ERPscanr project repository: automated meta-analysis of the ERP literature.

Event-related potentials (ERP) are a common signal of analysis in psychology and neuroscience experiments, with a large existing literature of ERP-related work. This project uses automated literature collection and text-mining of published research articles to summarize the ERP literature, examining patterns and associations within and between ERP components.

This results of this project are hosted online on the project website.

To goal of this project is explore and summarize the existing ERP-related literature. In order to do so, we first manually curated a dictionary of all known ERP components that we could find. This list of ERP components and labels is defined in the terms sub-folder, and viewable in the SearchTerms notebook.

For data collection, this project employs two main approaches for collecting literature data:

• The 'Words' approach collects text and metadata, such as authors, journals, keywords and date of publication, from all articles that are found based on the search terms. This data is primarily used to characterize and build profiles of ERP components.
• The 'Count' approach collects data on co-occurrence of ERP terms and other pre-defined terms of interest, including cognitive and/or disorder-related association terms. This data is primarily used to examine patterns and similarities across ERP components, based on their associated topics.

Overall, the goal is to analyze patterns in the literature and make inferences about ERP components, by analyzing the collected data.

The easiest way to examine the outputs of this project is by visiting the project website, which includes individual profiles for all examined ERP components and group level analyses.

To explore how this project was done, and see the underlying code, you can explore this repository. As a starting point, the notebooks describe the approach used in this project. For doing literature analyses in general, see the LISC tool.

This project is described in the following paper:

Donoghue T & Voytek B (2021). Automated meta-analysis of the event-related potential
(ERP) literature. Scientific Reports, 12, 1867. DOI: 10.1038/s41598-022-05939-9

Direct link: https://doi.org/10.1038/s41598-022-05939-9

This project was written in Python 3 and requires Python >= 3.7 to run.

This project requires standard scientific python libraries, listed in requirements.txt, which can be installed with Anaconda Distribution.

Additional requirements include:

• lisc >= 0.2.0

This project repository is organized in the following way:

• build_site/ contains scripts to create the project website
• code/ contains code written and used for this project
• docs/ contains files that create and define the project website
• notebooks/ contains a collection of Jupyter notebooks that step through the project
• scripts/ contains stand alone scripts that run the data collection and analysis
• terms/ contains all the search terms used for the literature collection

Note that in order to re-run the analyses using the existing dataset, you will need to download the dataset (see below) and add this to a data folder within the project folder.

The code in this repository can also be used to run a new data collection, using the terms as defined in the terms sub-folder, and data collection scripts available in the scripts sub-folder.

Post data-collection, the main analyses of the data are done in the notebooks.

This project uses literature data.

The literature dataset that was collected and analyzed in this project is openly available at this OSF repository.