Bilingual Code-Switching and Creative Thinking Project

ISEF Category: Behavioral and Social Sciences

Subcategory: Cognitive Psychology  ·  Difficulty: Intermediate  ·  Setup: School Lab  ·  Time: 1 to 2 Months

The Hook

Your brain may not think in one straight language lane. For bilingual students, switching languages can feel like changing gears in a car. This project asks whether that shift changes how many new ideas you can generate on an Alternate Uses Task, or AUT. If you design it well, you can test creativity, language control, and task order at the same time.

What Is It?

Code-switching means moving between two languages in the same conversation or thought stream. In this project, you are not testing whether bilingual students are better than monolingual students. You are testing whether using both languages while solving a problem changes performance on a divergent-thinking task.

The AUT asks people to list unusual uses for a common object, like a brick or paper clip. Think of it like trying to find many exits in one maze. Fluency counts how many ideas you give, flexibility counts how many different kinds of ideas you give, and originality counts how rare or unusual the ideas are. If code-switching changes any of those scores, that gives you a clean research question about language and creative thinking.

Why This Is a Good Topic

This is a strong science fair topic because you can test it with simple prompts, written responses, and clear scoring rules. It connects to bilingual education, classroom problem-solving, and how students handle open-ended tasks. You can learn how to design a fair comparison, score creative output, and control for language background without a hospital lab or expensive gear.

Research Questions

• How does code-switching before the AUT affect fluency scores in bilingual students?
• What is the effect of code-switching during the AUT on originality scores compared with single-language thinking?
• Does the order of language condition change AUT flexibility scores?
• To what extent does language dominance predict the size of any code-switching effect on divergent thinking?
• Which AUT prompts show the largest difference between code-switching and single-language conditions?
• How does bilingual proficiency relate to response quality after code-switching?

Basic Materials

• Printed AUT prompts in both language conditions.
• Participant consent and assent forms.
• Timed response sheets or a laptop with a response form.
• Randomized prompt order list.
• Digital stopwatch or timer app.
• Scoring rubric for fluency, flexibility, and originality.
• Spreadsheet for data entry and coding.

Advanced Materials

• Lab computer with PsychoPy or similar experiment software.
• Standardized bilingual proficiency measure.
• Headset microphone for spoken response capture.
• Counterbalanced task script and response coding sheet.
• Statistical software for mixed-effects analysis.
• Optional eye-tracking system for attention-shift analysis.
• Optional EEG setup for cognitive load comparisons.

Software & Tools

• PsychoPy: Builds and randomizes timed AUT tasks and can record response timing.
• JASP: Runs t tests, ANOVA, and mixed models with a simple interface.
• R: Handles mixed-effects models, effect sizes, and publication-ready graphs.
• Google Sheets: Organizes coded responses and helps you track scoring across participants.

Experiment Steps

1. Define the exact bilingual groups you will compare, including how you will measure language dominance and proficiency.
2. Choose one AUT scoring system so fluency, flexibility, and originality stay consistent across every participant.
3. Plan a counterbalanced order for the language conditions so practice effects do not look like code-switching effects.
4. Decide which background variables you will control, such as age, proficiency, and familiarity with the task.
5. Set your coding rules before you collect data, then stick to them when you score each response.
6. Pick the statistical test that matches your design, then plan how you will compare the two language conditions.

Common Pitfalls

• Mixing bilinguals with very different dominance levels, which can hide any code-switching effect.
• Scoring only raw idea count, which misses whether the responses are actually more varied or original.
• Letting participants see repeated prompts in the same order, which creates practice effects that look like language effects.
• Using prompts that depend on culture-specific vocabulary, which gives one language condition an unfair head start.
• Changing instructions across conditions, which makes the language shift and the task shift happen at the same time.

What Makes This Competitive

A stronger version of this project does more than compare average AUT scores. It separates fluency, flexibility, and originality, then checks whether language dominance, proficiency, and condition order change the result. If you use a crossover design, clean scoring rules, and the right statistical model, you can ask a sharper question than whether bilinguals simply score higher. That kind of design shows real control over confounders and measurement.

Project Variations

• Compare written AUT responses with spoken AUT responses to see whether output mode changes the size of the code-switching effect.
• Test balanced bilinguals and dominant bilinguals separately to see whether language dominance changes divergent-thinking scores.
• Replace the AUT with a category fluency task to see whether code-switching affects idea generation in a different way.

Learn More

• PubMed: Search review articles on bilingualism, code-switching, and divergent thinking.
• NCBI Bookshelf: Read free chapters on language processing and executive control.
• ERIC: Search education and psychology studies on bilingual learners and creative thinking.
• Open Science Framework: Find preregistration templates and examples for behavioral studies.
• MIT OpenCourseWare: Look for cognitive psychology and language lectures to study task design.