The human brain can store salient information, such as numbers, instructions or other details that are important for completing a task, for brief periods of time, via a cognitive system known as working memory (WM). This system is known to be constructive in nature, meaning that it does not only store information, but also organizes it in ways that can help tackle real-world problems more efficiently.

Researchers at Peking University recently carried out a study aimed at better understanding the neural mechanisms underpinning the hierarchical organization of information that is temporarily stored by the human brain. their findings, published in Nature Human Behaviorsuggest that the WM’s hierarchical organization of sequences is encoded in the two-dimensional (2D) geometry of neural representations.

“Instead of passively retaining information, WM reorganizes complex sequences into hierarchically embedded chunks to overcome capacity limits and facilitate flexible behavior,” Ying Fan, Muzhi Wang and their colleagues wrote in their paper. “To investigate the neural mechanisms underlying hierarchical organization in WM, we performed two electroencephalography (EEG) and one magnetoencephalography (MEG) experiments, wherein humans retain in WM a temporal sequence of items, that is, syllables, which are organized into chunks, that is, multisyllabic words.”

Fan, Wang and their colleagues carried out three experiments, during which participants were asked to memorize a series of syllable sequences for brief periods of time. These sequences were organized hierarchically into words and multi-word sequences.

The participants were later asked to recall the hierarchy in which the syllable sequences they were originally presented with were organized. During two of the experiments, the researchers recorded the participants’ neural activity using EEG, while in the third they used MEG.

“We demonstrate that the one-dimensional sequence is represented by two-dimensional neural representational geometry in WM arising from left prefrontal and temporoparietal regions, with separate dimensions encoding item position within a chunk and chunk position in the sequence,” wrote Fan, Wang and their colleagues. “Critically, this two-dimensional geometry is observed consistently in different experimental settings, even during tasks not encouraging hierarchical reorganization in WM and correlates with WM behavior.”

The researchers’ observations suggest that sequences of syllables were encoded as 2D neural representations, with distinct dimensions encoding their local rank (ie, the position of a syllable within a word) and the other their global rank (ie, the position of words within a sequence). Notably, this 2D neural geometry, originating in the prefrontal and temporoparietal cortex, was found to be the same irrespective of stimuli that people were asked to memorize and the specific task they were completing.

“Overall, these findings strongly support that complex sequences are reorganized into factorized multidimensional neural representational geometry in WM, which also speaks to general structure-based organizational principles given WM’s involvement in many cognitive functions,” wrote Fan, Wang and their colleagues.

The findings of this recent study contribute to the understanding of working memory processes, shedding new light on the processes underpinning the hierarchical organization of sequences of information that are temporarily stored in the brain. In the future, they could pave the way for more research examining the 2D geometrical representations of these sequences identified by Fan, Wang and their colleagues, which could lead to new interesting discoveries.

© 2024 Science