Resting state functional connectivity and cognitive task-related activation of the human claustrum
Introduction
In its mediolateral dimension, the claustrum is thin (submillimeter at certain points), but its rostrocaudal and dorsoventral dimensions are roughly equivalent to that of the striatum. Decades of tract tracing studies in rodents and non-human primates indicates that the claustrum is bidirectionally connected with many cortical areas (Edelstein and Denaro, 2004; Crick and Koch, 2005; Goll et al., 2015; Riche and Lanoir, 1978; Mathur et al., 2009; Mathur, 2014; Reser et al., 2017; Sherk, 1986; Park et al., 2012; White et al., 2017; Wang et al., 2017) and is estimated by volume to be among the most highly connected structures in the brain (Torgerson et al., 2015). These observations have fueled several hypotheses that the claustrum: 1) binds multimodal sensory information for the generation of conscious perception (Crick and Koch, 2005; Koubeissi et al., 2014); 2) coordinates somatosensory and motor cortical information (Smith et al., 2012) and; 3) acts as a cortico-cortical relay center supporting attention (Mathur, 2014; Goll et al., 2015).
While the exact function of the claustrum is still largely theoretical (Van Horn, 2019), recent work suggests a role in cognition. Comprehensive analyses of mice on how the cortical mantle connects with the claustrum demonstrate that the claustrum weakly innervates primary sensorimotor cortices, while heavily innervating frontal cortices including the anterior cingulate cortex (ACC) and the prelimbic area of the medial prefrontal cortex (PFC) (White et al., 2017). A strong input from the ACC to the claustrum also exists in rats (Smith and Alloway, 2010; White et al., 2017) and mice (Qadir et al., 2018), and this input encodes a top-down preparatory signal that is proportional to task difficulty (White et al., 2018). These findings suggest that the claustrum may subserve frontal cortical function, including top-down executive processes (Atlan et al., 2018; Goll et al., 2015; Mathur, 2014; White and Mathur, 2018). While some neuroimaging studies have been performed on the claustrum (Krimmel et al., 2019; Smith et al., 2017), evidence for a role of the human claustrum supporting any of the aforementioned functional hypotheses, including cognitive processing, is particularly lacking.
While the anatomical boundaries of the human claustrum (outside of the ventral subdivision) can be resolved with relative ease using high-resolution structural magnetic resonance imaging (MRI), functionally resolving this structure for analysis of blood-oxygenation level-dependent (BOLD) signal with functional MRI (fMRI) is challenging, as signal extracted from the claustrum is heavily mixed with the signal from the neighboring insula and putamen using standard methods. Analysis of BOLD data using standard methods results in similar patterns of functional connectivity (correlation of signal between regions) when comparing claustrum, insula, and putamen. This contrasts with data from multiple tract tracing studies, which instead show unique patterns of anatomical connectivity across these regions (Nakashima et al., 2000; Mathur et al., 2009; Pan et al., 2010; Sato et al., 2013; Wang et al., 2017). Thus, standard fMRI analyses are not capable of functionally resolving the claustrum and may yield inaccurate functional connectivity and activation results.
In an effort to elucidate claustrum function, the current study has three goals: 1) devise new fMRI methodology called Small Region Confound Correction (SRCC) to functionally distinguish the claustrum from the insula and putamen, creating a corrected claustrum timeseries; 2) perform resting state functional connectivity analyses with the corrected claustrum timeseries to reveal functional coupling of the claustrum in humans; and 3) test the hypothesis that the claustrum, owing to strong connectivity with frontal cortices and recent data in mice suggesting its involvement in top-down cognitive processing (White et al., 2018), is activated during a cognitive conflict task. Our data indicate that extraction of unique claustrum signal using fMRI is possible and that the claustrum displays expansive functional connectivity that overlaps with regions involved in cognitive control. During task performance, we found the claustrum to be active at the onset of – or switch to – cognitive conflict task engagement.
Section snippets
Overview
Three datasets were analyzed. The first dataset, 7T-Rest, is publicly available and consisted of scans from 20 healthy humans scanned with 7 Tesla (T) MRI (Gorgolewski et al., 2015). The second dataset, 3T-Rest, consisted of scans from 35 healthy humans acquired with 3T MRI. The third dataset, 3T-Task, used the same subjects as 3T-Rest and consisted of scans from 33 healthy humans performing a cognitive interference task acquired with 3T MRI. In 7T-Rest, we performed seed-based whole-brain
Resting state connectivity of the claustrum at 7T: methodological approach
We used high spatial resolution fMRI data in 7T-Rest and hand drawn ROIs of claustrum, insula, and putamen to analyze whole brain functional connectivity of these ROIs. Despite excellent resolution, we found that functional connectivity between the claustrum and the insula/putamen in this standard analysis was high (average FC of L/R claustrum with insula/putamen ranged from 0.32 to 0.5; Fig. 1D). This contrasts with known unique connectivity of the claustrum relative to insula and putamen (
Discussion
In this study, we provide a novel approach we term Small Region Confound Correction to detect the activity and functional connectivity of the human claustrum. In doing so, we find that the claustrum is strongly functionally connected to cingulate and prefrontal cortices at rest and that there is considerable overlap between claustrum connectivity maps and cognitive task-related networks. Supporting a role of the claustrum in cognition and task switching specifically, we show that the claustrum
Conflicts of interest
No conflicts of interest to disclose.
Author contributions
DAS and BNM conceived of the research. DAS, SK, MGW, and BNM designed research. MP generated claustrum masks. DAS, SK, FSB, MGW and BNM analyzed data. DAS, SK, and BNM wrote the manuscript.
Acknowledgments
This work was supported by National Institute on Alcohol Abuse and Alcoholism grants K22AA021414 and R01AA024845 (B.N.M.), National Institute of General Medical Sciences grant T32008181 (M.G.W.), National Institute of Neurological Disorders and Stroke grant T32NS063391 (M.G.W.), National Institute on Drug Abuse grant R03DA042336 (F.S.B.), and National Center for Complementary and Integrative Health grant R01AT007176 (D.A.S.); The authors also are grateful for the assistance of Dr. Rao
References (50)
- et al.
A component based noise correction method (CompCor) for BOLD and perfusion based fMRI
Neuroimage
(2007) - et al.
A core system for the implementation of task sets
Neuron
(2006) - et al.
Attention: the claustrum
Trends Neurosci.
(2015) - et al.
Electrical stimulation of a small brain area reversibly disrupts consciousness
Epilepsy Behav.
(2014) Towards a network theory of cognition
Neural Network.
(2000)- et al.
Towards a consensus regarding global signal regression for resting state functional connectivity MRI
Neuroimage
(2017) - et al.
Reduction of motion-related artifacts in resting state fMRI using aCompCor
Neuroimage
(2014) - et al.
An anterograde and retrograde tract-tracing study on the projections from the thalamic gustatory area in the rat: distribution of neurons projecting to the insular cortex and amygdaloid complex
Neurosci. Res.
(2000) - et al.
A wavelet method for modeling and despiking motion artifacts from resting-state fMRI time series
Neuroimage
(2014) - et al.
Sources and implications of whole-brain fMRI signals in humans
Neuroimage
(2017)
Projections from the insular cortex to pain-receptive trigeminal caudal subnucleus (medullary dorsal horn) and other lower brainstem areas in rats
Neuroscience
Resting-state fMRI in the human connectome project
Neuroimage
Anterior cingulate cortex input to the claustrum is required for top-down action control
Cell Rep.
Cluster-extent based thresholding in fMRI analyses: pitfalls and recommendations
Neuroimage
The claustrum supports resilience to distraction
Curr. Biol.
The gratton effect remains after controlling for contingencies and stimulus repetitions
Front. Psychol.
The multi-source interference task: validation study with fMRI in individual subjects
Mol. Psychiatr.
What is the function of the claustrum?
Philos. Trans. R. Soc. Lond. Ser. B Biol. Sci.
Enhancing the detection of BOLD signal in fMRI by reducing the partial volume effect
Comput. Math. Method. Med.
When structure affects function–the need for partial volume effect correction in functional and resting state magnetic resonance imaging studies
PLoS One
The claustrum: a historical review of its anatomy, physiology, cytochemistry and functional significance
Cell. Mol. Biol.
The human brain is intrinsically organized into dynamic, anticorrelated functional networks
Proc. Natl. Acad. Sci. U. S. A
Claustral afferents of superior parietal areas PEc and PE in the macaque
J. Comp. Neurol.
A high resolution 7-tesla resting-state fMRI test-retest dataset with cognitive and physiological measures
Scientific Data
Resting-state functional connectivity reflects structural connectivity in the default mode network
Cerebr. Cortex
Cited by (46)
Delayed escape behavior requires claustral activity
2024, Cell ReportsThe claustrum and consciousness: An update
2023, International Journal of Clinical and Health PsychologyPain-induced adaptations in the claustro-cingulate pathway
2023, Cell ReportsInfluence of claustrum on cortex varies by area, layer, and cell type
2023, NeuronCitation Excerpt :Because of its very widespread connections with the cortex, the CLA has been proposed to underlie various high-level functions including salience detection,2,3,4 attentional control,5,6 and consciousness.7,8 In addition, recent studies suggest that CLA is involved in associative learning,9,10 slow-wave activity,11,12 distractor suppression,6,13,14 attentional set shifting,15 task switching,16 and behavioral engagement.17,18 Although its function remains unclear, the anatomy of inputs and outputs to the CLA in rodents has been well characterized.13,19,20,21,22
- 1
co-senior authors.