hasha / cosmo2017

Data and scripts for the CoSMo 2017 summer school

Scripts for CoSMo 2017 summer school

Notes by Jan Drugowitsch, July 2017

Datasets

phs_subj_id.mat: Palmer, Huk & Shadlen (2006) data
phs_data_README.txt: describes the format of phs_subj_id.mat files
rs_subj_id.mat: Roitman & Sahdlen (2002) data
rs_data_README.txt: describes the format of rs_subj_id.mat files

Behavioral data plots

plot_psych_chron.m: Plots chronometric / psychometric curves
plot_speed_accuracy.m: Plots chronometric / psychometric curve with RT median splits
plot_rt_quant.m: Generates RT quantile plots
plot_rt_dist.m: Plots RT distributions
plot_pcorrect_over_time.m: Plots how p(correct) changes over time

Fitting behavior

fit_psych_chron.m: Fits psychometric / chronometric curve with diffusion model
plot_fitted_rt_dists.m: Fits behavior with diffusion model and plots RT distributions

Decision-making models

collapse_gain.m: Script comparing rewards between constant and collapsing-bound DDMs
plot_fpt_vary_bound_example.m: Plots RT distribution examples for constant vs. collapsing bounds
fpt_moments.m: Returns the RT moments for given RT distributions
accum_evidence_reward.m: Plots expected reward for accumulating evidence
accum_evidence_discrete.m: Plots discrete-time evidence accumulation with point-wise prior
accum_evidence_continuous.m: Plots continuous evidence accumulation with point-wise prior
accum_evidence_gauss.m: Plots continuous evidence accumulation with Gaussian prior

Dynamic programming

dp_discretized_g.m: Returns a vector of discretized beliefs
dp_value_iteration_point_hyp.m: Performs value iteration to find value function for point-wise prior
dp_getvalues_gauss_hyp.m: Uses backwards induction to find value function for Gaussian prior
dp_g_trans_point_hyp.m: Returns the belief transition matrix for point-wise prior
dp_g_trans_gauss_hyp.m: Returns the belief transition matrix for Gaussian prior
dp_valueintersect.m: Interpolates the belief at which two value functions intersect plot_dp_bound_direct_maximization.m: Performs direct diffusion model reward maximization
plot_dp_valueintersect_point.m: Plots value function intersection for point-wise prior
plot_dp_valueintersect_gauss.m: Plots value function intersection for Gaussian prior
plot_g_trans_point_hyp.m: Plots belief transition matrix examples for point-wise prior
plot_g_trans_gauss_hyp.m: Plots belief transition matrix examples for Gaussian prior
plot_dp_diffusion_point.m: Plots optimal diffusion model example for point-wise prior
plot_dp_diffusion_gauss.m: Plots optimal diffusion model example for Gaussian prior
plot_const_bound_gauss.m: Plots bound in belief for constant-bounded DDM for Gaussian prior