Researchers have frequently studied the machinations of memory -- specifically, how neurons stored the information gained from experience so that the same information can be recalled later. However, less is known about the underlying neurobiology of how we “learn to learn”--the mechanisms our brains use to go beyond drawing from memory to utilise past experiences in meaningful, novel ways.
A greater understanding of this process could point to new methods to enhance learning and to design precision cognitive behavioural therapies for neuropsychiatric disorders like anxiety, schizophrenia, and other forms of mental dysfunction.
To explore this, the researchers conducted a series of experiments using mice, which were assessed for their ability to learn cognitively challenging tasks. Prior to the assessment, some mice received “cognitive control training” (CCT). They were put on a slowly rotating arena and trained to avoid the stationary location of a mild shock using stationary visual cues while ignoring locations of the shock on the rotating floor. CCT mice were compared to control mice. One control group also learned the same place avoidance, but it did not have to ignore the irrelevant rotating locations.
The use of the rotating arena place avoidance methodology was vital to the experiment, the scientists noted, because it manipulated spatial information, dissociating the environment into stationary and rotating components. Previously, the lab had shown that learning to avoid shock on the rotating arena required using the hippocampus, the brain’s memory and navigation centre, as well as the persistent activity of a molecule (protein kinase M zeta) that is crucial for maintaining increases in the strength of neuronal connections and for storing long-term memory.
“In short, there were molecular, physiological, and behavioural reasons to examine long-term place avoidance memory in the hippocampus circuit as well as a theory for how the circuit could persistently improve,” explained Fenton.
Analysis of neural activity in the hippocampus during CCT confirmed that the mice were using relevant information for avoiding shock and ignoring the rotating distractions in the vicinity of the shock. Notably, this process of ignoring distractions was essential for the mice learning to learn as it allowed them to do novel cognitive tasks better than the mice that did not receive CCT.
Remarkably, the researchers could measure that CCT also improved how the mice’s hippocampal neural circuitry functioned to process information. The hippocampus is a crucial part of the brain for forming long-lasting memories as well as for spatial navigation, and CCT improved how it operated for months.
“The study shows that two hours of cognitive control training causes learning to learn in mice and that learning to learn is accompanied by improved tuning of a key brain circuit for memory. Consequently, the brain becomes persistently more effective at suppressing noisy inputs and more consistently effective at enhancing the inputs that matter,” observed Fenton.
The paper’s other authors were Ain Chung and Eliott Levy, NYU doctoral students at the time of the research; Claudia Jou a doctoral student at the City University of New York’s Hunter College and the Graduate Center; Alejandro Grau-Perales and Dino Dvorak, NYU postdoctoral fellows at the time of the study; and Nida Hussain, a student at NYU’s College of Arts and Science at the time of the study.