Modeling decision bias in dockerHDDM: A Systematic Comparison and Evaluation

Authors

Siyu Wu (邬思宇), Wanke Pan(潘晚坷), Hu Chuan-Peng (胡传鹏)

Preprints

English: Modeling Decision Bias in dockerHDDM: A Systematic Comparison and Evaluation
中文: 使用dockerHDDM对决策偏差进行建模

Overview

This repository includes all code for reproducing the simulation and data analysis in our preprint, which systematically demonstrates how to appropriately model decision bias in the diffusion-decision model (DDM) with dockerHDDM. This preprint will help readers dispel the myth about accuracy-coding, stimulus-coding, z-bias, v-bias.

Folder Structure

The folder structure is similar to our preprints.

HDDM/
│
├── 1_Simulation/  
│   ├── plot_simulation_CRF.ipynb   # Plot simulated data conditional response functions    
│   ├── 1_1_starting_point_bias/
│   │   ├── data_generation.py    # Simulate DDM data with starting point bias
│   │   ├── param_extraction.py   # Parameter extraction & summary
│   │   ├── simulation_all.ipynb  # 9 different model specifications
│   │   ├── plot_simulation_fit.ipynb # Plot fitting & recovery
|   |   ├── ppc_simulation_zbias.ipynb # Results of posterior predictive check
│   │   ├── simulated_data.csv    # Generated simulated dataset
│   │   └── subject_params.csv    # Simulated subject-level parameters
│   └── 1_2_drift_bias/
│       ├── data_generation.py    # Simulate DDM data with drift bias
│       ├── param_extraction.py   # Parameter extraction & summary
│       ├── simulation_all.ipynb  # 9 different model specifications
│       ├── plot_simulation_fit.ipynb # Plot fitting & recovery
|       ├── ppc_simulation_vbias.ipynb # Results of posterior predictive check
│       ├── simulated_data.csv    # Generated simulated dataset
│       └── subject_params.csv    # Simulated subject-level parameters
│
├── 2_Empirical_Reanalysis/          
│   ├── 2_1_starting_point_bias/
│   │   ├── starting_point_bias.ipynb    # 9 different model specifications
│   │   ├── plot_fit.ipynb  
│   │   └── ppc_starting_point_bias.ipynb  # Results of posterior predictive check
│   ├── 2_2_drift_bias/
│   │   ├── drift_bias.ipynb   # 9 different model specifications
│   │   ├── plot_fit.ipynb  
│   │   └── ppc_starting_point_bias.ipynb  # Results of posterior predictive check
│   └── Data_White_2014/             # Real experimental data (White et al., 2014)
│       ├── Memory_proportion/         # Memory proportion condition
│       ├── Memory_rule/              # Memory rule condition
│       ├── Percept_proportion/       # Perceptual proportion condition
│       ├── Percept_rule/             # Perceptual rule condition
│       └── plot_CRF.ipynb            # Conditional response function visualization
│
└── 3_Supp_Combine_Bias/                
    └── combine_bias_recover.ipynb # Simulation and parameter recovery

Prerequisites

• Python environment
  • Python 3.10+
• Required packages

  pip install hddm kabuki pandas numpy matplotlib seaborn arviz

  • or use Docker (recommended), see:

Pan, W., Geng, H., Zhang, L., Fengler, A., Frank, M. J., Zhang, R.-Y., &Chuan-Peng, H. (2025). dockerHDDM: A User-Friendly Environment for Bayesian Hierarchical Drift-Diffusion Modeling. Advances in Methods and Practices in Psychological Science, 8(1). https://doi.org/10.1177/25152459241298700

This repository is licensed under the Creative Commons Attribution 4.0 International License.

Contact

If you have any questions, please contact us:

• Hu Chuan-Peng: hcp4715@hotmail.com