Related Experiment Video
Updated: May 6, 2025
05:41
A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis
Published on: February 6, 2020
9.3K
Analysis of Operant Self-administration Behaviors with Supervised Machine Learning: Protocol for Video Acquisition
Leo F Pereira Sanabria1, Luciano S Voutour1, Victoria J Kaufman1
1Department of Psychology, Michigan State University, East Lansing, Michigan 48823.
Eneuro
|January 8, 2025
Summary
We developed a new method using machine learning for automated analysis of animal behavior during operant self-administration. This approach efficiently processes video data to classify complex behavioral patterns.
Area of Science:
- Behavioral Neuroscience
- Machine Learning Applications
- Animal Behavior Analysis
Background:
- Automated analysis of operant self-administration behavior is limited.
- Supervised machine learning can analyze complex behavioral profiles from video recordings.
- Efficient behavioral analysis is crucial for understanding animal models.
Purpose of the Study:
- To provide a methodology for automated analysis of operant self-administration behavior.
- To demonstrate the utility of pose estimation for behavioral data generation.
- To develop predictive behavioral classifiers from training session data.
Main Methods:
- Video acquisition using Raspberry Pi or GoPro cameras.
- Pose estimation using DeepLabCut (DLC) and Simple Behavioral Analysis (SimBA) software.
- Comparison of Med-PC lever response data with pose-estimation-derived quadrant time data.
Main Results:
- Proof of concept for using pose estimation outputs from DLC.
- Generation of quadrant time results from pose estimation data.
- Successful acquisition of behavioral classifiers using SimBA.
Conclusions:
- The presented methodology enables efficient, automated analysis of operant self-administration.
- Pose estimation provides a viable alternative to traditional behavioral data collection.
- This approach facilitates the generation of predictive behavioral classifiers for operant tasks.
