Cognitive science has long been aided by imaging tools—such as functional MRI and PET scans—that can reveal correlations between brain activity in particular regions and certain tasks or behaviors. These imaging tools contributed to a breakthrough in the understanding of cognitive function in the mid- to late-20th century when researchers began to consider the way that various regions in the cortex might interact with one another as a network to support high-order functions. One such set of regions—the “default network”—was consistently identified based on the way that it tended to increase its activity during off-task portions of attention-demanding tasks (Schulman et al., 1997). Neuroimaging studies have also implicated these brain regions in several other activities, including decision-making, imagination, scene construction, theory of mind, and more (Andrews-Hanna, 2011). However, in more recent years, high-resolution and individual-based studies have illuminated these regions and revealed more complexity. Through the undergoing iCere study—one of the largest and most comprehensive of its kind—at the Harvard Center for Brain Science, the Buckner Lab and its collaborators hope to use these new approaches to demystify the organization and function of the default network and other networks in the brain.
Over the past decade, researchers have found that certain regions traditionally considered part of the default network were recruited differently under varying conditions. For example, particular regions have been differentially associated with the “self” and self-referential processing (Davey et al., 2016; Qin & Northoff, 2011), emotional mentation (Andrews-Hanna, 2011), social cognition (Spreng & Andrews-Hanna, 2015), and more. Furthermore, altered activation in different regions of the default network has been explored as a potential mechanism in various disorders including autism spectrum disorder (Kennedy & Courchesne, 2008) and depression (Sheline et al., 2009). The variability in the activation of regions within the network led Andrews Hanna and colleagues in the Buckner Lab to argue in their 2010 paper that the “default network” is actually comprised of multiple subsystems that converge on “hub regions” but are functionally dissociable. That is, different subnetworks serve different functions but are linked through their common connection to a set of critical midline regions.
Recent work in the Buckner Lab at Harvard’s Center for Brain Science has challenged this former hypothesis, though, and introduced a new one: that the default network (among several others) is actually comprised of multiple fully distinct networks that have now been linked to unique functions. A 2017 study by Braga and Buckner using detailed neuroimaging in individuals identified two completely separable networks within the canonical default network bounds. Rather than utilizing group-averaging with scans from many subjects, this study involved repetitive scanning and analysis of individuals, allowing for high-precision network identification in each subject. Importantly, the two networks were found to parallel one another, each “possessing adjacent regions in eight or more cortical zones” and exhibiting spatial variability across individuals—strong evidence to suggest that they may have been blurred together in previous group-averaged studies. Other well-studied networks (frontoparietal network; dorsal attention network) displayed similar patterns of potential differentiation in intrinsic functional connectivity and may indicate the need for a shift in the way researchers approach network analyses.
One of the parallel distributed networks identified within the default network, dubbed default network A (DN-A), showed correlations with the parahippocampal cortex and has since been implicated in episodic projection (i.e., remembering or imagining events). (DiNicola et al., 2020). The other network, dubbed default network B (DN-B), was not coupled with the parahippocampal cortex, but did include the temporoparietal junction and was preferentially recruited for tasks involving various kinds of theory of mind (ToM) (DiNicola et al., 2020). DN-B was found to resemble the regions that had been previously linked to many forms of social inference (see Lieberman et al. 2007, Schulz et al. 2020 for reviews), and studies have long implicated component regions such as the temporoparietal junction in identifying and interpreting specific emotions and emotional situations (Skerry & Saxe, 2015). Despite these findings, the precise boundary conditions of recruitment for several of these and several other networks remains unclear.
For this reason, the iCere project—now one year underway—is working to collect and analyze brain data from dozens of individuals to study the organization of individually-defined networks in the cerebral cortex and cerebellum, as well as cognitive processes like episodic projection, social cognition, memory, and more. Researchers are making progress in refining our understanding of somatomotor representation, mapping the cerebellum, and exploring subcortical structures.
One initiative involving the iCere data seeks to enhance our analysis of higher-order cognitive networks by combining imaging data with independent behavioral measures. As part of the iCere series of scans, the Buckner Lab is acquiring data from repeatedly sampled individuals who answer questions while in the scanner (as in DiNicola et al., 2020). Responsivity of networks like DN-A and DN-B (identified individually in each subject) can then be quantified as in prior work. To gain more information about why responsivity may differ across questions, properties of the questions and the strategies used to answer them are also being quantified using online behavioral questionnaires, an approach based on Andrews-Hanna and colleagues (2010). In other words, independent data coming from online participants about the thought processes they use when answering questions will help researchers assess the observed responses in the brain data from scan participants. The hope is that this exploration will provide insight into exactly the kinds of cognitive operations that these newly parcellated networks support.
These new studies, of course, build on years of theory and research at the Center for Brain Science and beyond that have served as a valuable framework. For this project, open questions from previous, similar analyses at the Buckner Lab inspired the development of expanded question and behavioral data sets, which will include both new online data and imaging data from additional repeatedly scanned individuals. Significant portions of this new exploration are driven by interdisciplinary inquiry from undergraduate members of the Buckner Lab who are investigating the relationship between these higher-order brain networks (in particular, DN-A and DN-B) and concepts like morality, self-consciousness, and theory of mind. Inspired by philosophical discourse and literature, these questions and the study results will be approached from many angles to develop new hypotheses and interpretations. The large amount of data being collected will provide ample opportunity to analyze these connections—and then to replicate and potentially triplicate findings in different groups. In this sense, the iCere project has the potential to be one of the most powerful tools currently available for studying the mind. In time, it may help us uncover just how our brains make us who we are.
Many thanks to Lauren DiNicola and Randy Buckner for their guidance.
About the Author
Samuel Murdock is a junior at Harvard College concentrating in Neuroscience and Philosophy.
References
Andrews-Hanna, J. R. (2011). The Brain’s default network and its adaptive role in internal mentation. The Neuroscientist, 18(3), 251–270. https://doi.org/10.1177/1073858411403316
Andrews-Hanna, J. R., Reidler, J. S., Sepulcre, J., Poulin, R., & Buckner, R. L. (2010). Functional-anatomic fractionation of the brain’s default network. Neuron, 65(4), 550–562. https://doi.org/10.1016/j.neuron.2010.02.005
Braga, R. M., & Buckner, R. L. (2017). Parallel interdigitated distributed networks within the individual estimated by intrinsic functional connectivity. Neuron, 95(2). https://doi.org/10.1016/j.neuron.2017.06.038
Davey, C. G., Pujol, J., & Harrison, B. J. (2016). Mapping the self in the brain’s Default Mode Network. NeuroImage, 132, 390–397. https://doi.org/10.1016/j.neuroimage.2016.02.022
DiNicola, L. M., Braga, R. M., & Buckner, R. L. (2020). Parallel distributed networks dissociate episodic and social functions within the individual. Journal of Neurophysiology, 123(3), 1144–1179. https://doi.org/10.1152/jn.00529.2019
Kennedy, D. P., & Courchesne, E. (2008). Functional abnormalities of the default network during self- and other-reflection in autism. Social Cognitive and Affective Neuroscience, 3(2), 177–190. https://doi.org/10.1093/scan/nsn011
Qin, P., & Northoff, G. (2011). How is our self related to midline regions and the default-mode network? NeuroImage, 57(3), 1221–1233. https://doi.org/10.1016/j.neuroimage.2011.05.028
Sheline, Y. I., Barch, D. M., Price, J. L., Rundle, M. M., Vaishnavi, S. N., Snyder, A. Z., Mintun, M. A., Wang, S., Coalson, R. S., & Raichle, M. E. (2009). The default mode network and self-referential processes in depression. Proceedings of the National Academy of Sciences, 106(6), 1942–1947. https://doi.org/10.1073/pnas.0812686106
Skerry, A. E., & Saxe, R. (2015). Neural representations of emotion are organized around abstract event features. Current Biology, 25(15), 1945–1954. https://doi.org/10.1016/j.cub.2015.06.009
Spreng, R. N., & Andrews-Hanna, J. R. (2015). The default network and social cognition. Brain Mapping, 165–169. https://doi.org/10.1016/b978-0-12-397025-1.00173-1
Over the past decade, researchers have found that certain regions traditionally considered part of the default network were recruited differently under varying conditions. For example, particular regions have been differentially associated with the “self” and self-referential processing (Davey et al., 2016; Qin & Northoff, 2011), emotional mentation (Andrews-Hanna, 2011), social cognition (Spreng & Andrews-Hanna, 2015), and more. Furthermore, altered activation in different regions of the default network has been explored as a potential mechanism in various disorders including autism spectrum disorder (Kennedy & Courchesne, 2008) and depression (Sheline et al., 2009). The variability in the activation of regions within the network led Andrews Hanna and colleagues in the Buckner Lab to argue in their 2010 paper that the “default network” is actually comprised of multiple subsystems that converge on “hub regions” but are functionally dissociable. That is, different subnetworks serve different functions but are linked through their common connection to a set of critical midline regions.
Recent work in the Buckner Lab at Harvard’s Center for Brain Science has challenged this former hypothesis, though, and introduced a new one: that the default network (among several others) is actually comprised of multiple fully distinct networks that have now been linked to unique functions. A 2017 study by Braga and Buckner using detailed neuroimaging in individuals identified two completely separable networks within the canonical default network bounds. Rather than utilizing group-averaging with scans from many subjects, this study involved repetitive scanning and analysis of individuals, allowing for high-precision network identification in each subject. Importantly, the two networks were found to parallel one another, each “possessing adjacent regions in eight or more cortical zones” and exhibiting spatial variability across individuals—strong evidence to suggest that they may have been blurred together in previous group-averaged studies. Other well-studied networks (frontoparietal network; dorsal attention network) displayed similar patterns of potential differentiation in intrinsic functional connectivity and may indicate the need for a shift in the way researchers approach network analyses.
One of the parallel distributed networks identified within the default network, dubbed default network A (DN-A), showed correlations with the parahippocampal cortex and has since been implicated in episodic projection (i.e., remembering or imagining events). (DiNicola et al., 2020). The other network, dubbed default network B (DN-B), was not coupled with the parahippocampal cortex, but did include the temporoparietal junction and was preferentially recruited for tasks involving various kinds of theory of mind (ToM) (DiNicola et al., 2020). DN-B was found to resemble the regions that had been previously linked to many forms of social inference (see Lieberman et al. 2007, Schulz et al. 2020 for reviews), and studies have long implicated component regions such as the temporoparietal junction in identifying and interpreting specific emotions and emotional situations (Skerry & Saxe, 2015). Despite these findings, the precise boundary conditions of recruitment for several of these and several other networks remains unclear.
For this reason, the iCere project—now one year underway—is working to collect and analyze brain data from dozens of individuals to study the organization of individually-defined networks in the cerebral cortex and cerebellum, as well as cognitive processes like episodic projection, social cognition, memory, and more. Researchers are making progress in refining our understanding of somatomotor representation, mapping the cerebellum, and exploring subcortical structures.
One initiative involving the iCere data seeks to enhance our analysis of higher-order cognitive networks by combining imaging data with independent behavioral measures. As part of the iCere series of scans, the Buckner Lab is acquiring data from repeatedly sampled individuals who answer questions while in the scanner (as in DiNicola et al., 2020). Responsivity of networks like DN-A and DN-B (identified individually in each subject) can then be quantified as in prior work. To gain more information about why responsivity may differ across questions, properties of the questions and the strategies used to answer them are also being quantified using online behavioral questionnaires, an approach based on Andrews-Hanna and colleagues (2010). In other words, independent data coming from online participants about the thought processes they use when answering questions will help researchers assess the observed responses in the brain data from scan participants. The hope is that this exploration will provide insight into exactly the kinds of cognitive operations that these newly parcellated networks support.
These new studies, of course, build on years of theory and research at the Center for Brain Science and beyond that have served as a valuable framework. For this project, open questions from previous, similar analyses at the Buckner Lab inspired the development of expanded question and behavioral data sets, which will include both new online data and imaging data from additional repeatedly scanned individuals. Significant portions of this new exploration are driven by interdisciplinary inquiry from undergraduate members of the Buckner Lab who are investigating the relationship between these higher-order brain networks (in particular, DN-A and DN-B) and concepts like morality, self-consciousness, and theory of mind. Inspired by philosophical discourse and literature, these questions and the study results will be approached from many angles to develop new hypotheses and interpretations. The large amount of data being collected will provide ample opportunity to analyze these connections—and then to replicate and potentially triplicate findings in different groups. In this sense, the iCere project has the potential to be one of the most powerful tools currently available for studying the mind. In time, it may help us uncover just how our brains make us who we are.
Many thanks to Lauren DiNicola and Randy Buckner for their guidance.
About the Author
Samuel Murdock is a junior at Harvard College concentrating in Neuroscience and Philosophy.
References
Andrews-Hanna, J. R. (2011). The Brain’s default network and its adaptive role in internal mentation. The Neuroscientist, 18(3), 251–270. https://doi.org/10.1177/1073858411403316
Andrews-Hanna, J. R., Reidler, J. S., Sepulcre, J., Poulin, R., & Buckner, R. L. (2010). Functional-anatomic fractionation of the brain’s default network. Neuron, 65(4), 550–562. https://doi.org/10.1016/j.neuron.2010.02.005
Braga, R. M., & Buckner, R. L. (2017). Parallel interdigitated distributed networks within the individual estimated by intrinsic functional connectivity. Neuron, 95(2). https://doi.org/10.1016/j.neuron.2017.06.038
Davey, C. G., Pujol, J., & Harrison, B. J. (2016). Mapping the self in the brain’s Default Mode Network. NeuroImage, 132, 390–397. https://doi.org/10.1016/j.neuroimage.2016.02.022
DiNicola, L. M., Braga, R. M., & Buckner, R. L. (2020). Parallel distributed networks dissociate episodic and social functions within the individual. Journal of Neurophysiology, 123(3), 1144–1179. https://doi.org/10.1152/jn.00529.2019
Kennedy, D. P., & Courchesne, E. (2008). Functional abnormalities of the default network during self- and other-reflection in autism. Social Cognitive and Affective Neuroscience, 3(2), 177–190. https://doi.org/10.1093/scan/nsn011
Qin, P., & Northoff, G. (2011). How is our self related to midline regions and the default-mode network? NeuroImage, 57(3), 1221–1233. https://doi.org/10.1016/j.neuroimage.2011.05.028
Sheline, Y. I., Barch, D. M., Price, J. L., Rundle, M. M., Vaishnavi, S. N., Snyder, A. Z., Mintun, M. A., Wang, S., Coalson, R. S., & Raichle, M. E. (2009). The default mode network and self-referential processes in depression. Proceedings of the National Academy of Sciences, 106(6), 1942–1947. https://doi.org/10.1073/pnas.0812686106
Skerry, A. E., & Saxe, R. (2015). Neural representations of emotion are organized around abstract event features. Current Biology, 25(15), 1945–1954. https://doi.org/10.1016/j.cub.2015.06.009
Spreng, R. N., & Andrews-Hanna, J. R. (2015). The default network and social cognition. Brain Mapping, 165–169. https://doi.org/10.1016/b978-0-12-397025-1.00173-1
