Quantitative EEG Normative Databases: A Comparative Investigation

Abstract

Introduction. No clearly defined or universally accepted standards exist which practitioners and researchers can use to determine which quantitative electroencephalographic (QEEG) database is suitable to their needs. Diverse computational and methodological approaches across QEEG databases have been vigorously defended by their respective proponents and commonly misunderstood by practitioners. The purpose of this paper is to facilitate widespread discussion from which a universal set of standards can be agreed upon and applied to QEEG databases. Method. A broad set of criteria was developed from an extensive literature review and included issues of sampling, acquisition, hardware/software, control of confounding variables, and additional issues associated with disclosure, accessibility, and the screening of potential users. These criteria were then applied to the Hudspeth, John, Sterman-Kaiser, and Thatcher databases. Results. Results revealed reasonable concordance in data acquisition methods despite departures in inclusion/exclusion criteria and sample sizes. Significant differences were apparent in the controls used for possible confounding variables and the relative importance given to these variables. Conclusions. Research, clinical, and ethical implications are discussed, and it is recommended that the QEEG scientific community establish peer-review procedures and processes which prevent database manufacturers from seducing peers and clinicians with technocratic information and techniques that appear to confuse the user or oversimplify the complexity and richness of QEEG applications.