Mymou: A standalone touchscreen in-cage testing device

Download and install the latest Android release here

1) Reward Delivery Interface (Arduino)

• A simple script that runs on an Arduino Uno board
• It receives Bluetooth commands and then switches on or off digital out channels on the Arduino board as required
• These digital output channels can then be connected to your reward delivery system (e.g. peristaltic pump, pellet dispenser) to allow automated reward administration

2) Artifical Neural Network (Python)

• This is for training the system for facial recognition
• Runs in Python 3.5
• Takes image files collected with the Mymou system and then tunes the neural network to correctly identify the subject from an image
• Should work up to 99% accuracy

3) Behavioural Testing Unit (Android)

• Designed to run on Android 6.0.0. (Samsung Galaxy Tab A 10.1")
• Delivers the task to the subjects
• Can be easily programmed to deliver any tasks desired by the experimenter
• Example tasks are provided
• Uses Bluetooth to control the Reward Delivery Interface
• Also has built in facial recognition to identify which subject is playing when

4) Tablet Holder (OpenSCAD)

• A scale 3D model of the tablet holder used to mount the tablet to the side of the cage

Detailed information for each of these parts of the system can be found in the respective folders