Pipeline for LAT analysis

Data location

Data can be downloaded at (https://doi.org/10.5281/zenodo.10529801)

Behavioral data and model fits located in Data/Behavior_and_models Data for testing the fitting procedure located in Data/fitting_pipeline_validation Neural data located in Data/FR_around_event

Correspondence behavioral and neural data (ID)

Monkey B (Beaker)

181019 = 3 181025 = 8 181029 = 12 181030 = 14 181106 = 18 181109 = 20 181110 = 22 181115 = 26

Monkey S (Scooter)

181102 = 28 181103 = 30 181107 = 32 181108 = 34 181112 = 36 181113 = 38 181116 = 40 181117 = 42 181119 = 44

Reproduce the main figures (all the data necessary is in the corresponding folder)

Figure_1/plot_figure_1/Plot_figure_1_github.m Figure_2/plot_figure_2/Plot_figure_2_github.m Figures_3_and_4/plot_figure_3/Plot_figure_3_github_revised.m Figures_3_and_4/plot_figure_4/Plot_figure_4_github_revised.m Figure_5/plot_figure_5/Plot_figure_5_github.m Figure_6/plot_figure_6/Plot_figure_6_github.m

Reproduce the analyses (and the supplementary figures)

You need to set the paths yourself at the start of each script.

Behavioral data (Fig1 - S1-2)

You need to download the VBMC toolbox (https://github.com/lacerbi/vbmc)

In Figure_1/code

�doExecuteMasterScriptResetRWVBMC(fsroot,monkey,N_channels,analysis) save_Model_VBMC plot_figure1 (behavioral results per monkey. Fig 1 and S1) plot_models_VBMC (model comparison, Fig S1F) compare_control_models (model comparison, Fig S1G-H) plot_models_VBMC_2L (2 learning rate model comparison)

Plot_choice_model_behavior (Fig S2)

To test the fitting pipeline:

Test_parameter_recovery (generate data for each model) fit_all_models_cluster_proba (for each model and each generated data set) Test_parameter_recovery_proba (if LOAD=0, will use ‘Fitting_pipeline_validation_results’ in Data/fitting_pipeline_validation to plot Fig S1I-K)

Neural data

Figure 2 ans S3

In Figure_2/code

Uses: window [-600 300] around target, size = 900ms sliding windows around target, -600 to 350ms, size = 300ms sliding windows around reward end, 0 to 300ms, size = 300ms

models_FR_figure2_value_function (for window [-600 300]) plot_FigS2_prop_sig_models

prepare_pseudopop_peak_belief_classifier_prog_restricted_FEF prepare_bootstrap_peak_belief_classifier_across_time_prog => same with 4bins and rotated correspond to Fig S3H-I

Run_belief_classifier_combined_time_prog_restricted_FEF (for window [-600 300]) Run_belief_classifier_combined_time_prog_NN_restricted_FEF (for window [-600 300]) => same with 4bins and rotated correspond to Fig S3H-I, noOne removes neurons with only one block for a template bin

get_projection_classifier_combined_time_prog_restricted_FEF (for window [-600 300])

plot_Fig2_ET_pseudo_pop => same with 4bins and rotated correspond to Fig S3H-I, noOne removes neurons with only one block for a template bin

Run_belief_classifier_combined_time_prog_restricted_FEF (for sliding windows) Run_belief_classifier_combined_time_prog_NN_restricted_FEF (for sliding windows) get_projection_classifier_combined_time_prog_restricted_FEF (for sliding windows)

plot_Fig2_cross_temporal_decoding

plot_Fig2_PCA

models_FR_figure2_true_template (for window [-600 300]) plot_FigS2_true_template_prop_sig

models_FR_figure2_value_function (for sliding windows) plot_FigS2_prop_sig_models_across_time

Figures 3, 4 and S4

In Figures_3_and_4/code

Uses: window [-600 300] around target, size = 900ms

belief_peak_decoder_single_session_update_NN_restricted_FEF.m Plot_fig3_and_4.m MDS_template.m

ET_around_reset (Fig S4G)

Figure 5 and S5

In Figure_5/code

Uses: sliding window around target, -500 to 800ms, size = 200ms sliding window around response, -500 to 600ms, size = 200ms

prepare_pseudopop_choice_prog_classifier_across_time prepare_bootstrap_peak_belief_choice_across_time Run_classifier_peak_belief_choice_restricted_FEF Accuracy_classifier_peak_belief_choice

prepare_pseudopop_peak_belief_value_classifier_across_time prepare_bootstrap_peak_belief_value_across_time Run_classifier_peak_belief_value_restricted_FEF Accuracy_classifier_peak_belief_value

prepare_pseudopop_peak_belief_cc_classifier_across_time prepare_bootstrap_peak_belief_cc_across_time Accuracy_classifier_peak_belief_chosen_color_2bins

prepare_pseudopop_stim_color_across_time prepare_bootstrap_stim_color_across_time Run_stim_color_classifier_across_time_with_NN_restricted_FEF Accuracy_classifier_stim_color

Figure 6 and S6

In Figure_6/code

Uses: sliding window around target, -400 to 800ms, size = 200ms

get_explained_variance_split_restricted_FEF get_boot_glm_all_values_split_correct get_mean_EV_across_split_and_locs get_preferred_loc_restricted_FEF get_boot_glm_all_values_split_pref

plot_correlation_chosen_unchosen plot_correlation_local_global plot_correlation_across_locations plot_EV_across_loc plot_correlation_chosen_unchosen_pref