Paranoia and belief updating during a crisis (Suthaharan et al., 2021)

 Download zip file 
Help downloading and running models
Perceptual model with three hierarchical layers defined by probability distributions: (1) reward belief, (2) contingency beliefs, (3) volatility beliefs used to investigate the relationship between real-world uncertainty, paranoia, and laboratory task behavior.
1 . Suthaharan P, Reed EJ, Leptourgos P, Kenney JG, Uddenberg S, Mathys CD, Litman L, Robinson J, Moss AJ, Taylor JR, Groman SM, Colett PR (2021) Paranoia and belief updating during the COVID-19 crisis Nature Human Behavior
Model Information (Click on a link to find other models with that property)
Model Type: Multilayer Hierarchical Gaussian filter (HGF);
Brain Region(s)/Organism:
Cell Type(s):
Gap Junctions:
Simulation Environment: R;
Model Concept(s): Paranoia; COVID-19;
README.txt *
Directory for belief updating during a crisis

Praveen Suthaharan, Erin J. Reed, Pantelis Leptourgos, Joshua Kenney, Stefan Uddenberg, Christoph D. Mathys, Leib Litman, Jonathan Robinson, Aaron J. Moss, Jane R. Taylor, Stephanie M. Groman, Philip R. Corlett*


This directory includes:

1. Code
2. Data

The Data directory contains choice data for the HGF and behavioral and social data for the analysis:

 1. HGF - contains choice data for each period of the pandemic
 2. Figure1_boxplot_data.xlsx - data used to plot Figure 1 boxplots; reversals achieved and points earned data can be found in (11)
 3. Figure2A_confirmed_cases.csv - data used to overlay line graph representing covid cases across the pandemic period
 4. Figure4A_mapData.csv - data used to put together the mask-policy map
 5. Figure4B_protest.csv - data used to put together protest plot
 6. Figure4c_dynata.csv - data used to measure mask-wearing belief
 7. Figure5_CTL.csv - data used to match CTL feature in regression
 8. Figure5_regression.csv - data used to match for regression
 9. Figure5_us_cases_deaths.csv - data used to add confirmed covid cases feature into regression
 10. Figure7_pandemic_cloudresearch.csv - data used to put together the top maps representing CloudResearch recruitment
 11. pandemic.csv - main file containing behavioral data of N=605 individuals recruited during the pandemic period
 12. pandemic.xlsx - same as (9) but includes 'key' tab to elucidate demographic coding
 13. replication.csv - main file containing behavioral data of N=405 individuals recruited for the replication study
 14. us-employment.csv - file used to calculate the statistics for us-employment

*NOTE: (11) & (13) are the main files that also contain HGF belief data. Also, (11) and (13) contain the data for reproducing 
       Figures 3/6 and Figure 8, respectively.

The Hierarchical Gaussian Filter (HGF) toolbox v5.3.1 is freely available for download in the TAPAS package at We installed and ran the package in MATLAB and Statistics Toolbox Release 2016a (MathWorks, Natick, MA).


See Supplementary Methods.pdf for more detail

Loading data, please wait...