8000 GitHub - shaox192/GoodBadAttention: Codes for "Is Attention Necessary for the Representational Advantage of Good Exemplars over Bad Exemplars?"
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

Codes for "Is Attention Necessary for the Representational Advantage of Good Exemplars over Bad Exemplars?"

Notifications You must be signed in to change notification settings

shaox192/GoodBadAttention

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

GoodBadAttention

Code base for Is Attention Necessary for the Representational Advantage of Good Exemplars over Bad Exemplars? (Shao & Beck (2024), European Journal of Neuroscience, 59(9), 2129-2415.)

Absract

Publicly available data:

  • BOLD beta estimates from each ROI (PPA, OPA, MPA) are available on OSF. There are outputs after finishing the preprocessing pipeline. Data are stored in HDF5 formats. Behavioral files not only contain subjects' responses and reactions times, but also has the order and timing of images presented.

  • Images used in the experiment and their representativeness scores are available here. These ratings scores and images were obtained from Torralbo et al., 2013.

  • Image rating experiment pre-registration details are also available here.


fMRI data preprocessing scripts

All data preprocessing scripts are in fmri_preproc/. We show all preprocessing scripts used to process our fMRI data. Details are in our Methods section. These scripts are written in tcsh, and needs manual adjustments, and visual inspections of processing results regularly, therefore are not directly runnable but serve as a demonstration of our procedures.

Similar pipeline was applied to both functional localizer for ROI extractions and main experimental runs for actual data with some differences. Below we also outline which steps we took for each type.

Code dependencies:

  1. FSL: FLIRT version 6.0
  2. AFNI: Version AFNI_20.0.4

Notes

  1. GLMs are in 3ddeconvolve* files. The one with .py extension was used to avoid manually input all trial regressors because we need one estimate for each trial for later MVPA analysis.
  2. ROI masks were manully isolated using te bucket files output after 3ddeconvolve on functional localizers. These masks do need to be intersected with whole brain mask, then warped to the exp 5ED1 erimental run space using T1 images because these two sessions take place on different days.

Main data processing scripts

Subsequent data processing that produced all of our results are in main_proc/. We included multiple analysis to understand the role of attention in the neural effects of statistical regularities in scene images. Below is a breakdown of code files used for each analysis.

> Prerequisite data files: Please download both neural and behavioral data files from here. and put them respectively into main_proc/neural_data/ and main_proc/behav_data/.

> Code dependencies: Use requirements.txt to install all dependencies.

About

Codes for "Is Attention Necessary for the Representational Advantage of Good Exemplars over Bad Exemplars?"

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published
0