TestTamer: A CBT-Based Game to Reduce Test Anxiety Among High School Students

Overview

Hypothesis The objective of this project is to determine the effectiveness of a web-based gamified cognitive behavioral therapy (CBT) intervention in reducing test anxiety among high school students. Test anxiety can negatively impact academic performance, emotional wellbeing, and long-term educational outcomes, yet many students lack access to structured coping interventions. This project hypothesized that an interactive, scenario-based CBT game incorporating mindfulness breathing exercises and cognitive restructuring techniques would produce statistically significant reductions in self-reported test anxiety across multiple sessions. It was further hypothesized that students who use the CBT game will show a greater decrease in heart rate compared to the control activity. The ultimate goal was to engineer an accessible, scalable digital mental health tool capable of autonomously supporting students before high-stakes exams. Methodology Prior to experimentation, the web-based intervention was engineered to create interactive breathing visuals and structured thought-reframing exercises based on established CBT principles. Six iterative prototype sessions were developed to progressively reinforce skills such as diaphragmatic breathing, identification of negative automatic thoughts, and cognitive reframing. Participant assent and parental consent were obtained before participation, and all data was stored anonymously on a secure device. High school volunteers were divided into two groups: an experimental group (n=12) that used the CBT-based game and a control group (n=6) that engaged with a webpage with test anxiety information and skills to reduce test anxiety. Both groups completed identical pre-session and post-session Likert-based anxiety scales across six sessions. Open-ended qualitative responses were collected to assess perceived usefulness, engagement, and likelihood of future use. Additionally, a subset of students in each group had heart rate measured using a pulse oximeter before and after each intervention. Results Statistical analysis using paired-sample t-tests demonstrated statistically significant reductions in anxiety scores within the experimental group from pre- to post-intervention sessions. The experimental group showed consistent mean decreases in anxiety following gameplay across all six sessions. In contrast, while the control group showed minor fluctuations in anxiety scores, reductions were smaller in magnitude and less consistent over time. An independent two-sample t-test comparing overall change scores revealed that the experimental group experienced significantly greater anxiety reduction than the control group. Effect size analysis (Cohen’s d) indicated a large practical impact for the experimental condition, suggesting that the reduction was both statistically and meaningfully significant. Physiological data further supported these findings, as the experimental group demonstrated a progressive decrease in mean heart rate across sessions, as opposed to the control. Carryover scores increased steadily in the experimental group across session intervals, indicating increasing independent use of CBT techniques outside of gameplay, whereas the control group showed minimal growth. Qualitative responses reinforced these trends, with a majority of experimental participants reporting feeling more prepared and calm before tests. Additionally, most participants in the experimental group indicated that they would use the final prototype before a real exam, demonstrating strong adoption intent compared to the control condition. Conclusions The results support the hypothesis that a web-based CBT game like TestTamer can produce significantly greater reductions in test anxiety compared to traditional test anxiety reduction tips and skills. The statistically significant decreases in anxiety, large effect sizes, downward physiological trends, and increased behavioral carryover collectively demonstrate the effectiveness of the engineered system. Unlike traditional in-person therapy models, this intervention provides a scalable, low-cost, and autonomous alternative for students who may lack access to individualized counseling. These findings suggest that interactive digital CBT tools can meaningfully improve both emotional regulation and real-world coping behavior in academic settings. Future research should incorporate larger and more diverse populations, measure real-world academic outcomes, and explore hybrid models that combine short in-person instruction with CBT reinforcement to further optimize anxiety reduction outcomes. Overall, this project demonstrates the feasibility and impact of engineering digital mental health interventions tailored to adolescent test anxiety.

Competition history

• CSEF 2026 Behavioral & Social Sciences (Senior Division) · Entry S-03-01

Related projects

Closest projects by meaning, across every fair and year in the corpus.

Browse more like this

• All Behavioural and Social Sciences projects
• CSEF Behavioural and Social Sciences projects
• Every CSEF 2026 project