Low-Cost Approaches in Neuroscience to Teach Machine Learning Using a Cockroach Model.

Area of Science:

• Neuroscience education
• Animal behavior analysis
• Machine learning applications

Background:

• Neuroscience education access is limited in underserved communities.
• Analyzing animal behavior is complex for non-experts.
• Developing accessible tools for behavioral research is crucial.

Purpose of the Study:

• To create an educational program for neuroscience using accessible tools.
• To quantify cockroach (Gromphadorhina portentosa) place preference and behavior.
• To demonstrate the utility of SLEAP (Estimates Animal Poses) software for behavioral analysis.

Main Methods:

• An educational program was developed using cockroaches on a linear track.
• Behavioral data was collected over 14 days with varying stimuli (air, vapor, nicotine).
• The machine learning software SLEAP was used for low-cost, cloud-based data analysis.

Main Results:

• SLEAP demonstrated high accuracy (within 0.5% margin of error) compared to manual scoring.
• Cockroaches exhibited an aversive response to vapor alone compared to air.
• SLEAP facilitated classification of x-y coordinate data into distinct behaviors using clustering methods.

Conclusions:

• The linear track and SLEAP provide an efficient method for studying nicotine preference in cockroaches.
• The developed educational program offers free access to machine learning for animal behavior studies.
• This initiative enhances neuroscience education accessibility through cost-effective technology.