Smell Test for Early Biomarker Screening

ISEF Category: Translational Medical Science

Subcategory: Disease Detection and Diagnosis  ·  Difficulty: Intermediate  ·  Setup: Home Setup  ·  Time: 1 to 2 Months

The Hook

Your nose can pick up warning signs before you notice them. In some diseases, smell changes show up early, sometimes years before bigger symptoms. That makes a simple scent test a smart place to start. You can turn everyday grocery-store smells into data.

What Is It?

This project asks a simple question, can you measure how well someone identifies common scents and use that score as a possible early signal of smell loss? Smell loss, or hyposmia, means reduced sense of smell. Doctors and researchers often study it because the olfactory system can change early in some neurodegenerative diseases.

Think of it like a phone camera losing focus. The image still looks like something, but details get fuzzy. In this project, you are not diagnosing anyone. You are building a screening tool that collects smell-identification scores and compares them with survey-based risk factors, such as age, family history, smoking, recent illness, or self-reported neurological symptoms.

Why This Is a Good Topic

This is a strong science fair topic because you can test a clear measurement, score scent recognition, and compare it across groups. You do not need a hospital lab to study the method. You can build a structured survey, standardize the scent test, and analyze whether certain factors relate to lower smell scores. The project connects to early disease screening, but it stays realistic for a high school researcher.

Research Questions

• How does age group affect odor identification scores on a grocery-store scent test?
• What is the effect of self-reported smoking history on smell test performance?
• Does recent upper respiratory illness change scent identification scores?
• To what extent do family history of neurodegenerative disease and smell scores correlate in a survey cohort?
• Which scent, coffee, vanilla, cinnamon, or peppermint, is most often identified correctly across participants?
• How does repeated exposure to the same scent set change test-retest consistency?
• What is the effect of test format, multiple choice versus free response, on odor identification accuracy?

Basic Materials

• Coffee, vanilla, cinnamon, and peppermint scent sources in sealed containers or odor cards.
• Identical small jars, sample cups, or scent vials with lids.
• Label stickers and a code key for randomizing samples.
• Printed consent and survey forms or a secure online survey form.
• Web app or form builder for administering the smell test.
• Timer or stopwatch.
• Notebook or spreadsheet for recording responses.
• Digital kitchen scale for keeping sample amounts consistent if needed.

Advanced Materials

• University or school IRB guidance for human-subject research planning.
• Standardized odor identification test materials for comparison.
• Air-tight odor delivery containers with controlled headspace.
• Calibrated environmental monitor for room temperature, humidity, and airflow.
• Statistical software for regression, ROC analysis, or mixed-effects modeling.
• ImageJ or similar tool if you use any visual rating scales or packaging checks.
• Optional olfactometer access for validation studies.

Software & Tools

• Google Forms: Collects survey responses, scent scores, and participant background data in one place.
• Google Sheets: Organizes responses, computes totals, and helps you spot patterns.
• Python: Runs correlation tests, regression models, and simple graphs for your smell score data.
• R: Handles statistical comparisons, confidence intervals, and subgroup analysis.
• ImageJ: Lets you document and compare sample containers or labels if you need visual quality checks.

Experiment Steps

1. Define the one smell score you will measure, such as total correct identifications or a weighted score for harder scents.
2. Choose a fixed scent set and decide how you will keep each sample consistent across participants.
3. Design your survey so the smell test and the risk-factor questions stay separate and easy to analyze later.
4. Plan controls that reduce bias, such as randomizing scent order and masking labels.
5. Build an analysis plan before you collect data, including how you will compare groups and handle incomplete responses.
6. Pilot the full workflow with a small group so you can fix confusing instructions and weak scents before the main study.

Common Pitfalls

• Letting scent strength vary between containers, which makes low scores reflect weak samples instead of real smell ability.
• Using scents that are too similar, which makes the test measure guessing instead of odor recognition.
• Changing the order of scents for different participants without randomization, which can create order effects.
• Collecting self-reported risk factors in a messy survey, which makes the later correlation analysis hard to trust.
• Ignoring recent colds, allergies, or nasal congestion, which can hide the link between smell score and the variable you want to study.

What Makes This Competitive

A stronger project would go beyond a simple yes-or-no smell quiz. You could validate your score with a clean test design, a clear statistical model, and a comparison group that makes the data more meaningful. You could also test whether one scent format works better than another, or whether your score predicts a risk pattern better than chance. The best projects treat the smell test like a measurement tool, not just a survey.

Project Variations

• Test whether odor identification differs between coffee, vanilla, cinnamon, and peppermint versus fruit scents or household cleaning scents.
• Compare free-response odor naming with multiple-choice odor recognition to see which format gives more reliable scores.
• Analyze whether smell scores change after a recent cold, seasonal allergies, or known nasal congestion, using the same web app design.

Learn More

• NIH PubMed: Search review articles on olfactory dysfunction, hyposmia, and neurodegenerative disease screening.
• NASA OSDR: Useful for learning how to structure human-data studies and metadata, even if your project is smaller.
• NCBI Bookshelf: Free textbook chapters on sensory systems and clinical research methods, found through the National Center for Biotechnology Information.
• NIH National Institute on Deafness and Other Communication Disorders: Background on smell disorders and basic olfaction science, found on the NIH site.
• CDC National Center for Health Statistics: Good for learning how health survey variables are defined and reported, found on the CDC site.