Introduction to the Linear Mixed Model for Psychophysiological Research

Abstract

Traditionally, psychophysiological studies have relied on paired t-tests and repeated-measures analyses of variance to analyze data. However, the linear mixed model (LMM) has recently emerged as a flexible and powerful alternative. This article introduces the LMM to encourage its use in psychophysiological research. First, we compare the LMM with traditional statistical approaches, highlighting key differences and advantages. Next, we describe the structure and parameters of the LMM using hypothetical data. Finally, we illustrate how to apply an LMM to real psychophysiological data. The LMM's capacity to process all available data by accommodating random effects—often overlooked by traditional methods—is expected to improve the validity of statistical inferences in psychophysiology.