Bridging Big Data: Procedures for Combining Non-equivalent Cognitive Measures from the ENIGMA Consortium

Abstract

Investigators in the cognitive neurosciences have turned to Big Data to address persistent replication and reliability issues by increasing sample sizes, statistical power, and representativeness of data. While there is tremendous potential to advance science through open data sharing, these efforts unveil a host of new questions about how to integrate data arising from distinct sources and instruments. We focus on the most frequently assessed area of cognition - memory testing - and demonstrate a process for reliable data harmonization across three common measures. We aggregated raw data from 53 studies from around the world which measured at least one of three distinct verbal learning tasks, totaling N = 10,505 healthy and brain-injured individuals. A mega analysis was conducted using empirical bayes harmonization to isolate and remove site effects, followed by linear models which adjusted for common covariates. After corrections, a continuous item response theory (IRT) model estimated each individual subjects latent verbal learning ability while accounting for item difficulties. Harmonization significantly reduced inter-site variance by 37% while preserving covariate effects. The effects of age, sex, and education on scores were found to be highly consistent across memory tests. IRT methods for equating scores across AVLTs agreed with held-out data of dually-administered tests, and these tools are made available for free online. This work demonstrates that large-scale data sharing and harmonization initiatives can offer opportunities to address reproducibility and integration challenges across the behavioral sciences.

ENIGMA-Meditation: Worldwide consortium for neuroscientific investigations of meditation practices

“…As data accumulates with ENIGMA-Meditation, subsets of the aggregated data with comparable study designs and protocols (cross-sectional or longitudinal paradigms of functional or structural MRI), and / or homogenous samples (novices or experts, patients with specific disorders or healthy populations) can also be pooled and harmonized separately to enable targeted participant-level meta-or mega-analyses that address narrower hypotheses (e.g., neuroscientific mechanisms of MBSR to improve psychological distress or MBCT for MDD, functional activation changes during open-monitoring meditation, resting-state FC changes pre-to-post meditation retreats in healthy adults, etc.). Furthermore, neurobehavioural associations underlying meditation states, self-report measures and clinical scales may potentially be examined by harmonizing distinct but related assessment scores across sites and datasets (107,108). Taken together, such collaborative well-powered approaches may help to distinguish consistent, generalizable findings from false positives and discordant outcomes that are prevalent among smaller individual studies (106,109).…”