Abstract
Event-related potentials (ERPs) are a common approach for investigating the neural basis of cognition and disease. There exists a vast and growing literature of ERP-related articles, the scale of which motivates the need for efficient and systematic meta-analytic approaches for characterizing this research. Here we present an automated text-mining approach as a form of meta-analysis to examine the relationships between ERP terms, cognitive domains and clinical disorders. We curated dictionaries of terms, collected articles of interest, and measured co-occurrence probabilities in published articles between ERP components and cognitive and disorder terms. Collectively, this literature dataset allows for creating data-driven profiles for each ERP, examining key associations of each component, and comparing the similarity across components, ultimately allowing for characterizing patterns and associations between topics and components. Additionally, by examining large literature collections, novel analyses can be done, such as examining how ERPs of different latencies relate to different cognitive associations. This openly available dataset and project can be used both as a pedagogical tool, and as a method of inquiry into the previously hidden structure of the existing literature. This project also motivates the need for consistency in naming, and for developing a clear ontology of electrophysiological components.
Highlights
Electroencephalography (EEG), and in particular evoked responses, have long been used to investigate relationships between neural activity, human cognition, and clinical disorders[1,2]
31,556 articles were identified across all 98 Event-related potentials (ERPs) components, reflecting publications from between 1964 and 2021
We find that ERP experiments continue to be a highly prevalent method, with the number of ERP articles per year continuing to grow (Fig. 2B)
Summary
Electroencephalography (EEG), and in particular evoked responses, have long been used to investigate relationships between neural activity, human cognition, and clinical disorders[1,2]. Related work in informatics includes literature-based discovery and hypothesis generation, in which databases of literature are used to curate knowledge, annotate terms, and infer data-driven hypotheses, often based on relatively simple term co-occurrence measures[17,18,19]. Such approaches have been employed in the biomedical literature[20] including within neuroscience, for example the NeuroSynth tool[21] for functional MRI, and the ’BrainSCANR’ project which analyzed patterns of associations in order to generate potential novel hypotheses[22]. There has been a relative lack of such work applied to EEG/ERP research, and there is no, to our knowledge, systematic, literature wide, attempt to analyze or curate the existing ERP literature
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
