How Brain Cells Work Together to Remember and Imagine Places
Summary: Researchers have created a computational model that helps explain how mental images drawn from memory can be accounted for by the activity of specific neurons.
Source: UCL
Overview
UCL researchers have developed a computational model that explains how mental images drawn from memory arise from coordinated activity across individual brain cells. The model, published in eLife and supported by Wellcome, the European Research Council and the Human Brain Project, integrates known neural responses with proposed mechanisms to show how experiences are stored and later reassembled as vivid mental imagery.
The work synthesizes evidence about neurons that signal our location and orientation while we move through space—such as place cells that fire depending on where we are, grid cells that provide a metric for space, and head direction cells that signal orientation. The model also proposes additional, as-yet-unconfirmed neuron types that would signal distances and directions to objects in the environment. A central feature of the model is the distinction between egocentric and allocentric representations: sensory input is initially encoded egocentrically (objects located relative to the observer — ahead, left, right), while memory-related brain areas such as the hippocampus store information allocentrically (relative to the surroundings and environment).
The proposed circuit explains how egocentric sensory representations are transformed into allocentric memory representations for storage, and how the process can work in reverse: allocentric memories can be transformed back into egocentric representations to support the imagination of a scene as it would appear from a particular viewpoint.
Model Details and Mechanisms
Lead author Dr. Andrej Bicanski (UCL Institute of Cognitive Neuroscience) explains that recalling an event can be thought of as re-experiencing it in imagination. For example, if you met someone at a train station, the model shows how the brain could encode that encounter — the person’s distance and direction relative to you and the layout of the station — and later reconstruct a mental image of that same scene from the original viewpoint.
The model maps established neural populations (place cells, head-direction cells, grid cells, allocentric boundary- and object-vector cells, and gain-field neurons) into a coherent system that supports spatial memory, scene construction and mental navigation. It predicts interactions across multiple brain regions, via intermediary areas such as the retrosplenial cortex, to convert between viewpoint-dependent and viewpoint-independent representations.
Functional Implications and Predictions
Professor Neil Burgess, the senior author, notes that the model provides a mechanistic account of spatial memory and imagery, covering cognitive concepts such as episodic future thinking and scene construction. Importantly, it also clarifies how different patterns of brain damage could produce distinct impairments in these abilities—for example, explaining specific aspects of amnesia.
The researchers used the model to simulate lesions to particular brain regions, confirming experimental findings from animal studies. For example, damage to the hippocampus impairs the ability to remember where objects were located relative to both the observer and the surrounding environment. Although the authors have not yet modelled a specific neurological disorder in detail, they suggest the framework could serve as a foundation for investigating how different patterns of neural damage may lead to functional deficits in conditions such as Alzheimer’s disease.
Research Context and Abstract Summary
Abstract (summary): A neural-level model of spatial memory and imagery
The authors present a model explaining how egocentric spatial experiences represented in parietal cortex interface with viewpoint-independent representations in medial temporal areas, through retrosplenial cortex, to support key aspects of spatial cognition. The account shows how previously reported neural responses can be integrated in a modular fashion and predicts new cell types (for example, egocentric and head-direction-modulated boundary- and object-vector cells). The model outlines how interactions among these neural populations across multiple regions support spatial memory, scene construction, novelty detection, trace cells, and mental navigation. Simulated behavior and neural activity were compared with experimental data, illustrating how object-vector cells can anchor items within a contextual representation based on environmental boundaries, and how grid cells could update viewpoint during planning and short-cutting by driving sequential place cell activity.
Funding and Publication
Funding: The study was supported by Wellcome, the European Research Council and the Human Brain Project.
Source: Barbara Benham — UCL
Publisher: Organized by NeuroscienceNews.com
Image source: NeuroscienceNews.com image is in the public domain.
Original research: Open access research titled “A neural-level model of spatial memory and imagery” by Andrej Bicanski and Neil Burgess, published in eLife on September 4, 2018. DOI: 10.7554/eLife.33752
MLA: UCL. “How Brain Cells Work Together to Remember and Imagine Places.” NeuroscienceNews. NeuroscienceNews, 7 September 2018.
APA: UCL (2018, September 7). How Brain Cells Work Together to Remember and Imagine Places. NeuroscienceNews. Retrieved September 7, 2018.
Chicago: UCL. “How Brain Cells Work Together to Remember and Imagine Places.” Accessed September 7, 2018.
Feel free to share this summary of the neuroscience research. This article synthesizes the researchers’ explanations and the abstract to describe how neural populations may encode and reconstruct spatial experiences and imagined scenes.