distributionalRL

This repo contains the code for generating figures in Lowet et al., Nature, 2025 [PDF]. In order to run, it is also necessary to download the data from Dryad, which should be placed in a directory alongside the code folder as follows (this repo is equivalent to the code folder):

├── parent_dir
│   ├── code
│   ├── ├── neural_analysis
│   ├── ├── behavior_analysis
│   ├── ├── ann_decoding
│   ├── ├── envs
│   ├── ├── ...
│   ├── data
│   ├── ├── neural-plots
│   ├── ├── behavior-plots
│   ├── ├── ann_decoding
│   ├── ├── behavior
│   ├── ├── ...

The necessary conda environment for running the code is available in envs/environment.yml. For glm_analysis, the tf.yml file environment should be used instead. To do so, ensure conda is installed and then execute the command conda env create -f environment.yml.

In order to match the plotting style used in the paper, it is necessary to copy the paper_export.mplstyle file (found in the utils folder) to your environment-specific matplotlib stylelib folder. On my computer, this is found in ~/anaconda3/envs/neural/lib/python3.8/site-packages/matplotlib/mpl-data/stylelib/, although yours may differ depending on your conda installation.

Organization

The code is formatted as Jupyter notebooks. There are ten such notebooks, each located within the relevant subfolder.

1. neural_analysis/recording_figs.ipynb plots neural recording data (mostly Figs. 2, 3, 4 and ED Figs. 2-4, 6-8, and 10d-e.
2. behavior_analysis/compare_optostim.ipynb plots optogenetic stimulation data (Fig. 5 and ED Fig. 7a-m, 11).
3. behavior_analysis/licking_all_sessions.ipynb plots licking data (Fig. 1c, ED Fig. 1f-g, 8b, 10c).
4. behavior_analysis/behavioral_decoding.ipynb plots (Fig. 1d, ED Fig. 1h, 9b-c).
5. neural_analysis/plot_smoothed_data.ipynb plots Fig. 1f-g., ED Fig. 2b, 10a-b
6. neural_analysis/plot_sample_data.ipynb plots Fig. 1h, 4d,g.
7. neural_analysis/glm_analysis.ipynb plots data from ED Fig. 5 and 9d-f.
8. neural_analysis/compare_fano.ipynb plots Fano factor analysis (ED Fig. 6).
9. ann_decoding/ann_decoding.ipynb plots data from ANN-based decoding (ED Fig. 4f-l).
10. behavioral_analysis/plot_facemap_components.ipynb plots ED Fig. 1e.

They should all run after following the instructions above. If they don't, post an issue and I'll investigate!