Systematic exploration of functional network contribution to metacognition in 4 sensory modalities
Poster No:
2023
Submission Type:
Abstract Submission
Authors:
Katarzyna Hat1, Paola Galdi2, Michał Wierzchoń3, Kristian Sandberg4
Institutions:
1Doctoral School in the Social Sciences, Jagiellonian Univesity, Kraków, Poland, 2University of Edinburgh, Edinburgh, United Kingdom, 3Consciousness Lab, Institute of Psychology, Jagiellonian Univesity, Kraków, Poland, 4Center of Functionally Integrative Neuroscience, Aarhus University, Aarhus, Denmark
First Author:
Co-Author(s):
Kristian Sandberg
Center of Functionally Integrative Neuroscience, Aarhus University
Aarhus, Denmark
Center of Functionally Integrative Neuroscience, Aarhus University
Aarhus, Denmark
Introduction:
Metacognition is the ability to assess one's cognitive processes correctly. It is closely related to and influences the success of various mental processes, e.g. learning, memory and consciousness. However, it is still unclear to which extent metacognition is a domain-general or domain-specific process (Rouault, 2018). Here, we investigate whether functional connectivity at rest carries information about the metacognitive capabilities in 4 sensory modalities (vision, audition, touch and nociception) and how they contribute to the domain-generality debate.
Methods:
Behavioral data: 4 homologous perceptual tasks were administered outside the scanner in 4 sensory modalities. Each task consisted of 2 alternative forced choice tasks followed by confidence judgement. Data was analysed with the Signal Detection Framework using the bhsdtr2 package (Paulewicz, 2020), and a type-2 d' (metacognitive sensitivity - meta-d') was estimated (Maniscalco, 2012).
MRI: 302 participants underwent resting-state fMRI acquisition on a Siemens Skyra 3T (eyes open with fixation cross; TR=801ms, voxel size=2,5mm isotropic, 18min of acquisition). Data were preprocessed with fMRIPrep 21.0.2 and rsDenoise (Kliemann, 2022) following Finn (2015), excluding Global Signal Regression. Denoised data were parcellated into 400 cortical parcels from Schaefer (2018), 17 subcortical parcels defined as in the HCP CIFTI files (Glasser, 2013), and 28 cerebellar parcels using the SUIT atlas (Diedrichsen, 2009). Average time series were extracted from parcels to estimate functional connectivity as pairwise Pearson's correlation.
Predictive framework: Functional connections were used to predict the meta-d' scores (behavioral performance included as a confound) using Elastic Net linear regression and nested cross-validation. We ran 3 kinds of predictions using the 17 Yeo networks plus the cerebellum and subcortical regions as additional networks (19 networks in total). First, a top-down loop starting with all connections (whole brain) and then consecutively removing one of the networks, thus creating 'artificial lesions' (Dubois, 2018) until further removing networks would decrease the model's performance. Second, a bottom-up loop starting with two networks offering the best prediction and then consecutively adding one network until the model performance stops improving. Third, predictions that used only 1 or 2 networks. We measured Pearson's correlation between predicted and actual meta-d' scores. The model's accuracy was estimated with mean squared error. Since 1-2 network predictions tested all possible combinations, we controlled for false discovery rate (FDR) with the Benjamini-Hochberg procedure, setting the significance level at α=0.05. Finally, we performed permutation testing (n=1000) on all top-down and bottom-up final sets, and 1-2 networks significant results.
MRI: 302 participants underwent resting-state fMRI acquisition on a Siemens Skyra 3T (eyes open with fixation cross; TR=801ms, voxel size=2,5mm isotropic, 18min of acquisition). Data were preprocessed with fMRIPrep 21.0.2 and rsDenoise (Kliemann, 2022) following Finn (2015), excluding Global Signal Regression. Denoised data were parcellated into 400 cortical parcels from Schaefer (2018), 17 subcortical parcels defined as in the HCP CIFTI files (Glasser, 2013), and 28 cerebellar parcels using the SUIT atlas (Diedrichsen, 2009). Average time series were extracted from parcels to estimate functional connectivity as pairwise Pearson's correlation.
Predictive framework: Functional connections were used to predict the meta-d' scores (behavioral performance included as a confound) using Elastic Net linear regression and nested cross-validation. We ran 3 kinds of predictions using the 17 Yeo networks plus the cerebellum and subcortical regions as additional networks (19 networks in total). First, a top-down loop starting with all connections (whole brain) and then consecutively removing one of the networks, thus creating 'artificial lesions' (Dubois, 2018) until further removing networks would decrease the model's performance. Second, a bottom-up loop starting with two networks offering the best prediction and then consecutively adding one network until the model performance stops improving. Third, predictions that used only 1 or 2 networks. We measured Pearson's correlation between predicted and actual meta-d' scores. The model's accuracy was estimated with mean squared error. Since 1-2 network predictions tested all possible combinations, we controlled for false discovery rate (FDR) with the Benjamini-Hochberg procedure, setting the significance level at α=0.05. Finally, we performed permutation testing (n=1000) on all top-down and bottom-up final sets, and 1-2 networks significant results.
Results:
We found different sets of networks for all loops. Permutation testing confirmed the significance of results obtained with top-down and bottom-up loops. The top-down sets included 5-9 networks, while the bottom-down sets included 2-3. Importantly, in no case was the bottom-up set a subset of the top-down set (Figure 1). Additional 1-2 network analysis, which partially overcame the shortcomings of the Elastic Net and loops' approach, revealed potential hubs of metacognitive processing for each modality (Figure 2), especially dorsal attention B & central visual networks for vision, ventral attention B for audition and ventral attention A for touch and pain.
·Figure 1
·Figure 2
Conclusions:
We obtained significant predictions of the metacognitive performance with both top-down and bottom-up approaches. This analysis suggests that metacognitive information can be carried via multiple pathways. Although the pattern of results differs between modalities, there are some overlaps worth further discussion and investigation. In particular, the 1-2 network results suggest the ventral attention networks could play the role of hubs in the metacognitive processes.
Higher Cognitive Functions:
Executive Function, Cognitive Control and Decision Making 2
Modeling and Analysis Methods:
Connectivity (eg. functional, effective, structural)
fMRI Connectivity and Network Modeling
Perception, Attention and Motor Behavior:
Consciousness and Awareness 1
Perception: Multisensory and Crossmodal
Keywords:
Cognition
Consciousness
Meta-Cognition
Perception
Other - functional networks
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.
Dubois, J. (2018). A distributed brain network predicts general intelligence from resting-state human neuroimaging data. Philosophical Transactions of the Royal Society B: Biological Sciences, 373(1756).
Finn, E. S. (2015). Functional connectome fingerprinting: identifying individuals using patterns of brain connectivity. Nature Neuroscience, 18(11), 1664–1671.
Glasser, M. F. (2013). The minimal preprocessing pipelines for the Human Connectome Project. NeuroImage, 80, 105–124.
Kliemann, D. (2019). Intrinsic Functional Connectivity of the Brain in Adults with a Single Cerebral Hemisphere. Cell Reports, 29(8), 2398-2407.e4.
Maniscalco, B. (2012). A signal detection theoretic approach for estimating metacognitive sensitivity from confidence ratings. Consciousness and Cognition, 21(1), 422–430.
Paulewicz, B. (2020). The bhsdtr package: A general-purpose method of Bayesian inference for signal detection theory models. Behavior Research Methods, 52(5), 2122–2141.
Rouault, M. (2018). Human metacognition across domains: insights from individual differences and neuroimaging. Personality Neuroscience, 1, e17.
Schaefer, A. (2018). Local-Global Parcellation of the Human Cerebral Cortex from Intrinsic Functional Connectivity MRI. Cerebral Cortex (New York, N.Y. : 1991), 28(9), 3095–3114.