Mapping the parameter space in EEG/MEG analyses (mapMEEG)

Mapping the parameter space in EEG/MEG analyses (mapMEEG)

Idea and history

This project was initiated at the OHBM Hackathon 2020 (initial issue here), intially as a git-0/ no code project. It is a project with an educational goal.

The idea was to provide help for less experienced EEG/ MEG researchers to get started on pre-registration featuring MEG/ EEG by creating a pre-registration template in a bottom-up manner.

More specifically, we pursued the following steps:

identify existing pre-registrations featuring EEG/ MEG
classify those pre-registrations according to a few criterion using this Google form.
- which creates an overview in this Google spreadsheet (read-only link).
- for a list of pre-registrations still to be classified see this list (read-only link).
systematically screen those pre-registrations for the following criteria:
- What are steps that are regularly under-specified?
- What are steps that are regularly forgotten?
- What are steps for which researchers regularly deviate from their initially pre-registered pipeline?
- What are good practices that facilitate readability and understanding in the user?
come up with a simple framework for an EEG/ MEG pre-registration pipeline
create an adaptive shiny app that allows users to fill in typical EEG/ MEG pipeline parameters and automatically creates human-readable text that can be copy-pasted into a pre-registration
- See this repository
- Current prototype hosted on https://johalgermissen.shinyapps.io/mapMEEG_proto/

Future steps to do

Spreadsheet overview of existing pre-registrations:
- Code more pre-registration from the queue.
- rate pre-registrations for level of elaboration/ detail-
- compare pre-registrations to finally published journal articles to identify deviations.
- compare pre-registrations to registered reports (level of detail, deviations).
- identify pre-processing steps specific to MEG.
Shiny app:
- Separate tabs for data collection, data pre-processing, data analyses.
- Separate tabs for EEG and MEG.
- Allow researchers to fill in a certain step multiple times and feature it at different time points within their pipeline
- Allow researchers to determine the order in which steps are outputed in the text document
- Helper links with:
  - short explanation on each parameter.
  - short example.
  - reference to respective software options in commonly used packages (e.g. EEGLAB, Fieldtrip, MNE Python)
  - reference to further literature

How to contribute

If you'd like to get involved, join the hbmhack-mapMEEG channel on the brainhack mattermost. Join our channel

Alternatively, you can open a new issue on this repository itself if there is something you would like to discuss directly here. We're also happy about direct pull requests to improve our app.