Current Reporting Practices in Human Neuroscience Research

Abstract

Concerns for the replicability, reliability, and generalizability of human neuroimaging research have led to intense debates over sample size and open science practices, with more recent attention on the contributions of sampling and recruitment practices. Key to understanding the state of neuroscience research is an assessment of reporting practices that influence replicability, reliability, and generalizability. In this structured review, we evaluated reporting practice across three domains: (1) demographic (e.g., reporting participant race-ethnicity, age, any measure of socioeconomic position), (2) methodological (e.g., reporting recruitment methods, inclusion and exclusion criteria, why participants were excluded from analyses), and (3) open science and generalizability (e.g., analyses were preregistered, target population was stated). Included were 919 published MRI and fMRI studies from 2019 in nine top-ranked journals (N = 3,856 records screened). Reporting across domains was infrequent, with participant racial or ethnic identity (14.8%), reasons for missing imaging data (31.2%), and identification of a target population (19.4%) being particularly low/underreported. Reporting likelihood varied by study characteristics (e.g., participant age group) and was correlated across domains. The median sample size of studies was 55 participants. Study sample size, reporting frequency, was positively associated with two-year citation counts, providing some evidence that the complete reporting of demographic characteristics, methodological decisions, and open science and generalizability practices may not be as valued as study sample size. Recommendations for structural interventions at the journal level are proposed.