Engrams and Memory: My Scientific “Eras Tour”

Abstract

Memories are widely thought to be encoded by sparse populations of neurons, often referred to as engrams. My entry into this field began with a 2001 report in this journal showing that overexpression of the transcription factor CREB in the amygdala enhances fear memory in rats (Josselyn et al., 2001). Over the following two decades, work from my laboratory and others converged on a central idea: neurons compete for inclusion in an engram, and this competition is strongly influenced by their relative intrinsic excitability and recent activity. In this Progression article, I reflect on the scientific “eras” that shaped this view, tracing a path from early molecular manipulations of CREB (the Debut era) to more modern approaches that identify, image, and manipulate memory-relevant neuronal ensembles. Together, these studies suggest that memories are not written onto a blank neural slate but emerge from competitive interactions within preexisting neural circuits. By revisiting these discoveries through a personal lens (and with occasional inspiration from Taylor Swift's “Eras”), this article reflects on how our ideas about engrams evolved from pure speculation and theory to experimentally testable mechanisms.