Propagating Cortical Waves Coordinate Sensory Encoding and Memory Retrieval in The Human Brain
Presented During:
Thursday, June 26, 2025: 11:30 AM - 12:45 PM
Brisbane Convention & Exhibition Centre
Room:
P2 (Plaza Level)
Poster No:
827
Submission Type:
Abstract Submission
Authors:
Yifan Yang1, David Leopold2, Jeff Duyn2, Xiao Liu3
Institutions:
1The Pennsylvania State University, State College, PA, 2National Institutes of Health, Bethesda, MD, 3Pennsylvania State University, State College, PA
First Author:
Co-Author(s):
Introduction:
Complex behavior entails a balance between taking in sensory information from the environment and utilizing previously learned internal information. Experiments in mice have shown that the brain alternates between two modes facilitating visual coding and memory processes respectively, with transitions marked by stereotypical neuronal spiking cascades spanning forebrain structures (Liu et al., 2021; Yang et al., 2023). In humans, multi-second global brain activity observed with functional MRI (fMRI) has been described as propagating waves, moving from low-order sensory-motor (SM) regions to high-order default mode network (DMN) regions (Gu et al., 2021; Raut et al., 2021). Utilizing large-scale fMRI datasets, we investigated whether the fMRI waves and neuronal spiking cascades are the expression of homologous brain dynamics, and if so, whether the alternation between enhanced visual encoding and memory function is similarly across the cycle of fMRI waves in humans.
Methods:
We analyzed large-scale datasets to study infra-slow dynamics across species. For mouse spiking activity, we analyzed the Allen Visual Coding dataset (Siegle et al., 2021), comprising ~22,000 neurons recorded from 32 mice. For human studies, we utilized the Human Connectome Project (HCP) 7T dataset with resting-state fMRI data from 184 subjects (Van Essen et al., 2012), and the Natural Scenes Dataset (NSD), which includes task-based fMRI data from 8 subjects, each completing 22,000–30,000 image recognition trials over the course of a year (Allen et al., 2022). Infra-slow spiking cascades and propagating fMRI waves were detected using delay-profile decomposition method (Gu et al., 2021). To quantify brain stimulus encoding, we developed a deep learning model to decode fMRI responses into text descriptions of visual stimuli, which were compared to ground truth image captions (Fig. 2A).
Results:
We found that spontaneous pupil dilations at rest are associated with cascade dynamics in mouse spiking activity (Fig. 1A–1C) whereas infra-slow (<0.1 Hz) propagating waves observed in human fMRI, which traverse from low-order sensory-motor (SM) regions to high-order default mode network (DMN) regions (Fig. 1D–1F) and is accompanied by thalamic progression from posterior/lateral sensorimotor nuclei to anterior/medial limbic nuclei (Fig. 1G), thereby establishing a cross-species link between two types of infra-slow global brain activity.
The trained deep learning model successfully decoded semantic information of visual stimuli based on evoked fMRI responses (Fig. 2D), with accuracy significantly higher than the untrained model (Fig. 2B). Semantic encoding accuracy predicted subsequent recognition performance: accurately encoding an image stimulus at its first appearance led to a higher rate of successful recognition at its second presentation (Fig. 2C).
The SM-to-DMN fMRI waves persisted during visual memory tasks (Fig. 2E). Memory encoding and recall were estimated through participants' behavioral responses (Fig. 2F). We found that both sensory and memory functions were systematically modulated across the fMRI wave cycle: semantic and memory encoding showed similar changes, peaking during SM-activated phase, whereas memory recall showed an opposite modulation, peaking at DMN-activated phase (Fig. 2G). These seconds-scale modulations of sensory and memory processing resembled those observed in mice over the spiking cascade cycles (Fig. 2E).
The trained deep learning model successfully decoded semantic information of visual stimuli based on evoked fMRI responses (Fig. 2D), with accuracy significantly higher than the untrained model (Fig. 2B). Semantic encoding accuracy predicted subsequent recognition performance: accurately encoding an image stimulus at its first appearance led to a higher rate of successful recognition at its second presentation (Fig. 2C).
The SM-to-DMN fMRI waves persisted during visual memory tasks (Fig. 2E). Memory encoding and recall were estimated through participants' behavioral responses (Fig. 2F). We found that both sensory and memory functions were systematically modulated across the fMRI wave cycle: semantic and memory encoding showed similar changes, peaking during SM-activated phase, whereas memory recall showed an opposite modulation, peaking at DMN-activated phase (Fig. 2G). These seconds-scale modulations of sensory and memory processing resembled those observed in mice over the spiking cascade cycles (Fig. 2E).
Conclusions:
Here we showed that infra-slow activity waves propagating across the human cortex are associated with opposing modulation of the encoding and retrieval of information during visual memory tasks, similar to what has been observed in mice previously. The findings suggest a conserved feature of mammalian brain physiology that coordinates exteroceptive sensory sampling with internal mnemonic processes. Additionally, they offer an important new perspective into the nature of behavioral variability during cognitive tasks.
Learning and Memory:
Learning and Memory Other 1
Perception, Attention and Motor Behavior:
Perception: Visual 2
Keywords:
Machine Learning
Memory
MRI
Neuron
Perception
Single unit recording
Vision
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Are you Internal Review Board (IRB) certified? Please note: Failure to have IRB, if applicable will lead to automatic rejection of abstract.
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
For human MRI, what field strength scanner do you use?
Which processing packages did you use for your study?
Provide references using APA citation style.
Gu Y (2021) Brain Activity Fluctuations Propagate as Waves Traversing the Cortical Hierarchy. Cerebral Cortex 31:3986–4005.
Liu X (2021) Single-neuron firing cascades underlie global spontaneous brain events. Proceedings of the National Academy of Sciences 118:e2105395118.
Raut R V (2021) Global waves synchronize the brain’s functional systems with fluctuating arousal. Sci Adv 7:eabf2709.
Siegle JH (2021) Survey of spiking in the mouse visual system reveals functional hierarchy. Nature 592:86–92.
Van Essen DC (2012) The Human Connectome Project: A data acquisition perspective. Neuroimage 62:2222–2231.
Yang Y (2023) Intrinsic forebrain arousal dynamics governs sensory stimulus encoding. bioRxiv.