification effects hold over time or whether students eventually stop responding to points and badges — what the paper calls saturation. The practical implication for educators isn't to abandon gamification or to fully embrace it. It's to treat it as a design question with a human variable. A leaderboard isn't a neutral tool; it's a tool that lands differently depending on the psychological profile of the person looking at it. The student who thrives on rankings may not be the one you assume. The one who quietly accumulates badges and improves their accuracy under gamified pressure might be the introvert you thought wasn't particularly competitive. Smiderle and colleagues have shown, in a real classroom over four real months, that the effect of adding game mechanics to learning is inseparable from the question of who is sitting at the keyboard. One-size-fits-all gamification is likely leaving some students behind while simultaneously failing others. The next step, personality-aware design, is harder to build, but this study makes a strong case that it's the right direction. This lecture was created by ennepō. Go to https://ennepo.ai to Discover, Create and Follow the latest research in your field. Read when you can. Listen when you want to.