Hubs are highly connected network components crucial for network functions. In the human brain, hubs are thought to facilitate communication between information-processing systems, and support cognitive functions that likely arise from brain-wide network interactions (Bertolero et al., 2015; Gratton et al., 2018; van den Heuvel and Sporns, 2013). Prior studies have identified hubs in frontoparietal (FP) association cortices that have extensive connections with distributed brain regions (Goldman-Rakic, 1988; Hagmann et al., 2008; Power et al., 2013). These FP hubs are behaviorally significant. For example, the connectivity pattern of cortical hubs correlates with behavioral task performance (Bertolero et al., 2018; Cole et al., 2013), and focal damage to cortical hubs is associated with behavior across cognitive domains (Reber et al., 2021; Warren et al., 2014).

The human thalamus also possesses hub-like network properties (Cole et al., 2010; Hwang et al., 2017). The thalamus consists of different constituent subregions, each with a unique functional, anatomical, and connectivity profile (Sherman and Guillery, 2013). Many thalamic subregions exhibit ‘many-to-one’ and ‘one-to-many’ connectivity motifs—a subregion receives converging projections from multiple cortical regions, and simultaneously projects to multiple cortical regions (Giguere and Goldman-Rakic, 1988; Guillery and Sherman, 2002; Selemon and Goldman-Rakic, 1988). This connectivity motif is a hallmark characteristic of higher-order thalamic nuclei (i.e., the mediodorsal nucleus [MD]), and can be supported by a specific group of thalamocortical projection cells, the ‘matrix’ cells (Jones, 2009). Matrix cells have a diffuse and distributed projection pattern; they innervate the superficial layers of multiple cortical regions, crossing receptive fields and functional boundaries. Consistent with these anatomical features, formal network analyses have found ‘connector hubs’ in the human thalamus. Thalamic connector hubs have strong functional connectivity with multiple cortical networks (Greene et al., 2020; Hwang et al., 2017), and each network can be associated with a distinct set of cognitive functions (Bertolero et al., 2015; Crossley et al., 2013; Yeo et al., 2015). This connectivity architecture suggests that a thalamic hub participates in functions that involve multiple networks and might contribute to many different cognitive functions.

However, the behavioral significance of this important network position of the thalamus has yet to be established. Large-scale meta-analyses of functional neuroimaging research, primarily functional magnetic resonance imaging (fMRI) studies, have found that tasks from many different cognitive domains (e.g., executive function, memory, perception) are associated with increased activity in overlapping thalamic subregions (Hwang et al., 2017; Yeo et al., 2015). Functional neuroimaging findings are nevertheless correlational, because increased blood-oxygen-level-dependent (BOLD) activity in response to a particular behavior is not evidence that this brain region is necessary for the studied behavior (Sutterer and Tranel, 2017). The lesion method, studying patients with focal damage to the thalamus, can provide a stronger test of whether thalamic hubs are necessary for human cognition.

Based on the prominent hub property of the thalamus, we hypothesize that thalamic hubs are involved in multi-domain processing involving multiple functional systems and contribute to behavior across multiple cognitive domains. To test this hypothesis, we combined neuropsychological evaluations from patients with focal thalamic lesions with network analyses of the human thalamocortical functional connectome. To evaluate behavior across cognitive domains, we analyzed neuropsychological tests that assess executive, language, memory, learning, visuospatial, and construction functions. To evaluate the hub properties of different thalamic subregions, we used two different measures: first, a well-established graph-theoretic metric, participation coefficient (PC), and second, the estimated density of matrix projection cells. PC estimates the distribution of coupling between a brain region with multiple brain systems in the functional connectome (Power et al., 2013; Warren et al., 2014). Hub regions with higher PC values are thought to participate in processes recruiting multiple functional networks (Shine et al., 2016). Matrix cells in the thalamus diffusely project to multiple brain regions, and thus thalamic hub regions should have a higher density of matrix cells. We predict that patients with focal lesions to thalamic subregions with high PC values and high density of matrix cells will exhibit more extensive impairment across cognitive domains. In contrast, lesions to thalamic subregions with lower PC values and lower density of matrix cells will exhibit more limited impairment.

We identified 20 patients (ages 18–70, 13 males) with focal lesions restricted to the thalamus, and 21 comparison patients (ages 19–77 years, 8 males) with lesions that spared the thalamus (Figure 1A). All patients were drawn from the Iowa Neurological Patient Registry. The registry contains data from patients with focal, stable brain lesions who have undergone neuropsychological assessment and brain imaging at least 3 months after lesion onset. Comparison patients had lesions that spared the thalamus, predominately lesioned the gray matter (GM) (>80% of total lesion volume), but were similar in size to those found in thalamic patients (≤5088 mm³, size of the largest observed thalamic lesion). Lesion sizes were not significantly different between groups (thalamus group: mean = 1364 mm³, SD = 1212 mm³; comparison group: mean = 1334 mm³, SD = 732 mm³; group difference in lesion size, p = 0.31). There were no significant group differences in age, full scale IQ, verbal IQ, and performance IQ between thalamus patients and comparison patients.

Figure 1

Download asset Open asset

We predicted that lesions to hub regions in the thalamus will affect behavior across different cognitive domains. To test the behavioral relevance of the thalamic hubs, we first compared neuropsychological outcomes between the thalamus and comparison patients. The goal of this comparison was to determine whether lesions of the thalamus were associated with cognitive impairment, beyond any nonspecific lesion effects. We analyzed outcome data from several neuropsychological tests, assessing functions from the following domains: (1) executive function using the Trail Making Test Part B (TMT Part B); (2) verbal naming using the Boston Naming Test (BNT); (3) verbal fluency using the Controlled Oral Word Association Test (COWA); (4) immediate learning using the first trial test score from the Rey Auditory-Verbal Learning Test (RAVLT); (5) total learning by summing scores from RAVLT, trials 1–5; (6) long-term memory recall using the RAVLT 30 min delayed recall score; (7) long-term memory recognition using the RAVLT 30 min delayed recognition score; (8) visuospatial memory using the Rey Complex Figure delayed recall score; (9) psychomotor function using the Trail Making Test Part A (TMT Part A); and (10) construction using the Rey Complex Figure copy test. We grouped these tests into the executive (TMT Part B), verbal (COWA and BNT), memory (RAVLT delayed recall and delayed recognition), learning (RAVLT immediate learning and total learning), psychomotor (TMT Part A), and visuospatial (Complex Figure tests) domains (Lezak et al., 2012). To adjust for demographic factors such as age and years of education, all test scores were transformed to z-scores using published population norms. Statistical significance of between-group comparisons were assessed with the randomized permutation tests unless otherwise noted.

We found that patients with thalamic lesions performed significantly worse than comparison patients on the following tests (Table 1): TMT Part B, BNT, RAVLT delayed recall, and RAVLT delayed recognition. Notably, each test had at least one thalamus patient that performed worse than 95% of the normative population (z-score <–1.645; Figure 1B). These results suggest that thalamic lesions were associated with more severe behavioral impairments in executive, verbal, and memory functions relative to comparison patients that had damage outside of the thalamus.

There are two potential models of thalamocortical connectivity that could explain the observed behavioral impairments. First, each cognitive domain is associated with a distinct thalamocortical system, thus different task impairments are associated with segregated lesion sites within the thalamus. Alternatively, thalamic hubs are involved in many cognitive processes across domains through their widespread connectivity with multiple systems, and thus lesions to a critical hub region could be associated with widespread impairment. The first explanation predicts little to no lesion overlap among different impaired tasks, whereas the second model predicts a high degree of lesion overlap among the impaired tasks. To discern between these two possibilities, we first examined lesion sites associated with impaired performance separately for each task (Figure 2A, left panel). Notably, we found an overlapping lesion site in the left anterior-medio-dorsal thalamus that is associated with impairment across different cognitive domains (Figure 2A, right panel). This result suggests that a patient with a focal thalamic lesion to this multi-domain lesion site could exhibit behavioral impairment across cognitive domains.

Figure 2 with 1 supplement see all

Download asset Open asset

We plotted the degree of impairment (expressed by z-score) across all tasks and cognitive domains separately for each patient (Figure 2B), and found that in 12 out of 20 patients, significant impairment (z < –1.645) was reported in more than two cognitive domains. We then examined whether patients with behavioral impairments across multiple domains had lesions to this identified overlapping site (Figure 2C; for each patient’s lesion, please see Figure 2—figure supplement 1). We found that in the 12 patients that exhibited impairment across multiple domains, there was indeed an overlapping lesion site in the left anterior-medio-dorsal thalamus. This overlapping site was notably absent in the eight patients that exhibited either no behavioral impairment or impairment in one single task (Figure 2C, left panel). This pattern was also observed when examining lesion sites from individual patients, as individual patients with impairments across multiple domains had lesions that overlapped with this left anterior-medio-dorsal thalamic region (Montreal Neurological Institute [MNI] coordinate: –7, 10, 8).

We further tested whether lesions to the left anterior-medio-dorsal thalamus were associated with broad impairment across more cognitive domains rather than being driven by more severe deficits in a limited number of domains. For this purpose, we included an expanded group of 145 comparison patients from the Iowa Neurological Patients Registry (ages 19–81 years, 67 males; for lesion coverage, see Figure 3—figure supplement 1). Unlike the first group of comparison patients (Figure 1), these comparison patients were not matched with the thalamus patients on the lesion size, but on the averaged severity of behavioral deficits across all 10 neuropsychology tests. We predicted that when matched on the severity of behavioral deficit, patients with lesions that overlapped with the left anterior-medio-dorsal thalamus would exhibit impairments on more cognitive domains, whereas comparison patients will exhibit more circumscribed deficits in fewer cognitive domains. To test this prediction, we first calculated an ‘average impairment score’ by averaging the normalized z-scores across all 10 neuropsychological tests, and a ‘multi-domain impairment score’ by summing the number of tests with significant behavioral deficits (defined as z < 1.645). Both scores regressed out the variance associated with differences in lesion size. We then fitted a linear regression model to the group of 145 comparison patients that had average impairment scores similar to thalamus patients (maximum = 1.39, minimum = –1.59). We found that 11 out of 12 thalamus patients with lesions that overlapped with the left anterior-medio-dorsal thalamus (Figure 2C) exhibited higher multi-domain impairment score when compared to comparison patients (Figure 3). This suggests that lesions to the left anterior-medio-dorsal thalamus are not merely associated with more severe behavioral impairment, but also associated with impairment across multiple cognitive domains to a greater extent than would be expected from lesions in other brain regions.

Figure 3 with 1 supplement see all

Download asset Open asset

Because it is possible that impairment across cognitive domains was driven by larger lesions that damaged many functionally specialized subregions in the thalamus, we tested whether there was a significant association between lesion size and the extent of behavioral impairment. We found that there was no significant correlation between lesion volume and number of cognitive domains impaired (r(19) = 0.21, p = 0.36). Furthermore, patients with impairment in more than two cognitive domains did not have larger lesions when compared to patients with impairment in less than two cognitive domains (randomized permutation test p = 0.42). Using the Morel thalamic nuclei atlas, we compared the number of anatomical nuclei in the thalamus overlapped with lesions associated with and without multi-domain impairment. Thalamic nuclei were defined using a published atlas derived from postmortem human brains (Krauth et al., 2010). We found no significant difference (multi-domain lesions: mean = 4.9 nuclei, SD = 1.89; single-domain lesions: mean = 5.92 nuclei, SD = 2.7; randomized permutation p = 0.39). Using a thalamic functional parcellation atlas we previously published that segmented the thalamus into different network parcellations (Hwang et al., 2017), we also did not find significant differences in the number of parcellations between these lesion sites (multi-domain lesions: mean = 3.25 parcels, SD = 1.87; single-domain lesions: mean = 3.88 parcels, SD = 2.59; randomized permutation p = 0.21; randomized permutation p = 0.62). These results indicate that thalamic lesions associated with more global deficits did not involve more thalamic subregions than lesions that did not.

We further evaluated the brain activity maps likely associated with the putative cognitive processes assessed by each of the neuropsychology tests we assessed. Specifically, we utilized the Neurosynth database (Yarkoni et al., 2011), which contains activation loci from thousands of published fMRI studies, to perform automatic meta-analyses and identify brain regions likely recruited for each task. We queried the following terms: ‘executive function’, ‘recall’, ‘recognition’, ‘fluency’, ‘naming’. We found that these terms were associated with distinct brain activity maps with minimal overlap (Figure 4). Maps for terms ‘naming’ and ‘fluency’ overlapped in the left frontal cortex, maps for ‘recall’ and ‘recognition’ overlapped in temporal cortices, and all four maps overlapped in a small region (three 2 mm³ voxels) in the left inferior frontal gyrus.

Figure 4

Download asset Open asset

We performed additional analyses to contrast the hub properties between thalamic lesion sites that were only observed in patients that exhibited impairment across multiple domains, versus lesion sites only observed in patients that exhibited no behavioral impairment or impairment in a single domain (Figure 5A). Lesion sites that overlapped between patients that exhibited multi-domain impairment and patients with no behavioral impairment were excluded. Our prediction was that lesions to thalamic regions exhibiting prominent hub properties—as measured with PC—will have an association with widespread negative behavioral outcomes across multiple domains. In contrast, lesions to thalamic regions with lower hub properties will have less widespread association. We calculated the PC value for each thalamic voxel using a large normative thalamocortical functional connectome dataset (Hwang et al., 2017). The purpose was to estimate hub properties of thalamic subregions in the healthy population and use this result to estimate hub properties of the lesion sites. Briefly, for each normative subject, we calculated thalamocortical functional connectivity between thalamic voxels and 400 cortical regions of interests (ROIs), spanning seven canonical cortical networks (Schaefer et al., 2018). We did not calculate functional connectivity between thalamic voxels. Voxel-wise PC values were calculated for each subject and averaged across subjects. A high PC value indicates that a hub region has distributed connectivity with multiple cortical networks. We found that when averaged across normative subjects, the anterior, medial, and dorsal thalamus exhibit strong connector hub properties (Figure 5B). We then compared the distribution of voxel-wise PC values between the multi-domain and single-domain lesion sites using the Komogorov-Smirnov test, and found that thalamic voxels in the multi-domain lesion sites had on average higher PC values compared to those in the single domain lesion sites (Figure 5C). The voxel-wise PC values were significantly higher in the multi-domain sites compared to the single-domain sites (Kolmogorov-Smirnov d = 0.16, p < 0.001; replication dataset: Kolmogorov-Smirnov d = 0.11, p < 0.001). We then statistically compared PC values between multi-domain and single-domain sites across individual subjects from the normative cohort, and found that the multi-domain sites exhibited significantly higher PC values (Figure 5D, t(234) = 6.472, p < 0.001; replication dataset: t(61) = 3.21, p = 0.002), confirming our central prediction.

Figure 5

Download asset Open asset

Lesions to cortical hubs are known to be associated with widespread behavioral impairments across multiple cognitive domains (Warren et al., 2014; Reber et al., 2021). To replicate this finding and validate our approach with the larger expanded comparison patient group (same sample as Figure 3), we examined whether comparison lesions associated with multi-domain behavioral impairments also exhibited stronger connector hub properties (higher PC values). We grouped the 145 expanded comparison patients into two sub-groups: those with (n = 58) and without (n = 87) behavioral impairments in more than one cognitive domain. We then mapped the lesion sites associated with these two patient groups, and found that comparison lesions associated with multi-domain impairment were primarily located in lateral frontal and posterior parietal associative regions, regions prior studies found to exhibit strong connector hub properties (Figure 6A). To assess the hub property of each comparison patient’s lesion, we calculated the PC value of each cortical GM voxel from the normative functional connectome (Reber et al., 2021). Consistent with findings from prior studies (Bertolero et al., 2015; Power et al., 2013), the lateral frontal and posterior parietal regions exhibit strong connector hub properties (Figure 6B). We then compared the distribution of voxel-wise PC values between the multi-domain and single-domain lesion sites in comparison patients, and found that voxels in the multi-domain lesion sites had on average higher PC values compared to those in the single-domain lesion sites (Figure 6C; Kolmogorov-Smirnov d = 0.11, p < 0.001; replication dataset: Kolmogorov-Smirnov d = 0.14, p < 0.001).

Figure 6

Download asset Open asset

We then compared the hub properties between thalamic lesions and comparison lesions associated with multi-domain impairment (Figure 6D), and found the comparison lesions had a wider distribution of PC values, likely because comparison lesions included in Figure 6 were significantly larger than thalamic lesions (comparison lesions: mean = 36440 mm³, SD = 33078 mm³, thalamic lesions: mean = 1559 mm³, SD = 1312 mm³, randomized permutation test p < 0.001). The peak values were comparable between thalamic lesions and comparison lesions associated with multi-domain impairment (PC = 0.57). We found no statistically significant difference in the multiple-domain impairment score between thalamic and expanded comparison patients with multi-domain impairment (Figure 6E, randomized permutation test p = 0.17).

Findings reported in Figure 5 suggest that lesion sites associated with impairment across cognitive domains have a diverse functional relationship with distributed systems involved in different cognitive functions. Thus, we mapped the cortical functional networks that show strong functional connectivity with voxels within the multi-domain lesion site. For every thalamic voxel, we calculated its average functional connectivity with seven cortical functional networks (Schaefer et al., 2018), including the visual, somatomotor (SM), limbic, dorsal attention (DA), cingulo-opercular (CO), FP, and default mode (DMN) networks. We then divided the functional connectivity estimates of each network by the total summated functional connectivity strength of each voxel. The purpose of this procedure was to derive a functional connectivity weight ratio estimate to assess the network selectivity of each voxel. If a voxel only interacts with a specific network, the majority of its functional connectivity strength should be devoted to that network, resulting in a high functional connectivity weight ratio, whereas connectivity with other networks should be considerably lower. In contrast, if a voxel is broadly interacting with multiple functional networks, then it should exhibit overlapping functional connectivity weight ratios for those networks. Consistent with the high PC values we observed, we found that thalamic voxels in the multi-domain lesion sites exhibit a diffuse functional connectivity relationship with cortical functional networks that are predominately located in heteromodal association areas, including the FP, DMN, limbic, and CO networks (Figure 7A). The centroids of functional connectivity weights were between 0.15 and 0.3 for these networks, and lower (close to 0) for the visual, SM, and DA networks.

Figure 7

Download asset Open asset

We repeated the functional connectivity weight ratio analysis to assess functional connectivity between thalamic voxels and cortical regions identified via the Neurosynth meta-analyses (Figure 4). Specifically, we calculated the functional connectivity between each thalamic voxel and cortical voxels associated with each queried term (‘executive’, ‘naming’, ‘fluency’, ‘recall’, ‘recognition’), then divided the averaged functional connectivity estimates of each term by the total summated functional connectivity strength of each voxel. The purpose was to determine whether there are voxels only selectively interacting with brain regions associated with specific cognitive processes. We found a distributed thalamocortical functional connectivity relationship with brain regions associated with these putative cognitive processes (Figure 7B). Higher functional connectivity weight ratios were found for regions implicated in ‘executive’ and ‘fluency’, and lower ratio for ‘naming’, ‘recall’, and ‘recognition’. Notably, we did not observe a pattern that would indicate strong functional specificity, as the highest observed weight ratio was less than 37%. This result suggests that the thalamic hub region is not selectively interacting with a specific cortical system, instead it diffusely interacts with distinct brain regions associated with these varied cognitive processes.

The thalamus is also known to have dozens of constituent nuclei that can be defined using chemoarchitectual and cytoarchitectual properties. We further examined which thalamic nuclei overlapped with the multi-domain lesion sites, by calculating the percentage of lesioned voxels within the multi-domain lesion sites for each thalamic nucleus (Krauth et al., 2010). We found that the multi-domain lesion sites overlap with several higher-order thalamic nuclei, including the anterior nucleus (AN), the MD, the ventromedial nucleus (VN), the intralaminar nucleus (IL), as well as other nuclei, including the ventro-anterior (VA) and the ventrolateral (VL) nuclei (Figure 7C).

Thalamic subdivisions are comprised of a mixture of ‘core’ and ‘matrix’ thalamocortical projections cells, and each type has a different projection pattern to the cortex (Jones, 2009; Jones, 2001). Importantly, these cell types also express distinct calcium-binding proteins: Calbindin-rich matrix cells project diffusively to the superficial layers of the cerebral cortex, unconstrained by functional borders between cortical regions; whereas in contrast, parvalbumin-rich core cells send topographic specific projections to the middle layers of the cerebral cortex. The distributed projection pattern of matrix cells is conceptually analogous to the inter-network connectivity property of connector hubs. Therefore, we further predicted that the multi-domain lesion sites would contain relatively higher concentrations of matrix cells when compared to the single-domain lesion sites. To test this prediction, we examined data from the Allen Human Brain Atlas that estimated the brain-wide expression of calbindin and parvalbumin proteins, CALB1 and PVALB, respectively (Gryglewski et al., 2018). This allowed us to estimate the relative density of matrix and core cells in each thalamic voxel (Müller et al., 2020). We found that CALB1 were more strongly expressed in the anterior-medio-dorsal thalamus (Figure 8A), suggesting that multi-domain lesion sites likely contain higher densities of matrix projection cells. We then compared the distribution of voxel-wise CALB1 expression between the multi-domain and single-domain lesion sites using the Komogorov-Smirnov test, and found that thalamic voxels in the multi-domain lesion sites had on average higher CALB1 expression when compared to those in the single-domain lesion sites (Figure 8B; Kolmogorov-Smirnov d = 0.483, p < 0.001). This suggests that the multi-domain lesion sites had more matrix cells when compared to the single-domain lesion sites. In contrast, thalamic voxels in the single-domain lesion sites had on average higher PVALB expression compared to those in the multi-domain lesion sites (Figure 8C; Kolmogorov-Smirnov d = 0.482, p < 0.001), suggesting relatively higher concentrations of core cells in the single-domain lesion sites. We then calculated the Dice coefficient to compare the spatial overlap between the CALB1 expression map and lesion masks. We found the CALB1 expression map to be more similar to the multi-domain lesion mask (Dice coefficient = 0.49), and less similar to the single-domain lesion site (Dice coefficient = 0.007).

Figure 8

Download asset Open asset

Prior studies that examined the network organization of the human brain have consistently identified the thalamus as a prominent hub. Within the thalamus there are regional differences in relative hubness, with the anterior, medial, and dorsal thalamus found to exhibit the strongest hub property and broadly connected with distributed functional networks (Cole et al., 2010; Hwang et al., 2017). This connectivity architecture likely allows the thalamus to flexibly participate in functions that support cognition across multiple domains. Findings from the current study provide empirical evidence that confirms the behavioral significance of this network architecture.

The lesion method is well suited for determining the relationship between thalamic hubs and cognition. Lesioning a hub region should weaken network interactions across multiple systems and have a widespread influence on behavior across cognitive domains. In contrast, lesioning a thalamic subregion with a specific cognitive function should have a more limited effect. Past studies have found that lesions to the human thalamus are associated with a wide range of cognitive impairments, including executive dysfunction (Hwang et al., 2020; Liebermann et al., 2013), amnesia (Graff-Radford et al., 1990; Pergola et al., 2016; von Cramon et al., 1985), aphasia (Crosson et al., 1986; Graff-Radford et al., 1984), and attention deficits (de Bourbon-Teles et al., 2014; Snow et al., 2009). However, it is not clear whether deficits reported in prior studies each localize to distinct thalamic regions, or whether a restricted lesion to a hub region can be associated with widespread effects. Our findings help to answer this question.

Specifically, we found significant impairment on the TMT Part B, RAVLT, and BNT tests in patients with lesions to an overlapping region in the left anterior-medio-dorsal thalamus. Of note, basic perceptual and motor functions in thalamic patients were comparable to comparison patients, given that we did not find a significant difference in the TMT Part A and Rey Complex Figure construction test. Thus, the impairment we observed is related to higher-order cognitive processes related to language, memory, and executive functions. How can a thalamic subregion be simultaneously associated with executive, language, and memory functions? Our framework suggests that each of these cognitive domains is associated with a different brain system, and these brain systems have converging connectivity with the multi-domain lesion site that we identified. In support, we found that this lesion site exhibited a strong connector hub property, with a diverse functional connectivity relationship with multiple cortical networks. Importantly, the connector hub property of this multi-domain lesion site was greater than lesion sites that were associated with more limited impairment, again confirming the behavioral relevance of network hubs.

We suspect that the left thalamus is involved because language-related functions are more lateralized to the left hemisphere, and cognitive performance using standard neuropsychological tests tends to be left-lateralized (Adolphs et al., 2020). Successful performance on the delayed recall and delayed recognition tests likely requires a long-term memory system within which the anterior thalamus is a pivotal component, interlinking the hippocampal and cortical systems (Aggleton et al., 2010). Prior studies on lesion symptom mapping have found that TMT and language-related tasks are associated with medial frontal, lateral frontal, and temporal cortices (Baldo et al., 2013; Glascher et al., 2012). These brain regions likely overlap with the FP, CO, and DF networks that we found to have strong functional connectivity with the multi-domain lesion site. Furthermore, these cortical regions are known to receive anatomical projections from the AN and MD, which we also found to overlap with the identified lesion site (Barbas et al., 1991; Giguere and Goldman-Rakic, 1988; Selemon and Goldman-Rakic, 1988). In addition to the AN and MD, the multi-domain lesion sites also overlapped with the IL and VN. These thalamic nuclei are known to have higher densities of matrix projection cells (Jones, 2009) that express Calbindin-binding proteins. Matrix cells broadly project to multiple cortical regions crossing receptive fields and functional boundaries (Jones, 2001). This indicates that matrix cells simultaneously project to segregated and distributed cortical regions, an anatomical feature that is consistent with the properties of brain network hubs (Müller et al., 2020). Therefore, matrix cells may be the anatomical connectivity substrate that allows the anterior-medio-dorsal thalamus to promote its connector hub functions that we will discuss below.

How do thalamic connector hubs support cognitive functions across these diverse domains? Studies on brain network organization may offer some insights. For example, brain functions engage functional segregation, where segregated systems can perform specialized functions without interference, and functional integration, where outputs from specialized systems, can be integrated between domains via connector hubs that interlink distributed systems (Bertolero et al., 2015; Cohen and D’Esposito, 2016; Shine et al., 2016). This modular-like organization balances segregated and integrated processes and can be reliably observed in the human brain (Meunier et al., 2010; Sporns and Betzel, 2016). Furthermore, an optimal modular structure has been found to correlate with behavioral performance on tasks across multiple domains (Bertolero et al., 2018), and lesions to connector hubs, including the thalamus, have been shown to disrupt modular organization more so than other non-hub lesions (Gratton et al., 2012; Hwang et al., 2017). Therefore, lesions to thalamic hubs may disrupt the optimal organization of multiple systems via its widespread thalamocortical connectivity, and affect both functional segregation and integration. This disruption will not be limited to one system nor to a single cognitive domain, a prediction that is consistent with our finding. Thus, one hypothetical function of thalamic connector hubs is to maintain a cognitively optimal architecture across multiple functional systems to support diverse cognitive processes.

Our results further indicate that we cannot fully dissociate the averaged severity of behavioral impairment from the diversity of behavioral impairment, as patients with multi-domain impairment were also more severely impaired across multiple tests. This suggests that there could be common latent processes that are necessary for performing different neuropsychological tests, and disruptions in these common processes might globally affect behavior without specificity. For example, it is possible that thalamic lesions are associated with distal disruptions (‘diaschisis’; Sutterer and Tranel, 2017; Von Monakow, 1911) of cortical systems involved in general task-control functions that impact behavior across multiple domains. We found that the multi-domain lesion site has strong functional connectivity with FP and CO networks, which are cortical networks hypothesized to be involved in domain general processes such as maintaining task-relevant information, guiding relevant sensory inputs toward the correct behavioral output, and adjusting behavior when errors are made (Dosenbach et al., 2008; Gratton et al., 2018). It is also possible that thalamic lesions disrupt processing in a hypothetical supraordinate, domain-general cortical system, known as the ‘multi-demand system’ (Cole et al., 2013; Duncan, 2010). The multi-demand system consists of a set of medial frontal, lateral frontal, and posterior parietal regions do not have clear functional specializations, but are instead broadly tuned to implement operations that might be required to perform the different neuropsychological tests that we utilized. This system partially overlaps with the CO and FP networks (Gratton et al., 2018), and a prior study found strong functional connectivity between this multi-domain system with the anterior and dorsal thalamus (Assem et al., 2020), which overlaps with the multi-domain lesion site that we identified. Thus, lesions to thalamic hubs may disrupt functions associated with the CO and FP networks, or the multi-demand system, which in turn globally affect functions that are necessary for cognition across domains.

Evidence for animal models further suggest that thalamic hubs can influence cortical activity. For example, inhibiting the VL or MD in rodents diminished cortical evoked activity in rodents (Bolkan et al., 2017; Guo et al., 2017), and stimulating the MD can enhance the information coding in rodent’s medial frontal cortex. In addition to modulating evoked activity, thalamus may also facilitate inter-regional communication via its diverging hub-like connectivity. For example, it has been found that inactivation of the thalamus diminishes the effectiveness of cortico-cortical communication (Theyel et al., 2010; Zhou et al., 2016). We hypothesize that thalamic modulations of evoked responses, promoting inter-regional communication, maintaining modular organization of functional systems, interacting with task-control networks, may together be generalized processes that are necessary for a wide range of behavioral tasks. Given that these mechanisms are not domain-specific, damage to thalamic hubs may broadly affect tasks that engage these processes for optimal performance.

One limitation of our study is that we did not have equal sampling of lesion locations throughout different thalamic subregions. Our results could be potentially biased by the higher concentration of anterior and medial thalamic lesions in our sample, and missed other multi-domain hubs (e.g., the posterior-medial thalamus) that we were under-powered to detect. Given that the posterior-medial thalamus has also been found to exhibit strong hub properties (Hwang et al., 2017; Greene et al., 2020), focal lesions to the posterior-medial thalamus may also be associated with behavioral deficits across multiple behavioral domains. Alternatively, because we found that the anterior but not the posterior thalamus had denser expression of the CALB1 protein, a marker for matrix thalamocortical projection cells, it is also possible that only the anterior-medial but not the posterior-medial thalamus is a multi-domain hub. A negative finding would suggest that PC, a graph metric we and many others used for mapping hubs in the human brain, may need to be combined with sources of data (e.g., behavior and gene expression) to more reliably identify multi-domain hubs. Our study does not have enough data to sufficiently adjudicate these two competing predictions, and this question will have to be addressed by future studies.

It is also possible that each of these neuropsychological tasks recruits a distinct, parallel thalamocortical circuity (Alexander et al., 1986), and that larger lesions can cover many small specialized subregions in the thalamus, thus affecting behavior across cognitive domains. Results from ours and previous studies do not support this explanation. First, lesion size did not correlate with the degree of multi-domain impairment. Second, we observed multi-domain impairment in several patients with restricted lesions to the left anterior-medio-dorsal thalamus. Third, the anterior-medio-dorsal lesion site did not cover more thalamic sub-parcellations. Fourth, previous large-scale meta-analyses of fMRI research have found that multiple tasks increased BOLD activity in overlapping thalamic subregions, but not segregated thalamic subregions (Hwang et al., 2017; Yeo et al., 2015). Fifth, our functional connectivity weight ratio analysis also showed that most, if not all voxels in the multi-domain lesion site have a broad functional connectivity relationship with multiple cortical networks. Finally, Neurosynth queries showed that cognitive processes putatively involved in the impaired neuropsychological tests each associated different sets of spatially segregated brain regions. In other words, the multi-domain lesion site does not appear to have strong functional specificity. Thus, while it is possible that the lesion method does not have the resolution and anatomical specificity to detect lesions that cover many small, spatially restricted, functionally-specific thalamic subregions, our results do not support this interpretation. Instead, the cross-domain impairment that we observed were likely associated with lesions to thalamic hubs that have a converging relationship with multiple systems.

To conclude, the principal contribution of our study is to demonstrate the behavioral significance of thalamic hubs. We found that a thalamic hub in the left anterior-medio-dorsal thalamus is critical for memory, executive, and language-related functions. This significant behavioral profile supports the prominent network position of the thalamic hub. These findings lead to the question: what is the function that is implemented by a thalamic hub that allows it to be broadly involved in cognition? Our findings constrain the possible answers—it must be processes that are not specific to a single cognitive function but can be generalized across domains. As discussed above, one possibility is that thalamic hubs maintain an optimal functional network structure, to promote both segregated and integrated functions, which are domain-general network processes that support cognition across multiple domains.

We have made all code and lesion-derived measures used in the manuscript freely available on github (https://github.com/kaihwang/LTH, (copy archived at swh:1:rev:fdcca3fd575911fa42fc79bccc468c2f9f6e6153)), including neuropsych assessment outcome, derivatives from lesion analyses, data used for functional connectivity analyses, and mRNA expression analyses. Functional connectivity analyses utilized publicly available datasets (Holmes et al., 2015; Nooner et al., 2012). The only data that we cannot post without restrictions are each patient's clinical MRI data and lesion data. Patients were enrolled into the Iowa Lesion Patient Registry the past few decades, and most did not consent to post their clinical MRI data publicly. To gain access to those data, the interested party will have to contact the PI of the lesion registry, Dr. Dan Tranel, and the corresponding author of this project, Dr. Kai Hwang. The user will require to sign a data use agreement. This institutional policy was designed to ensure the appropriate use of the data for academic and not commercial purposes. A study plan of the proposed research will have to be submitted, and we will work with the interested party to obtain the necessary IRB approval from both institutions.

1. 1. Homes et al

   (2015) Brain Genomics Superstruct Project initial data release with structural

   ID functional. Brain Genomics Superstruct Project.

   https://www.nature.com/articles/sdata201531
2. 1. Nooner et al

   (2012) The enhanced Nathan Kline Institute-Rockland Sample

   ID NKI-RS. NKI-Rockland sample.

   http://fcon_1000.projects.nitrc.org/indi/enhanced/

1. 1. Adolphs R
   2. Bruss J
   3. Manzel K
   4. Corbetta M
   5. Tranel D
   6. Boes AD

   (2020) Multivariate Lesion-Behavior Mapping of General Cognitive Ability and Its Psychometric Constituents

   The Journal of Neuroscience 40:8924–8937.

   https://doi.org/10.1523/JNEUROSCI.1415-20.2020

   • PubMed
   • Google Scholar
2. 1. Aggleton JP
   2. O’Mara SM
   3. Vann SD
   4. Wright NF
   5. Tsanov M
   6. Erichsen JT

   (2010) Hippocampal-anterior thalamic pathways for memory: uncovering a network of direct and indirect actions

   The European Journal of Neuroscience 31:2292–2307.

   https://doi.org/10.1111/j.1460-9568.2010.07251.x

   • PubMed
   • Google Scholar
3. 1. Alexander GE
   2. DeLong MR
   3. Strick PL

   (1986) Parallel organization of functionally segregated circuits linking basal ganglia and cortex

   Annual Review of Neuroscience 9:357–381.

   https://doi.org/10.1146/annurev.ne.09.030186.002041

   • PubMed
   • Google Scholar
4. 1. Assem M
   2. Glasser MF
   3. Van Essen DC
   4. Duncan J

   (2020) A Domain-General Cognitive Core Defined in Multimodally Parcellated Human Cortex

   Cerebral Cortex 30:4361–4380.

   https://doi.org/10.1093/cercor/bhaa023

   • PubMed
   • Google Scholar
5. 1. Avants BB
   2. Tustison N
   3. Song G

   (2009)

   Advanced normalization tools (ANTS

   The Insight Journal 2:1–35.

   • Google Scholar
6. 1. Baldo JV
   2. Arévalo A
   3. Patterson JP
   4. Dronkers NF

   (2013) Grey and white matter correlates of picture naming: evidence from a voxel-based lesion analysis of the Boston Naming Test

   Cortex; a Journal Devoted to the Study of the Nervous System and Behavior 49:658–667.

   https://doi.org/10.1016/j.cortex.2012.03.001

   • PubMed
   • Google Scholar
7. 1. Barbas H
   2. Henion TH
   3. Dermon CR

   (1991) Diverse thalamic projections to the prefrontal cortex in the rhesus monkey

   The Journal of Comparative Neurology 313:65–94.

   https://doi.org/10.1002/cne.903130106

   • PubMed
   • Google Scholar
8. 1. Behzadi Y
   2. Restom K
   3. Liau J
   4. Liu TT

   (2007) A component based noise correction method (CompCor) for BOLD and perfusion based fMRI

   NeuroImage 37:90–101.

   https://doi.org/10.1016/j.neuroimage.2007.04.042

   • PubMed
   • Google Scholar
9. 1. Bertolero MA
   2. Yeo BTT
   3. D’Esposito M

   (2015) The modular and integrative functional architecture of the human brain

   PNAS 112:E6798–E6807.

   https://doi.org/10.1073/pnas.1510619112

   • Google Scholar
10. 1. Bertolero MA
    2. Yeo BTT
    3. Bassett DS
    4. D’Esposito M

    (2018) A mechanistic model of connector hubs, modularity and cognition

    Nature Human Behaviour 2:765–777.

    https://doi.org/10.1038/s41562-018-0420-6

    • PubMed
    • Google Scholar
11. 1. Bolkan SS
    2. Stujenske JM
    3. Parnaudeau S
    4. Spellman TJ
    5. Rauffenbart C
    6. Abbas AI
    7. Harris AZ
    8. Gordon JA
    9. Kellendonk C

    (2017) Thalamic projections sustain prefrontal activity during working memory maintenance

    Nature Neuroscience 20:987–996.

    https://doi.org/10.1038/nn.4568

    • PubMed
    • Google Scholar
12. 1. Bowie CR
    2. Harvey PD

    (2006) Administration and interpretation of the Trail Making Test

    Nature Protocols 1:2277–2281.

    https://doi.org/10.1038/nprot.2006.390

    • PubMed
    • Google Scholar
13. 1. Brett M
    2. Leff AP
    3. Rorden C
    4. Ashburner J

    (2001) Spatial normalization of brain images with focal lesions using cost function masking

    NeuroImage 14:486–500.

    https://doi.org/10.1006/nimg.2001.0845

    • PubMed
    • Google Scholar
14. 1. Buckner RL
    2. Krienen FM
    3. Castellanos A
    4. Diaz JC
    5. Yeo BTT

    (2011) The organization of the human cerebellum estimated by intrinsic functional connectivity

    Journal of Neurophysiology 106:2322–2345.

    https://doi.org/10.1152/jn.00339.2011

    • PubMed
    • Google Scholar
15. 1. Choi EY
    2. Yeo BTT
    3. Buckner RL

    (2012) The organization of the human striatum estimated by intrinsic functional connectivity

    Journal of Neurophysiology 108:2242–2263.

    https://doi.org/10.1152/jn.00270.2012

    • PubMed
    • Google Scholar
16. 1. Cohen JR
    2. D’Esposito M

    (2016) The Segregation and Integration of Distinct Brain Networks and Their Relationship to Cognition

    The Journal of Neuroscience 36:12083–12094.

    https://doi.org/10.1523/JNEUROSCI.2965-15.2016

    • PubMed
    • Google Scholar
17. 1. Cole M.W
    2. Pathak S
    3. Schneider W

    (2010) Identifying the brain’s most globally connected regions

    NeuroImage 49:3132–3148.

    https://doi.org/10.1016/j.neuroimage.2009.11.001

    • PubMed
    • Google Scholar
18. 1. Cole M
    2. Reynolds JR
    3. Power JD
    4. Repovs G
    5. Anticevic A
    6. Braver TS

    (2013) Multi-task connectivity reveals flexible hubs for adaptive task control

    Nature Neuroscience 16:1348–1355.

    https://doi.org/10.1038/nn.3470

    • PubMed
    • Google Scholar
19. 1. Crossley NA
    2. Mechelli A
    3. Vértes PE
    4. Winton-Brown TT
    5. Patel AX
    6. Ginestet CE
    7. McGuire P
    8. Bullmore ET

    (2013) Cognitive relevance of the community structure of the human brain functional coactivation network

    PNAS 110:11583–11588.

    https://doi.org/10.1073/pnas.1220826110

    • PubMed
    • Google Scholar
20. 1. Crosson B
    2. Parker JC
    3. Kim AK
    4. Warren RL
    5. Kepes JJ
    6. Tully R

    (1986) A case of thalamic aphasia with postmortem verification

    Brain and Language 29:301–314.

    https://doi.org/10.1016/0093-934X(86)90050-7

    • PubMed
    • Google Scholar
21. 1. Crowe SF

    (1998) The differential contribution of mental tracking, cognitive flexibility, visual search, and motor speed to performance on parts A and B of the Trail Making Test

    Journal of Clinical Psychology 54:585–591.

    https://doi.org/10.1002/(SICI)1097-4679(199808)54:5<585::AID-JCLP4>3.0.CO;2-K

    • PubMed
    • Google Scholar
22. 1. de Bourbon-Teles J
    2. Bentley P
    3. Koshino S
    4. Shah K
    5. Dutta A
    6. Malhotra P
    7. Egner T
    8. Husain M
    9. Soto D

    (2014) Thalamic control of human attention driven by memory and learning

    Current Biology 24:993–999.

    https://doi.org/10.1016/j.cub.2014.03.024

    • PubMed
    • Google Scholar
23. 1. Dosenbach NUF
    2. Fair DA
    3. Cohen AL
    4. Schlaggar BL
    5. Petersen SE

    (2008) A dual-networks architecture of top-down control

    Trends in Cognitive Sciences 12:99–105.

    https://doi.org/10.1016/j.tics.2008.01.001

    • PubMed
    • Google Scholar
24. 1. Duncan J

    (2010) The multiple-demand (MD) system of the primate brain: mental programs for intelligent behaviour

    Trends in Cognitive Sciences 14:172–179.

    https://doi.org/10.1016/j.tics.2010.01.004

    • PubMed
    • Google Scholar
25. 1. Fastenau PS
    2. Denburg NL
    3. Hufford BJ

    (1999) Adult norms for the Rey-Osterrieth Complex Figure Test and for supplemental recognition and matching trials from the Extended Complex Figure Test

    The Clinical Neuropsychologist 13:30–47.

    https://doi.org/10.1076/clin.13.1.30.1976

    • PubMed
    • Google Scholar
26. 1. Fonov VS
    2. Evans AC
    3. McKinstry RC
    4. Almli CR
    5. Collins DL

    (2009) Unbiased nonlinear average age-appropriate brain templates from birth to adulthood

    NeuroImage 47:S102.

    https://doi.org/10.1016/S1053-8119(09)70884-5

    • Google Scholar
27. 1. Giguere M
    2. Goldman-Rakic PS

    (1988) Mediodorsal nucleus: areal, laminar, and tangential distribution of afferents and efferents in the frontal lobe of rhesus monkeys

    The Journal of Comparative Neurology 277:195–213.

    https://doi.org/10.1002/cne.902770204

    • PubMed
    • Google Scholar
28. 1. Glascher J
    2. Adolphs R
    3. Damasio H
    4. Bechara A
    5. Rudrauf D
    6. Calamia M
    7. Paul LK
    8. Tranel D

    (2012) Lesion mapping of cognitive control and value-based decision making in the prefrontal cortex

    PNAS 109:14681–14686.

    https://doi.org/10.1073/pnas.1206608109

    • Google Scholar
29. 1. Goldman-Rakic PS

    (1988) Topography of cognition: parallel distributed networks in primate association cortex

    Annual Review of Neuroscience 11:137–156.

    https://doi.org/10.1146/annurev.ne.11.030188.001033

    • PubMed
    • Google Scholar
30. 1. Graff-Radford NR
    2. Eslinger PJ
    3. Damasio AR
    4. Yamada T

    (1984) Nonhemorrhagic infarction of the thalamus: behavioral, anatomic, and physiologic correlates

    Neurology 34:14–23.

    https://doi.org/10.1212/wnl.34.1.14

    • PubMed
    • Google Scholar
31. 1. Graff-Radford NR
    2. Tranel D
    3. Van Hoesen GW
    4. Brandt JP

    (1990) Diencephalic amnesia

    Brain 113:1–25.

    https://doi.org/10.1093/brain/113.1.1

    • PubMed
    • Google Scholar
32. 1. Gratton C
    2. Nomura EM
    3. Pérez F
    4. D’Esposito M

    (2012) Focal brain lesions to critical locations cause widespread disruption of the modular organization of the brain

    Journal of Cognitive Neuroscience 24:1275–1285.

    https://doi.org/10.1162/jocn_a_00222

    • PubMed
    • Google Scholar
33. 1. Gratton C
    2. Sun H
    3. Petersen SE

    (2018) Control networks and hubs

    Psychophysiology 55:13032.

    https://doi.org/10.1111/psyp.13032

    • PubMed
    • Google Scholar
34. 1. Greene DJ
    2. Marek S
    3. Gordon EM
    4. Siegel JS
    5. Gratton C
    6. Laumann TO
    7. Gilmore AW
    8. Berg JJ
    9. Nguyen AL
    10. Dierker D
    11. Van AN
    12. Ortega M
    13. Newbold DJ
    14. Hampton JM
    15. Nielsen AN
    16. McDermott KB
    17. Roland JL
    18. Norris SA
    19. Nelson SM
    20. Snyder AZ
    21. Schlaggar BL
    22. Petersen SE
    23. Dosenbach NUF

    (2020) tegrative and Network-Specific Connectivity of the Basal Ganglia and Thalamus Defined in Individuals

    Neuron 105:742–758.

    https://doi.org/10.1016/j.neuron.2019.11.012

    • PubMed
    • Google Scholar
35. 1. Gryglewski G
    2. Seiger R
    3. James GM
    4. Godbersen GM
    5. Komorowski A
    6. Unterholzner J
    7. Michenthaler P
    8. Hahn A
    9. Wadsak W
    10. Mitterhauser M
    11. Kasper S
    12. Lanzenberger R

    (2018) Spatial analysis and high resolution mapping of the human whole-brain transcriptome for integrative analysis in neuroimaging

    NeuroImage 176:259–267.

    https://doi.org/10.1016/j.neuroimage.2018.04.068

    • PubMed
    • Google Scholar
36. 1. Guillery RW
    2. Sherman SM

    (2002) Thalamic relay functions and their role in corticocortical communication: generalizations from the visual system

    Neuron 33:163–175.

    https://doi.org/10.1016/s0896-6273(01)00582-7

    • PubMed
    • Google Scholar
37. 1. Guimerà R
    2. Nunes Amaral LA

    (2005) Functional cartography of complex metabolic networks

    Nature 433:895–900.

    https://doi.org/10.1038/nature03288

    • PubMed
    • Google Scholar
38. 1. Guo ZV
    2. Inagaki HK
    3. Daie K
    4. Druckmann S
    5. Gerfen CR
    6. Svoboda K

    (2017) Maintenance of persistent activity in a frontal thalamocortical loop

    Nature 545:181–186.

    https://doi.org/10.1038/nature22324

    • PubMed
    • Google Scholar
39. 1. Hagmann P
    2. Cammoun L
    3. Gigandet X
    4. Meuli R
    5. Honey CJ
    6. Wedeen VJ
    7. Sporns O

    (2008) Mapping the structural core of human cerebral cortex

    PLOS Biology 6:e159.

    https://doi.org/10.1371/journal.pbio.0060159

    • PubMed
    • Google Scholar
40. 1. Holmes AJ
    2. Hollinshead MO
    3. O’Keefe TM
    4. Petrov VI
    5. Fariello GR
    6. Wald LL
    7. Fischl B
    8. Rosen BR
    9. Mair RW
    10. Roffman JL
    11. Smoller JW
    12. Buckner RL

    (2015) Brain Genomics Superstruct Project initial data release with structural, functional, and behavioral measures

    Scientific Data 2:150031.

    https://doi.org/10.1038/sdata.2015.31

    • PubMed
    • Google Scholar
41. 1. Hwang Kai
    2. Bertolero MA
    3. Liu WB
    4. D’Esposito M

    (2017) The human thalamus is an integrative hub for functional brain networks

    The Journal of Neuroscience 37:5594–5607.

    https://doi.org/10.1523/JNEUROSCI.0067-17.2017

    • PubMed
    • Google Scholar
42. 1. Hwang K
    2. Bruss J
    3. Tranel D
    4. Boes AD

    (2020) Network Localization of Executive Function Deficits in Patients with Focal Thalamic Lesions

    Journal of Cognitive Neuroscience 32:2303–2319.

    https://doi.org/10.1162/jocn_a_01628

    • PubMed
    • Google Scholar
43. Software

    1. Hwang K

    (2021) Data and code for Neuropsychological evidence of multi-domain network hubs in the human thalamus

    Software Heritage.

    https://archive.softwareheritage.org/swh:1:dir:d877ded181e271c37e39077f6dedd3536f88f561;origin=https://github.com/kaihwang/LTH;visit=swh:1:snp:c557c5d1c9b45ef760ce4f96845da0cac3f845b4;anchor=swh:1:rev:fdcca3fd575911fa42fc79bccc468c2f9f6e6153
44. 1. Jones EG
    2. Hendry SHC

    (1989) Differential Calcium Binding Protein Immunoreactivity Distinguishes Classes of Relay Neurons in Monkey Thalamic Nuclei

    The European Journal of Neuroscience 1:222–246.

    https://doi.org/10.1111/j.1460-9568.1989.tb00791.x

    • PubMed
    • Google Scholar
45. 1. Jones EG

    (2001) The thalamic matrix and thalamocortical synchrony

    Trends in Neurosciences 24:595–601.

    https://doi.org/10.1016/s0166-2236(00)01922-6

    • PubMed
    • Google Scholar
46. 1. Jones EG

    (2009) Synchrony in the interconnected circuitry of the thalamus and cerebral cortex

    Annals of the New York Academy of Sciences 1157:10–23.

    https://doi.org/10.1111/j.1749-6632.2009.04534.x

    • PubMed
    • Google Scholar
47. 1. Kortte KB
    2. Horner MD
    3. Windham WK

    (2002) The trail making test, part B: cognitive flexibility or ability to maintain set?

    Applied Neuropsychology 9:106–109.

    https://doi.org/10.1207/S15324826AN0902_5

    • PubMed
    • Google Scholar
48. 1. Krauth A
    2. Blanc R
    3. Poveda A
    4. Jeanmonod D
    5. Morel A
    6. Székely G

    (2010) A mean three-dimensional atlas of the human thalamus: generation from multiple histological data

    NeuroImage 49:2053–2062.

    https://doi.org/10.1016/j.neuroimage.2009.10.042

    • PubMed
    • Google Scholar
49. Book

    1. Lezak MD
    2. Howieson DB
    3. Bigler ED
    4. Tranel D

    (2012)

    Neuropsychological Assessment

    Oxford University Press.

    • Google Scholar
50. 1. Liebermann D
    2. Ploner CJ
    3. Kraft A
    4. Kopp UA
    5. Ostendorf F

    (2013) A dysexecutive syndrome of the medial thalamus

    Cortex; a Journal Devoted to the Study of the Nervous System and Behavior 49:40–49.

    https://doi.org/10.1016/j.cortex.2011.11.005

    • PubMed
    • Google Scholar
51. 1. Meunier D
    2. Lambiotte R
    3. Bullmore ET

    (2010) Modular and hierarchically modular organization of brain networks

    Frontiers in Neuroscience 4:200.

    https://doi.org/10.3389/fnins.2010.00200

    • PubMed
    • Google Scholar
52. 1. Meyers JE
    2. Meyers KR

    (1995) Rey complex figure test under four different administration procedures

    The Clinical Neuropsychologist 9:63–67.

    https://doi.org/10.1080/13854049508402059

    • Google Scholar
53. 1. Müller EJ
    2. Munn B
    3. Hearne LJ
    4. Smith JB
    5. Fulcher B
    6. Arnatkevičiūtė A
    7. Lurie DJ
    8. Cocchi L
    9. Shine JM

    (2020) Core and matrix thalamic sub-populations relate to spatio-temporal cortical connectivity gradients

    NeuroImage 222:117224.

    https://doi.org/10.1016/j.neuroimage.2020.117224

    • PubMed
    • Google Scholar
54. 1. Nachev P
    2. Coulthard E
    3. Jäger HR
    4. Kennard C
    5. Husain M

    (2008) Enantiomorphic normalization of focally lesioned brains

    NeuroImage 39:1215–1226.

    https://doi.org/10.1016/j.neuroimage.2007.10.002

    • PubMed
    • Google Scholar
55. 1. Nooner KB
    2. Colcombe SJ
    3. Tobe RH
    4. Mennes M
    5. Benedict MM
    6. Moreno AL
    7. Panek LJ
    8. Brown S
    9. Zavitz ST
    10. Li Q
    11. Sikka S
    12. Gutman D
    13. Bangaru S
    14. Schlachter RT
    15. Kamiel SM
    16. Anwar AR
    17. Hinz CM
    18. Kaplan MS
    19. Rachlin AB
    20. Adelsberg S
    21. Cheung B
    22. Khanuja R
    23. Yan C
    24. Craddock CC
    25. Calhoun V
    26. Courtney W
    27. King M
    28. Wood D
    29. Cox CL
    30. Kelly AMC
    31. Di Martino A
    32. Petkova E
    33. Reiss PT
    34. Duan N
    35. Thomsen D
    36. Biswal B
    37. Coffey B
    38. Hoptman MJ
    39. Javitt DC
    40. Pomara N
    41. Sidtis JJ
    42. Koplewicz HS
    43. Castellanos FX
    44. Leventhal BL
    45. Milham MP

    (2012) The NKI-Rockland Sample: A Model for Accelerating the Pace of Discovery Science in Psychiatry

    Frontiers in Neuroscience 6:152.

    https://doi.org/10.3389/fnins.2012.00152

    • PubMed
    • Google Scholar
56. 1. Pergola G
    2. Danet L
    3. Barbeau EJ
    4. Eustache P
    5. Planton M
    6. Raposo N
    7. Sibon I
    8. Albucher JF
    9. Bonneville F
    10. Peran P
    11. Pariente J

    (2016) Thalamic amnesia after infarct: The role of the mammillothalamic tract and mediodorsal nucleus

    Neurology 86:1928.

    https://doi.org/10.1212/WNL.0000000000002730

    • PubMed
    • Google Scholar
57. 1. Power JD
    2. Schlaggar BL
    3. Lessov-Schlaggar CN
    4. Petersen SE

    (2013) Evidence for hubs in human functional brain networks

    Neuron 79:798–813.

    https://doi.org/10.1016/j.neuron.2013.07.035

    • PubMed
    • Google Scholar
58. 1. Reber J
    2. Hwang K
    3. Bowren M
    4. Bruss J
    5. Mukherjee P
    6. Tranel D
    7. Boes AD

    (2021) Cognitive impairment after focal brain lesions is better predicted by damage to structural than functional network hubs

    PNAS 118:e2018784118.

    https://doi.org/10.1073/pnas.2018784118

    • Google Scholar
59. 1. Ruff RM
    2. Light RH
    3. Parker SB
    4. Levin HS

    (1996)

    Benton Controlled Oral Word Association Test: reliability and updated norms

    Archives of Clinical Neuropsychology 11:329–338.

    • PubMed
    • Google Scholar
60. 1. Schaefer A
    2. Kong R
    3. Gordon EM
    4. Laumann TO
    5. Zuo X-N
    6. Holmes AJ
    7. Eickhoff SB
    8. Yeo BTT

    (2018) Local-Global Parcellation of the Human Cerebral Cortex from Intrinsic Functional Connectivity MRI

    Cerebral Cortex 28:3095–3114.

    https://doi.org/10.1093/cercor/bhx179

    • PubMed
    • Google Scholar
61. 1. Selemon LD
    2. Goldman-Rakic PS

    (1988) Common cortical and subcortical targets of the dorsolateral prefrontal and posterior parietal cortices in the rhesus monkey: evidence for a distributed neural network subserving spatially guided behavior

    The Journal of Neuroscience 8:4049–4068.

    https://doi.org/10.1523/JNEUROSCI.08-11-04049.1988

    • PubMed
    • Google Scholar
62. Book

    1. Sherman SM
    2. Guillery RW

    (2013)

    Functional Connections of Cortical Areas: A New View from the Thalamus

    MIT Press.

    • Google Scholar
63. 1. Shine JM
    2. Bissett PG
    3. Bell PT
    4. Koyejo O
    5. Balsters JH
    6. Gorgolewski KJ
    7. Moodie CA
    8. Poldrack RA

    (2016) The Dynamics of Functional Brain Networks: Integrated Network States during Cognitive Task Performance

    Neuron 92:544–554.

    https://doi.org/10.1016/j.neuron.2016.09.018

    • PubMed
    • Google Scholar
64. 1. Snow JC
    2. Allen HA
    3. Rafal RD
    4. Humphreys GW

    (2009) Impaired attentional selection following lesions to human pulvinar: evidence for homology between human and monkey

    PNAS 106:4054–4059.

    https://doi.org/10.1073/pnas.0810086106

    • PubMed
    • Google Scholar
65. 1. Sporns O
    2. Betzel RF

    (2016) Modular Brain Networks

    Annual Review of Psychology 67:613–640.

    https://doi.org/10.1146/annurev-psych-122414-033634

    • PubMed
    • Google Scholar
66. 1. Sutterer MJ
    2. Tranel D

    (2017) Neuropsychology and cognitive neuroscience in the fMRI era: A recapitulation of localizationist and connectionist views

    Neuropsychology 31:972–980.

    https://doi.org/10.1037/neu0000408

    • PubMed
    • Google Scholar
67. 1. Theyel BB
    2. Llano DA
    3. Sherman SM

    (2010) The corticothalamocortical circuit drives higher-order cortex in the mouse

    Nature Neuroscience 13:84–88.

    https://doi.org/10.1038/nn.2449

    • PubMed
    • Google Scholar
68. 1. Tombaugh TN
    2. Hubley AM

    (1997) The 60-item Boston Naming Test: norms for cognitively intact adults aged 25 to 88 years

    Journal of Clinical and Experimental Neuropsychology 19:922–932.

    https://doi.org/10.1080/01688639708403773

    • PubMed
    • Google Scholar
69. 1. Vakil E
    2. Blachstein H

    (1993) Rey Auditory-Verbal Learning Test: structure analysis

    Journal of Clinical Psychology 49:883–890.

    https://doi.org/10.1002/1097-4679(199311)49:6<883::AID-JCLP2270490616>3.0.CO;2-6

    • PubMed
    • Google Scholar
70. 1. van den Heuvel MP
    2. Sporns O

    (2013) Network hubs in the human brain

    Trends in Cognitive Sciences 17:683–696.

    https://doi.org/10.1016/j.tics.2013.09.012

    • PubMed
    • Google Scholar
71. 1. von Cramon DY
    2. Hebel N
    3. Schuri U

    (1985) A contribution to the anatomical basis of thalamic amnesia

    Brain 108 (Pt 4):993–1008.

    https://doi.org/10.1093/brain/108.4.993

    • PubMed
    • Google Scholar
72. 1. Von Monakow C

    (1911)

    Localization of brain functions

    Z Psychol Neurol 17:185–200.

    • Google Scholar
73. 1. Warren DE
    2. Power JD
    3. Bruss J
    4. Denburg NL
    5. Waldron EJ
    6. Sun H
    7. Petersen SE
    8. Tranel D

    (2014) Network measures predict neuropsychological outcome after brain injury

    PNAS 111:14247–14252.

    https://doi.org/10.1073/pnas.1322173111

    • PubMed
    • Google Scholar
74. 1. Yarkoni T
    2. Poldrack RA
    3. Nichols TE
    4. Van Essen DC
    5. Wager TD

    (2011) Large-scale automated synthesis of human functional neuroimaging data

    Nature Methods 8:665–670.

    https://doi.org/10.1038/nmeth.1635

    • PubMed
    • Google Scholar
75. 1. Yeo BTT
    2. Krienen FM
    3. Sepulcre J
    4. Sabuncu MR
    5. Lashkari D
    6. Hollinshead M
    7. Roffman JL
    8. Smoller JW
    9. Zöllei L
    10. Polimeni JR
    11. Fischl B
    12. Liu H
    13. Buckner RL

    (2011) The organization of the human cerebral cortex estimated by intrinsic functional connectivity

    Journal of Neurophysiology 106:1125–1165.

    https://doi.org/10.1152/jn.00338.2011

    • PubMed
    • Google Scholar
76. 1. Yeo BTT
    2. Krienen FM
    3. Eickhoff SB
    4. Yaakub SN
    5. Fox PT
    6. Buckner RL
    7. Asplund CL
    8. Chee MWL

    (2015) Functional Specialization and Flexibility in Human Association Cortex

    Cerebral Cortex 25:3654–3672.

    https://doi.org/10.1093/cercor/bhu217

    • PubMed
    • Google Scholar
77. 1. Zhou H
    2. Schafer RJ
    3. Desimone R

    (2016) Pulvinar-Cortex Interactions in Vision and Attention

    Neuron 89:209–220.

    https://doi.org/10.1016/j.neuron.2015.11.034

    • PubMed
    • Google Scholar

Author details

1. Kai Hwang

   1. Department of Psychological and Brain Sciences, The University of Iowa & The University of Iowa College of Medicine, Iowa City, United States
   2. Cognitive Control Collaborative, The University of Iowa & The University of Iowa College of Medicine, Iowa City, United States
   3. Iowa Neuroscience Institute, The University of Iowa & The University of Iowa College of Medicine, Iowa City, United States
   4. Department of Psychiatry, The University of Iowa & The University of Iowa College of Medicine, Iowa City, United States

   Contribution

   Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Resources, Software, Supervision, Validation, Visualization, Writing - original draft, Writing – review and editing

   For correspondence

   kai-hwang@uiowa.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-1064-7815

2. James M Shine

   Brain and Mind Center, The University of Sydney, Sydney, Australia

   Contribution

   Formal analysis, Methodology, Resources, Writing - original draft, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-1762-5499

3. Joel Bruss

   1. Iowa Neuroscience Institute, The University of Iowa & The University of Iowa College of Medicine, Iowa City, United States
   2. Department of Neurology, The University of Iowa & The University of Iowa College of Medicine, Iowa City, United States
   3. Department of Pediatrics, The University of Iowa & The University of Iowa College of Medicine, Iowa City, United States

   Contribution

   Data curation, Formal analysis, Methodology, Software, Writing – review and editing

   Competing interests

   No competing interests declared

4. Daniel Tranel

   1. Department of Psychological and Brain Sciences, The University of Iowa & The University of Iowa College of Medicine, Iowa City, United States
   2. Iowa Neuroscience Institute, The University of Iowa & The University of Iowa College of Medicine, Iowa City, United States
   3. Department of Neurology, The University of Iowa & The University of Iowa College of Medicine, Iowa City, United States

   Contribution

   Data curation, Funding acquisition, Project administration, Resources, Supervision, Writing – review and editing

   Competing interests

   No competing interests declared

5. Aaron Boes

   1. Iowa Neuroscience Institute, The University of Iowa & The University of Iowa College of Medicine, Iowa City, United States
   2. Department of Psychiatry, The University of Iowa & The University of Iowa College of Medicine, Iowa City, United States
   3. Department of Neurology, The University of Iowa & The University of Iowa College of Medicine, Iowa City, United States
   4. Department of Pediatrics, The University of Iowa & The University of Iowa College of Medicine, Iowa City, United States

   Contribution

   Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Writing - original draft, Writing – review and editing

   Competing interests

   No competing interests declared

• 1,605

  views

• 272

  downloads

• 34

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.