Diversity and Representation in Developmental EEG: Participant Recruitment and Documentation

• In Event: Towards a More Open, Reproducible, and Representative Field of Developmental EEG Research

Abstract

As developmental EEG research becomes increasingly influential in informing policy (e.g., Troller-Renfree et al., 2021) and clinical practice (e.g., McPartland et al., 2020), it is important to ensure such work is reproducible and generalizable. However, compared to other areas of developmental science and other neuroimaging modalities, the representation and reporting standards of developmental EEG has received little attention. We carried out a systematic review of six of the top journals for developmental psychology with the most published EEG papers in the last ten years (since 2011) to provide an initial assessment of previous and current practices involving participant recruitment and documentation. We first identified the journals with the most published EEG manuscripts that included children using PubMed. Given our focus in the use of EEG to study developmental psychology, we excluded clinical journals using EEG to diagnose seizures (e.g., Epilepsia or Epilepsy and Behavior). From this analysis, our systematic review included the following journals: Developmental Cognitive Neuroscience, Developmental Psychobiology, Developmental Science, Child Development, PLoS One, and Scientific Reports. We screened titles and abstracts of 969 studies to identify 546 empirical articles reporting on EEG from children (newborns to 17-year-olds). Thus far, sample size and demographics have been recorded for 389 articles. Results revealed that most studies did not report the minimum demographic information necessary to examine study generalizability. Of the studies reviewed, only 34.3% reported race, 19.7% reported ethnicity, 37.3% reported SES, and 89.3% reported gender. Importantly, these results did not seem to change based on the year the study was published from 2011 to 2022. However, reporting differed between journals. For example, most papers published in Child Development (71.4%) and Developmental Psychobiology (63.0%) reported race, whereas this was not true of the other journals reviewed (<26.5%). Of the studies who reported race, samples consisted of mostly White (70.6%) participants. Moreover, the median total sample size was 65 participants. This number reflects the number of participants before removing participants (e.g., due to lack of artifact-free trials) for many studies, indicating that in the best-case scenario most studies were adequately powered (.80) to detect medium effect sizes (d ≥ .69, r ≥ .33). Although far from acceptable, these results are considerably better than similar analyses of EEG and fMRI in adults (Clayson et al., 2019; Goldfarb & Brown, 2022). Moreover, the differences between journals highlight that most studies (especially if federally funded) likely have this information but fail to report it in manuscripts. It further suggests that improvements in reporting practices can be achieved likely from changes in editorial requirements. Future analyses will increase our sample to examine other potential moderators of these findings like participant age, study design (e.g., longitudinal vs. cross-sectional), and type of EEG analysis. Our findings highlight the need for increased transparency in reporting developmental EEG results, as well as research on larger and more diverse samples of children. Finally, in order to encourage increased reporting and recruitment of diverse samples, we will discuss barriers to increasing representation in developmental EEG studies, as well as potential solutions.

Authors

• Santiago Morales, University of Southern California
  Presenting Author

• Lauren Oh

• Nidah Mohammed

• Leona Abrahamian, University of Southern California

• Jingqi Li, University of Southern California

• Jenae Cipolla, University of Southern California