Understanding the biology of memory is a major goal of contemporary neuroscientists. Short-term or "working" memory is an important process that enables us to interact in meaningful ways with others and to comprehend the world around us on a moment-to-moment basis. A study published this week in Science (February 18) presents a strikingly simple yet robust mathematical model of how short-term memory circuits in the brain are likely to store, process, and make rapid decisions about the information the brain receives from the world.
A classic although purely practical example of working memory is our ability to look up a telephone number, remember it just long enough to dial it, and then promptly forget it. However, working memory is fundamental to many other cognitive processes including reading, writing, holding a conversation, playing or listening to music, decision-making, and thinking rationally in a general sense.
Cold Spring Harbor Laboratory computational neuroscientist Carlos Brody explores how brain neurons interact with each other to form the circuits or "neural networks" that underlie working memory and other rapid and flexible cognitive processes.
In the new study, Brody's group developed a mathematical model for interpreting data collected at Universidad Nacional Autónoma de México by his collaborator Rodolfo Romo. Romo's group measured brain neuron activity of macaque monkeys while the animals performed a simple task that involves working memory.
In one version of the task, animals were trained to compare an initial stimulus (a vibration applied to a fingertip) with a second stimulus applied a few seconds later and to immediately provide a "yes" or "no" answer to the question "was the first vibration faster than the second?"
This behavior requires the animals to load the initial stimulus into their working memory ("loading phase"), hold information about that stimulus in their working memory ("memory phase"), and compare that information to the second stimulus and make a decision based on the comparison (the "decision phase").
At the outset of the study, Brody and Cold Spring Harbor Laboratory postdoctoral fellow Christian Machens hoped to develop a mathematical model--based on known properties of "spiking" neurons--that would explain how the brain carries out just the memory phase of the behavior.
To their surprise, the simple "mutual inhibition" model they developed yielded a neural network architecture that explains not only the memory phase, but also the loading phase and the decision phase of the behavior. The model makes several predictions about the neurological basis of working memory that can be tested to confirm the likelihood that the model is a significant advance toward understanding fundamental properties of brain structure and function.
The above post is reprinted from materials provided by Cold Spring Harbor Laboratory. Note: Content may be edited for style and length.
Cite This Page: