Memory recall involves a transient break in excitatory-inhibitory balance

Renée S. Koolschijn, Anna Shpektor, William T. Clarke, I. Betina Ip, David Dupret, Uzay Emir, Helen C. Barron

TL;DR

Memory recall isn't just about retrieving information; it involves a temporary shift in the brain's excitation-inhibition balance, particularly in visual cortex. This shift, marked by a transient increase in glutamate relative to GABA, is orchestrated by the hippocampus, suggesting it acts as a gatekeeper for memory access.

What Problem Does This Paper Address?

Our brains are adept at storing and recalling memories. The hippocampus is widely believed to play a crucial role in this process by acting as a "memory index," coordinating the reactivation of neural patterns stored across the neocortex. However, the precise mechanism by which the hippocampus achieves this coordination, especially how it "opens the door" to access dormant memories, has remained elusive. This paper investigates how the balance between excitatory (glutamate) and inhibitory (GABA) neurotransmission in neocortical circuits might be dynamically modulated to facilitate memory recall.

Methods

The researchers employed a novel, combined functional Magnetic Resonance Imaging (fMRI) and Magnetic Resonance Spectroscopy (fMRS) technique. This allowed them to simultaneously measure:

  • fMRI: Blood Oxygen Level-Dependent (BOLD) signal, reflecting neural activity, in both the hippocampus and visual cortex.
  • fMRS: Time-resolved concentrations of glutamate and GABA in the primary visual cortex (V1).

Participants underwent a three-stage task:

  1. Associative Learning: Learning associations between auditory and visual cues.
  2. Conditioning: Learning which visual cues predicted a rewarding or neutral outcome.
  3. Inference Test (during fMRI-fMRS scan): Participants heard auditory cues and had to infer the associated outcome by recalling the learned visual cue. This stage was designed to probe memory recall.

Trials were classified as "remembered" or "forgotten" based on participants' performance during the inference test and a subsequent surprise associative test, ensuring a conservative measure of recall.

Datasets

The study utilized data from 22 healthy human volunteers. No new benchmarks were introduced; the study relied on existing behavioral paradigms and neuroimaging techniques.

Key Insights and Ideas

  • Memory Recall as a Transient E/I Shift: The core idea is that recalling a memory isn't a passive retrieval but an active process that temporarily disrupts the brain's usual state of balanced excitation and inhibition.
  • Hippocampus as a "Gatekeeper": The hippocampus doesn't just store memories; it actively influences neocortical circuits to allow access. This suggests a dynamic, rather than static, role for the hippocampus in memory.
  • Disinhibition as the Mechanism: The proposed mechanism is "disinhibition" – the hippocampus might selectively reduce inhibitory activity in neocortical areas, thereby allowing the relevant memory representations to become active.
  • fMRS for Dynamic Neurochemistry: The study highlights the power of time-resolved fMRS to capture rapid neurochemical changes linked to cognitive events, moving beyond static measurements.

Key Results

  • Increased Glu/GABA Ratio During Recall: During the "question" period of the inference test (when participants were recalling to make a decision), there was a significant increase in the ratio of glutamate to GABA in the visual cortex for "remembered" trials compared to "forgotten" trials.
  • Hippocampal BOLD Predicts Neocortical Changes: Activity in the hippocampus (measured by fMRI BOLD signal) during "remembered" trials significantly predicted the observed changes in the glu/GABA ratio in visual cortex. Specifically, higher hippocampal activity was associated with lower GABA and a higher glu/GABA ratio.
  • Specificity of the Effect: This relationship between hippocampal activity and visual cortex neurochemistry was specific to the recall period and was not observed in control brain regions.
  • Behavioral Relevance: Performance on the inference test was directly predicted by participants' ability to recall the auditory-visual associations learned earlier.

Limitations and Future Work

  • fMRS Resolution: While the study achieved event-related fMRS, the temporal resolution is still limited compared to fMRI. Further improvements in fMRS acquisition and analysis could provide even finer temporal detail.
  • Metabolite Pools: fMRS measures bulk metabolite concentrations, which don't perfectly map to neurotransmitter release. The interpretation relies on established relationships between these pools and neuronal activity.
  • Clinical Populations: The authors suggest that future work could investigate these mechanisms in clinical populations with memory or E/I balance disorders (e.g., schizophrenia, autism, epilepsy), as disturbances in E/I balance are implicated in these conditions.
  • Cross-Species Comparisons: The use of a VR environment that mirrors paradigms used in rodents opens doors for future cross-species investigations into the cellular and circuit mechanisms underlying these findings.

Summary updated

Abstract

The brain has a remarkable capacity to acquire and store memories that can later be selectively recalled. These processes are supported by the hippocampus which is thought to index memory recall by reinstating information stored across distributed neocortical circuits. However, the mechanism that supports this interaction remains unclear. Here, in humans, we show that recall of a visual cue from a paired associate is accompanied by a transient increase in the ratio between glutamate and GABA in visual cortex. Moreover, these excitatory-inhibitory fluctuations are predicted by activity in the hippocampus. These data suggest the hippocampus gates memory recall by indexing information stored across neocortical circuits using a disinhibitory mechanism.
BibTeX
@inproceedings{koolschijn2021memory,
  author = {Renée S. Koolschijn and Anna Shpektor and William T. Clarke and I. Betina Ip and David Dupret and Uzay Emir and Helen C. Barron},
  title = {{Memory recall involves a transient break in excitatory-inhibitory balance}},
  booktitle = {eLife},
  year = {2021},
  doi = {10.7554/elife.70071},
}
Figure