Arrangement of specialized and general control regions in the LPFC evaluated using precision fMRI
Poster No:
940
Submission Type:
Abstract Submission
Authors:
Zach Ladwig1, Nathan Labora2, Megan Dorn1, Joanna Hernandez1, Rodrigo Braga1, Caterina Gratton2
Institutions:
1Northwestern University, Chicago, IL, 2Florida State University, Tallahassee, FL
First Author:
Co-Author(s):
Introduction:
There is debate regarding whether regions of the human lateral prefrontal cortex support specific cognitive functions or respond to many diverse demands (Duncan & Owen 2000). Recent evidence from several single-subject precision fMRI studies suggest that there are LPFC regions which respond to specific domains (e.g., language, theory of mind, episodic projection, auditory processing, visual processing) and distinct but nearby general control regions which respond to many cognitive demands (Fedorenko et al., 2011, DiNicola et al., 2020, Noyce et al., 2017, Assem et al., 2020). However, these sets of regions have been identified in different individuals using different paradigms, so an open question is how these regions are arranged relative to each other within single individuals. To address this question, we deeply sampled a small number of individuals using a comprehensive task-based fMRI battery in order to understand the relative arrangement of specialized and general control regions in the LPFC.
Methods:
A preliminary dataset of two highly sampled subjects (2F) was used for these analyses, each with 13-15 hours of task fMRI. Tasks included sentence processing, episodic projection, theory of mind, and a wide variety of cognitive control tasks targeting working memory, attention, and inhibition. Between 45 and 90 minutes of data were collected per task. Each of these tasks were taken from or adapted directly from the precision fMRI literature. MRI data underwent structural and functional preprocessing including motion correction, distortion correction, registration, normalization, projection to the surface, and spatial smoothing. Contrast maps were generated for each subject using a task-specific, block design GLM. In these two subjects, we completed a split-half analysis for each task and found that with sufficient data (typically 30 minutes per contrast), it was possible to obtain reliable subject-specific activation maps that reliably located fine-scale organizational patterns.
Results:
We found that task contrasts targeting distinct high-level cognitive functions (language, cognitive control, episodic projection, and theory of mind) activated distributed sets of distinct but nearby regions throughout the brain and including the lateral prefrontal cortex. That is, language task contrasts activated a distributed set of regions different from those activated during theory of mind task contrasts, etc. This was shown first by using split-half analysis to define thresholded task-specific ROIs in one half of the data and quantifying relative activation to various cognitive demands in the second half (Figure 1). This was also shown by visualizing the overlap (or lack thereof) of thresholded contrast maps on the surface (Figure 2). In both subjects, we saw separation between cognitive domains in the LPFC as well as other brain regions. In addition, we found in both subjects that there were "general control" regions activated by many cognitive control contrasts that were also non-overlapping with but nearby to the domain-specialized regions. In several cases, regions with distinct functional profiles lay side-by-side in an interdigitated fashion. There was one exception - language and auditory processing demands activated overlapping regions in both subjects.
·Figure 1: Quantitative separation of high level functions in the LPFC
·Figure 2: Visualization of thresholded contrast maps in the LPFC shows high level separation and interdigitation.
Conclusions:
These results support the view that high-level functions are supported by distinct sets of distributed regions which are interdigitated with one another throughout the cortex, rather than by large swaths of multifunctional cortex or slowly changing cortical gradients. Further, they support the idea that there are regions which respond to diverse cognitive control demands and these too are distributed throughout the cortex and interdigitated with specialized regions. We plan to extend this work to include more subjects, more cognitive domains, and the consideration of large-scale functional networks defined by functional connectivity.
Emotion, Motivation and Social Neuroscience:
Social Cognition
Higher Cognitive Functions:
Executive Function, Cognitive Control and Decision Making 1
Imagery
Language:
Language Comprehension and Semantics
Modeling and Analysis Methods:
Activation (eg. BOLD task-fMRI) 2
Keywords:
ADULTS
Cognition
Cortex
FUNCTIONAL MRI
Language
NORMAL HUMAN
1|2Indicates the priority used for review
Provide references using author date format
2. Fedorenko, E., Behr, M. K. & Kanwisher, N. (2011) Functional specificity for high-level linguistic processing in the human brain. Proc. Natl. Acad. Sci. U. S. A. 108, 16428–16433
3. DiNicola, L. M., Braga, R. M. & Buckner, R. L. (2020) Parallel distributed networks dissociate episodic and social functions within the individual. J. Neurophysiol. 123, 1144–1179
4. Noyce, A. L., Cestero, N., Michalka, S. W., Shinn-Cunningham, B. G. & Somers, D. C. (2017) Sensory-Biased and Multiple-Demand Processing in Human Lateral Frontal Cortex. J. Neurosci. 37, 8755–8766
5. Assem, M., Glasser, M. F., Van Essen, D. C. & Duncan, J. A (2020) Domain-General Cognitive Core Defined in Multimodally Parcellated Human Cortex. Cereb. Cortex 30, 4361–4380