Friday, December 25, 2009

How our brains keep multiple things in mind.

Siegel et al. have made a fundamental advance in revealing how the brain manages to keep multiple things in mind in our working (short term) memory system. A review by Vogel and Fukada in the same issue of PNAS gives a nice description of the context and nature of the experiments:
...brain oscillations are thought to provide a vehicle for coordinating and sharing information within a given cortical region as well as a means of communicating signals between different brain areas. Oscillations can occur across a number of different frequency bands, ranging from very slow cycles (4–7 Hz, theta band) to very fast cycles (25–100 Hz, gamma band). In the context of working memory, oscillations in the gamma band have been proposed to play a fundamental role in linking up the various attributes of the memoranda (e.g., position, shape, color, etc.) across numerous individual neurons into a unified working memory representation. However, if working memories are all represented in the same gamma oscillation, how do we manage to keep from blurring all of the active memories together? One solution that has been proposed in a number of computational models has been to keep the memories separated by positioning each one in a different phase within the oscillation. That is, individual memories can be kept segregated, so long as they are “out of phase” with one another in the oscillation.

...the study by Siegel et al. appears to provide the first demonstration of such a phase-coding scheme in the brain for working memory. To do this, they recorded the local field potential over the prefrontal cortex while monkeys performed a sequential short-term memory task. In this task, monkeys are shown two pictures, one at a time, that they had to remember. After a short delay, memory was tested by presenting three pictures simultaneously; two of which were the pictures they had seen earlier in the trial, and one was a novel picture. To perform correctly, the monkey responded by initially looking at the first picture in the original sequence, then looking at the second picture, but not the novel picture. This task requires them to actively remember both of the pictures from the sequence and the order of presentation. By examining the gamma oscillation over prefrontal cortex during the blank delay period while these memories were being maintained in mind, Siegel et al. found that the two remembered objects were represented in distinct phase orientations of the oscillation depending on the order of presentation. That is, the first object of the sequence was preferentially coded in one phase orientation, and the second object was always in a separate phase orientation. Thus, they found direct evidence that the brain kept these two active memories separated by keeping them out of phase.
From Siegal et al's abstract:
We recorded neuronal activity from the prefrontal cortices of monkeys remembering two visual objects over a brief interval. We found that during this memory interval prefrontal population activity was rhythmically synchronized at frequencies around 32 and 3 Hz and that spikes carried the most information about the memorized objects at specific phases. Further, according to their order of presentation, optimal encoding of the first presented object was significantly earlier in the 32 Hz cycle than that for the second object. Our results suggest that oscillatory neuronal synchronization mediates a phase-dependent coding of memorized objects in the prefrontal cortex. Encoding at distinct phases may play a role for disambiguating information about multiple objects in short-term memory.

No comments:

Post a Comment