Summary: A new study shows that the human brain organizes action-outcome associations in cognitive maps, much like how it maps physical space for navigation. These action maps are formed in the hippocampus and communicate with the motor system to support goal-directed behaviors and flexible decision-making.
Using virtual reality tasks and fMRI, researchers observed that the brain represents similar actions closer together in this map-like structure, influencing how we perceive and choose between them. This challenges traditional views of memory systems, revealing that action planning is guided by interconnected neural networks that link memory, perception, and motor control.
Key Facts:
- Cognitive Action Maps: The brain organizes actions and their outcomes using hippocampal maps.
- Motor Integration: These maps interact with the motor system to guide behavior and decision-making.
- Beyond Space: The study expands the role of cognitive maps from spatial navigation to action planning.
Source: Max Planck Institute
How are the relations between different action plans organized and structured in the brain to support our rich behavioural repertoire?
Irina Barnaveli and Christian Doeller together with Simone Viganò and Daniel Reznik from the Max Planck Institute for Human Cognitive and Brain Sciences and with Patrick Haggard from the Institute of Cognitive Neuroscience, University College London, argue in their recent study that the brain organizes action-outcome associations in a cognitive map-like structure.
The researchers further demonstrate in their study, published in Nature Communications, that these cognitive maps, located in the hippocampal system, communicate with the motor system during action evaluation, suggesting that goal-directed action planning skills rely on multiple neural systems.
The human capacity to develop a diverse and highly complex repertoire of action plans is truly remarkable.
Many of our behaviours are rooted in associations between actions and their outcomes, which we form and leverage flexibly. For instance, the same keypress action can lead to different outcomes depending on whether it’s performed on a computer keyboard, a radio, or another context.
We often have multiple alternative choices, so that action selection requires comparing across available action-outcomes, which is a non-trivial and taxing problem.
“How do we compare the many action plans stored in memory and select the most appropriate one?
“We propose that these action-outcome associations could be organized in a cognitive map within the hippocampal system, potentially supporting efficient action selection within the rich human behavioural repertoire” explains Irina Barnaveli, first author of the study.
“The hippocampus is a part of the brain involved in forming memories and navigating in space. Navigation relies heavily on the creation of maps of space. Our study suggests that we create similar maps for organizing and selecting action plans, linking perception to action.”
In the immersive virtual reality, participants performed a motor interaction task, in which they learned how to control the flying and catching of a virtual ball using different actions. Later, they compared learned actions while their brain activity was monitored using functional magnetic resonance imaging (fMRI).
The scientists found patterns of brain activity that are typically interpreted as signatures of ‘cognitive mapping’, suggesting that the brain abstracts and organizes action plans in a map-like representation.
This map is also evident from participants’ behaviour: the closer the actions are within the hypothesised map, the more similar they are perceived by the participants.
Crucially, this cognitive map in the hippocampal system exchanges information with the motor system to relate multiple action plans to each other.
“The map-like representations might therefore represent how humans interact with their environment in a very general sense, well beyond the specific case of spatial navigation. By supporting action selection, cognitive maps could contribute to optimizing the acquisition and exploitation of a wide repertoire of action plans.
“This discovery further challenges the classical distinction between declarative and procedural memory, suggesting that goal-directed actions rely on multiple neural systems that integrate action generation, motor planning, and memory”, concludes Christian Doeller, last author of the study.
About this neuroscience research news
Author: Bettina Hennebach
Source: Max Planck Institute
Contact: Bettina Hennebach – Max Planck Institute
Image: The image is credited to Neuroscience News
Original Research: Open access.
“Hippocampal-entorhinal cognitive maps and cortical motor system represent action plans and their outcomes” by Irina Barnaveli et al. Nature Communications
Abstract
Hippocampal-entorhinal cognitive maps and cortical motor system represent action plans and their outcomes
Efficiently interacting with the environment requires weighing and selecting among multiple alternative actions based on their associated outcomes. However, the neural mechanisms underlying these processes are still debated.
We show that forming relations between arbitrary action-outcome associations involve building a cognitive map.
Using an immersive virtual reality paradigm, participants learned 2D abstract motor action-outcome associations and later compared action combinations while their brain activity was monitored with fMRI.
We observe a hexadirectional modulation of the activity in entorhinal cortex while participants compared different action plans.
Furthermore, hippocampal activity scales with the 2D similarity between outcomes of these action plans. Conversely, the supplementary motor area represents individual actions, showing a stronger response to overlapping action plans.
Crucially, the connectivity between hippocampus and supplementary motor area is modulated by the similarity between the action plans, suggesting their complementary roles in action evaluation.
These findings provide evidence for the role of cognitive maps in action selection, challenging classical models of memory taxonomy and its neural bases.
