Tutorials

Welcome to the SPICE tutorials section. These tutorials will guide you through various aspects of using SPICE for cognitive modeling.

Available Tutorials

Data Preparation in SPICE
- Learn how to prepare and format your data for SPICE
- Understand the DatasetRNN class and data utilities
- Overview dataset splitting and preprocessing techniques
Basic Rescorla-Wagner Model
- Introduction to SPICE using a simple Rescorla-Wagner learning model
- Learn how to set up and train your first SPICE model
- Understand the basics of combining RNNs with equation discovery
Rescorla-Wagner with Forgetting
- Extend the basic model with forgetting mechanisms
- Learn how to work with multiple cognitive mechanisms
- Understand how SPICE discovers interaction effects
Working with Hardcoded Equations
- Learn how to use predefined equations in SPICE
- Understand when and why to use hardcoded equations
- Compare performance with discovered equations
Modeling Individual Differences
- Learn how to capture individual differences in cognitive models
- Work with participant-specific parameters
- Analyze and interpret individual variations
Case Study: Weinhardt et al. 2024
- Complete case study from recent research
- Advanced modeling techniques
- Real-world application example

Each tutorial is available in two formats:

As an interactive Jupyter notebook in the tutorials/ directory of the repository
As a web page in this documentation

To run the interactive notebooks:

Clone the SPICE repository:

git clone https://github.com/whyhardt/SPICE.git
cd SPICE

Install SPICE and its dependencies:

pip install -e .
pip install -r requirements-dev.txt

If you encounter any issues while following the tutorials: