Whole-Brain Dynamics of Compositional Cognition: Domain-Specific and Domain-General Interactions
Poster No:
741
Submission Type:
Abstract Submission
Authors:
Joshua Tan1, Rebekah Wong1, Isabella Orlando2, Jungwoo Kim3, Eli Müller4, Claire O'Callaghan2, James Shine2
Institutions:
1University of Sydney, Camperdown, NSW, 2The University of Sydney, Sydney, NSW, 3SKKU, Suwon, Gyeonggi-do, 4University of Sydney, Sydney, NSW
First Author:
Co-Author(s):
Introduction:
Compositional cognition refers to the recombination of prior knowledge to generate new thoughts and behaviours (Fusi et al., 2016). This process involves context-dependent (domain-specific) and context-independent (domain-general) components that allow the application of learned skills to novel situations. Previous studies have examined compositionality from a static, cortical perspective, linking the process to large-scale cognitive control networks (Figure 1a; Cole et al., 2013). However, the dynamic interactions across the entire brain, including subcortical structures, remain poorly understood. Here, we explore how different brain regions contribute to domain-specific and domain-general processes and how their time-series dynamics interact to facilitate compositional cognition (Figure 1b). This investigation provides insight into the neural mechanisms underlying complex cognitive behaviours.
Methods:
Eighty-seven right-handed participants (mean age = 24 years, SD = 4 years, 51 female) completed the Concreted Permuted Rules of Operations (C-PRO) cognitive paradigm while undergoing simultaneous fMRI scanning (Ito et al., 2017). The task involved twelve rules from three domains (logic, sensory, motor), with one rule from each domain forming a set (64 possible combinations). Each set was performed in a mini-block of three consecutive trials (ITI = 1570 ms). Participants completed eight runs of 7 min 36 s each in a 3T MRI scanner after 30 minutes of training. Data were preprocessed using fMRIprep, denoising, and a high-pass filter (0.01). Time-series data were extracted and z-scored for 482 regions (400 cortical, Schaefer et al., 2018; 54 subcortical, Tian et al., 2020; 28 cerebellar, Diedrichsen, 2006). Task-evoked BOLD activity was modelled using a generalized linear-mixed model, with separate regressors for each mini-block. Changes specific to each domain were modelled with group-level GLMs resulting in a three-dimensional space where each axis maps to a specific domain. Regions were grouped as domain-specific and domain-general depending on their coordinates in this space. The original BOLD time-series was projected onto this coordinate space and time-series dynamics were analysed to compare domain-specific versus domain-general recruitment. Graph theory metrics assessed regional communication across the brain.
Results:
A sparsely distributed network of regions was recruited across all mini-blocks (domain-general). This included cortical regions from the ventrolateral prefrontal cortex, superior and inferior parietal lobe, and superior temporal lobe. Crus I, putamen, and the thalamus were also recruited (pfdr < 0.05). This group of regions differed from regions that were recruited for specific domains. Domain-general regions demonstrated signs of integration indicated by a higher participation coefficient, whereas domain-specific regions were segregated (mean difference = 3.22, p = 0.02)
Time-series dynamics were analysed by projecting original BOLD time-series of domain-specific and domain-general regions onto the three-dimensional space. Over time, domain-specific regions moved further along the axis of their bias, with motor regions moving later than the other two domains. Domain-general regions were consistently in-between all three axis, showing no bias to any axis across the duration of the trial.
Time-series dynamics were analysed by projecting original BOLD time-series of domain-specific and domain-general regions onto the three-dimensional space. Over time, domain-specific regions moved further along the axis of their bias, with motor regions moving later than the other two domains. Domain-general regions were consistently in-between all three axis, showing no bias to any axis across the duration of the trial.
Conclusions:
From a whole-brain approach, we identified a distributed set of regions that span the cortex, subcortex and cerebellum that are recruited for compositional cognition. We showed that compositional cognition can be broken down into domain-general and domain-specific regions, and that these regions were separate to one another. Finally, we also showed that these two classes of regions have different temporal signatures and communicate with the rest of the brain in different manners. Therefore, interactions throughout the whole brain are important in generating new thoughts and behaviours.
Higher Cognitive Functions:
Executive Function, Cognitive Control and Decision Making 1
Modeling and Analysis Methods:
Activation (eg. BOLD task-fMRI) 2
fMRI Connectivity and Network Modeling
Keywords:
Basal Ganglia
Cerebellum
Cognition
fMRI CONTRAST MECHANISMS
Sub-Cortical
Systems
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?
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.
Diedrichsen, J. (2006). A spatially unbiased atlas template of the human cerebellum. NeuroImage, 33(1), 127–138. https://doi.org/10.1016/j.neuroimage.2006.05.056
Fusi, S., Miller, E. K., & Rigotti, M. (2016). Why neurons mix: High dimensionality for higher cognition. Current Opinion in Neurobiology, 37, 66–74. https://doi.org/10.1016/j.conb.2016.01.010
Ito, T., Kulkarni, K. R., Schultz, D. H., Mill, R. D., Chen, R. H., Solomyak, L. I., & Cole, M. W. (2017). Cognitive task information is transferred between brain regions via resting-state network topology. Nature Communications, 8(1), 1027. https://doi.org/10.1038/s41467-017-01000-w
Schaefer, A., Kong, R., Gordon, E. M., Laumann, T. O., Zuo, X.-N., Holmes, A. J., Eickhoff, S. B., & Yeo, B. T. T. (2018). Local-Global Parcellation of the Human Cerebral Cortex from Intrinsic Functional Connectivity MRI. Cerebral Cortex, 28(9), 3095–3114. https://doi.org/10.1093/cercor/bhx179
Tian, Y., Margulies, D. S., Breakspear, M., & Zalesky, A. (2020). Topographic organization of the human subcortex unveiled with functional connectivity gradients. Nature Neuroscience, 23(11), 1421–1432. https://doi.org/10.1038/s41593-020-00711-6