SIT

Code and data for the social influence task (SIT), accompanying the paper:

Zhang, L. & Gläscher, J. (2020). A brain network supporting social influences in human decision-making. Science Advances, 6, eabb4159.
DOI: 10.1126/sciadv.abb4159.

---

Outreach:

• A 1.4-min #SciComm video in lay English is available on YouTube and bilibili.
• A 1-hour talk on this paper is available on YouTube and bilibili. The slides deck is available here.
• Part of the experimental setup was previously covered by a European television channel Arte Xenius (in German and French).
• A Twitter thread is compiled to summarize the main findings; see here for an unroll version.
• Media coverage (selection): COSMOS, UNIVIE, UKE (German), APA.at (German), EurekAlert, ScienceDaily, medicalxpress, SingularityHub.

---

This repository contains:

root
 ├── data # Preprocessed behavioral data & fMRI BOLD time series data
 │    ├── behavioral
 │    ├── fMRI
 ├── code # Matlab, R, & Stan code to run analyses and produce figures
 │    ├── behavioral
 │    ├── fMRI
 │    ├── stanmodel

Note 1: to properly run all scripts, you may need to set the root of this repository as your working directory.
Note 2: to properly run all modeling analyses, you may need to install the {RStan} package in R.
Note 3: to reproduce the Matlab figures, you may need the NaN Suite, the color brewer toolbox, the niceGroupPlot kit, and the offsetAxes function.

---

Behavioral analyses

• Figure 1B: plot_single_sub_data.m
• Figure 1D-E: plot_main_behav_within_trial.m
• Figure 1F-G: plot_acc_bet_within_trial.m
• Figure 1H-I: plot_main_behav_between_trial.m

Computational modeling

• Hierarchical Bayesian models written in the Stan language: code/stanmodel*
• Figure 2E-H: plot_m6b_winning.r --> The stanfit object needs to be downloaded at Figshare.
• Figure 2I-J: plot_param_and_behav.m
• Figure 3A: plot_dec_var_corr.m

* Interested in how to code computational models in Stan? Feel free to check out my BayesCog lectures (recipient of the 2020 SIPS Commendation, Society for the Improvement of Psychological Science).

fMRI BOLD time-series analyses

• Figure 3D-F, 4D: plot_time_series.m*
• core function for the time-series analyses: ts_corr_basic.m --> relies on normalise.m
• permutation test: ts_perm_test.m

* See our tutorial paper (Zhang & Lengersdorff et al., 2020) for more details regarding the justification/solidification of prediction error signals.

fMRI connectivity analyses

• Figure 4C,G,I: plot_connectivity_strength.m

---

For bug reports, please contact Lei Zhang ([email protected], or @lei_zhang_lz).

Thanks to Markdown Cheatsheet and shields.io.

---

LICENSE

This license (CC BY-NC 4.0) gives you the right to re-use and adapt, as long as you note any changes you made, and provide a link to the original source. Read here for more details.