MultiLatentStateInstrumentalLearning

Copyright 2021, Warren Woodrich Pettine

This code contains simulations and data analysis used in "Pettine, W. W., Raman, D. V., Redish, A. D., Murray, J. D. “Human generalization of internal representations through prototype learning with goal-directed attention,” March 2023, Nature Human Behavior.

https://www.nature.com/articles/s41562-023-01543-7

For examples of running the code base, see the examples.py file in the main directory.

Installation

All code was written using python version 3.7. Please see the requirements file for required packages. To install the requirements, just run.

pip install -r requirements.txt

Organization

The code is organized in the following packages.

• simulations
• subjects
• utils
• analysis
• plots

See package modules for details on each function.

In addition, the main directory contains the module 'examples.py' that runs the algorithmic model on relevant experiments.

simulations

The simulations package contains all code necessary to simulate the various models. It is organized in the following modules:

• algorithmicmodels
• neuralnetwork
• parameters
• experiments

algorithmicmodels.py

This module contains the main code for the prototype and exemplar algorithmic models.

neuralnetwork.py

This module contains the code to run the neural network simulations. This module is unfinished.

parameters.py

Module for creating parameters dictionaries to create agents, networks and the worlds with with they interact.

experiments.py

Module that creates specific experiments used in the paper.

utils

This package contains the module utils.py, which organizes Various useful functions, such as those for unit testing.

analysis

This package contains Code for higher-level analyses of behavior and model fun

• general
• plotfunctions

general.py

Functions used for general analysis

plotfunctions.py

Functions used to plot the data