Hippocampal subfield and medial temporal cortical persistent activity during working memory reflects ongoing encoding

• Published in: Frontiers in systems neuroscience 2015-03, Vol.9, p.30-30
• Main Authors: Nauer, Rachel K, Whiteman, Andrew S, Dunne, Matthew F, Stern, Chantal E, Schon, Karin
• Format: Article
• Language: English
• Subjects: delayed matching-to-sample; high-resolution fMRI; Hippocampus; Medial Temporal Lobes; Neuroscience; working memory
• ISSN: 1662-5137; 1662-5137
• DOI: 10.3389/fnsys.2015.00030

Previous neuroimaging studies support a role for the medial temporal lobes in maintaining novel stimuli over brief working memory (WM) delays, and suggest delay period activity predicts subsequent memory. Additionally, slice recording studies have demonstrated neuronal persistent spiking in entorhinal cortex, perirhinal cortex (PrC), and hippocampus (CA1, CA3, subiculum). These data have led to computational models that suggest persistent spiking in parahippocampal regions could sustain neuronal representations of sensory information over many seconds. This mechanism may support both WM maintenance and encoding of information into long term episodic memory. The goal of the current study was to use high-resolution fMRI to elucidate the contributions of the MTL cortices and hippocampal subfields to WM maintenance as it relates to later episodic recognition memory. We scanned participants while they performed a delayed match to sample task with novel scene stimuli, and assessed their memory for these scenes post-scan. We hypothesized stimulus-driven activation that persists into the delay period-a putative correlate of persistent spiking-would predict later recognition memory. Our results suggest sample and delay period activation in the parahippocampal cortex (PHC), PrC, and subiculum (extending into DG/CA3 and CA1) was linearly related to increases in subsequent memory strength. These data extend previous neuroimaging studies that have constrained their analysis to either the sample or delay period by modeling these together as one continuous ongoing encoding process, and support computational frameworks that predict persistent activity underlies both WM and episodic encoding.