Editors' ChoiceNeuropsychiatric disorders

The neural substrates of super memory

See allHide authors and affiliations

Science Translational Medicine  29 Mar 2017:
Vol. 9, Issue 383, eaan0770
DOI: 10.1126/scitranslmed.aan0770

Abstract

Specific neural networks support superior memory in world-class memory athletes.

Deficits in memory function are a prominent feature of multiple brain disorders, including schizophrenia, Alzheimer’s disease, and traumatic brain injury. Classically, translational neuroscience studies have examined the differences in brain structure and function that exist between individuals with brain disorders and healthy control subjects. These studies have identified several key brain areas including prefrontal cortex and hippocampus in memory dysfunction, and newer functional imaging and electrophysiological studies have also implicated interactions between these regions. Despite these advances, the field has yet to achieve a clear mechanistic understanding of memory function that can be translated into new clinical therapeutics.

To expand on previous studies, Dresler et al. implemented an alternative approach based on comparing functional brain activity in control subjects with brain activity in world-class memory athletes. Using functional magnetic resonance imaging, they identified group differences in functional connectivity across key brain regions associated with working memory. They then tested whether mnemonic memory training induced these same network connectivity patterns in individuals who previously had no training. Not only did mnemonic memory training induce this connectivity pattern in naïve subjects, but the strength of these neural changes was directly correlated with memory gains exhibited by individuals 4 months after their initial training session. Critically, these neural changes and memory gains were absent in subjects who underwent a less effective memory training protocol or received no memory training at all.

Analysis of resting brain activity immediately before memory encoding revealed that the memory gains that result from mnemonic training are linked to increased functional connectivity between interconnected memory-related brain regions. These network-level interactions may prime the brain for subsequent encoding. Conversely, during periods of active memory encoding, enhanced functional connectivity within (and not between) these brain regions reflects training-induced memory gains. Together, these findings suggest that treatments that aim to enhance memory by targeting activity within brain regions may have to be delivered during encoding, whereas those that target network level brain processes may have to be delivered immediately before encoding. Thus, this study highlights memory-related brain processes that can be exploited for therapeutic development and optimization.

Highlighted Article

View Abstract

Navigate This Article