CoChat

Revisiting the global workspace orchestrating the hierarchical organization of the human brain

Open access
Deco, Gustavo, Vidaurre, Diego, Kringelbach, Morten L. · 2021 · Nature Human Behaviour

The DOI matches its Crossref record; 144 of the 144 references it cites were found in a registry or an index.

Abstract

A central challenge in neuroscience is how the brain organizes the information necessary to orchestrate behaviour. Arguably, this whole-brain orchestration is carried out by a core subset of integrative brain regions, a ‘global workspace’, but its constitutive regions remain unclear. We quantified the global workspace as the common regions across seven tasks as well as rest, in a common ‘functional rich club’. To identify this functional rich club, we determined the information flow between brain regions by means of a normalized directed transfer entropy framework applied to multimodal neuroimaging data from 1,003 healthy participants and validated in participants with retest data. This revealed a set of regions orchestrating information from perceptual, long-term memory, evaluative and attentional systems. We confirmed the causal significance and robustness of our results by systematically lesioning a generative whole-brain model. Overall, this framework describes a complex choreography of the functional hierarchical organization of the human brain.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41562-020-01003-6

Cites

144 of 144 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Overview of general theoretical framework. a, The causal bidirectional flow of information between any two brain regions is determined by computing the pairwise NDTE. The figure shows two arbitrary regions, X and Y, superimposed on the human brain, with different shades of colours marking different regions, and we show the timecourses of X and Y in blue and red, respectively. The statistical significance is determined at the individual level using the circular time-shifted surrogates method23 and at the group level using P level aggregation across individuals. b, The functional hierarchical organization is given by the full NDTE matrix, where the rows contain the target regions and the columns contain the source regions. For each brain region, the total incoming flow of information, Gin, is simply the sum of all sources (that is, the sum over the rows in the matrix). Similarly, the total outgoing flow of information, Gout, is the sum over all targets (that is, the sum over the columns). c, The FRIC is the smallest set of brain regions (highlighed in red) that integrate and orchestrate function in a given task. It can be identified as the most highly connected brain regions that (1) are more densely connected within themselves than to regions with lower connectivity, whilst (2) having the highest level of incoming directed flow (Gin) and (3) the lowest outgoing directed flow (Gout; see Methods). d, The global neuronal workspace must be relevant to all tasks and situations (highlighted with red circles for each task) and must therefore comprise the common FRIC members across many different tasks, that is, the intersection of FRICs from tasks and rest (highlighted with red circles on the right panel, which also shows networks lower in the hierarchy in different colours). e, To establish the causal importance of the FRIC, we fit a whole-brain model to the resting NDTE empirical data and extract the underlying effective connectivity (see Results and Methods). f, The whole-brain model is then systematically lesioned for regions belonging to the FRIC and compared with lesioning non-FRIC members. For illustration, the cartoon highlights the lesion sites with red dots and other regions with blue dots. Overall, this confirms the causal importance of these regions in the orchestration of the functional hierarchical organization of the human brain.
Fig. 1 Overview of general theoretical framework. a, The causal bidirectional flow of information between any two brain regions is determined by computing the pairwise NDTE. The figure shows two arbitrary regions, X and Y, superimposed on the human brain, with different shades of colours marking different regions, and we show the timecourses of X and Y in blue and red, respectively. The statistical significance is determined at the individual level using the circular time-shifted surrogates method23 and at the group level using P level aggregation across individuals. b, The functional hierarchical organization is given by the full NDTE matrix, where the rows contain the target regions and the columns contain the source regions. For each brain region, the total incoming flow of information, Gin, is simply the sum of all sources (that is, the sum over the rows in the matrix). Similarly, the total outgoing flow of information, Gout, is the sum over all targets (that is, the sum over the columns). c, The FRIC is the smallest set of brain regions (highlighed in red) that integrate and orchestrate function in a given task. It can be identified as the most highly connected brain regions that (1) are more densely connected within themselves than to regions with lower connectivity, whilst (2) having the highest level of incoming directed flow (Gin) and (3) the lowest outgoing directed flow (Gout; see Methods). d, The global neuronal workspace must be relevant to all tasks and situations (highlighted with red circles for each task) and must therefore comprise the common FRIC members across many different tasks, that is, the intersection of FRICs from tasks and rest (highlighted with red circles on the right panel, which also shows networks lower in the hierarchy in different colours). e, To establish the causal importance of the FRIC, we fit a whole-brain model to the resting NDTE empirical data and extract the underlying effective connectivity (see Results and Methods). f, The whole-brain model is then systematically lesioned for regions belonging to the FRIC and compared with lesioning non-FRIC members. For illustration, the cartoon highlights the lesion sites with red dots and other regions with blue dots. Overall, this confirms the causal importance of these regions in the orchestration of the functional hierarchical organization of the human brain.CC BY · PubMed Central
Fig. 2 Comparing functional and structural hierarchical organization. a, Functional hierarchy shown as cortical renderings of each of the incoming (Gin), outgoing (Gout) and total (Gtot) directional information flow computed from the NDTE matrix of 1,003 HCP participants (see Fig. 1 and Methods) in the Glasser parcellation (using renderings of cortical flattening and 3D renderings with midline, right, left, top and bottom views). As shown, the outgoing information, Gout, is highest in sensory areas, while the incoming information, Gin, is highest in higher-order, integrative transmodal areas. b, The structural hierarchy shown for myelination of brain regions (myelin-weighted T1w/T2w) with the same renderings at the voxel level (top) and in the Glasser parcellation (bottom). c, Scatterplots between the functional hierarchy (Gin, Gout and Gtot) and structural hierarchy (myelination). The linear correlations are shown by the red line (with standard error in shaded gray) overlaid on the scatterplots (see Results for r values). This shows that myelination is highly correlated with Gtot and mainly driven by correlation with Gout. On the other hand, there is a much lower correlation with the incoming flow, that is, an integrative measure of Gin. This means that the static measure of myelination is likely to mostly reflect the driving flow in sensory areas, but provides much less information on integrative areas. d, The same as a but for the DK80 parcellation. e, Myelination in the DK80 parcellation. f, Scatterplots between the functional hierarchy (Gin, Gout and Gtot) and structural hierarchy (myelination) in the DK80 parcellation. The linear correlations are shown by the red line (with standard error in shaded gray) overlaid on the scatterplots (see Results for r values). The results in this coarser-scale DK80 parcellation are fully consistent with the finer-scale Glasser parcellation.
Fig. 2 Comparing functional and structural hierarchical organization. a, Functional hierarchy shown as cortical renderings of each of the incoming (Gin), outgoing (Gout) and total (Gtot) directional information flow computed from the NDTE matrix of 1,003 HCP participants (see Fig. 1 and Methods) in the Glasser parcellation (using renderings of cortical flattening and 3D renderings with midline, right, left, top and bottom views). As shown, the outgoing information, Gout, is highest in sensory areas, while the incoming information, Gin, is highest in higher-order, integrative transmodal areas. b, The structural hierarchy shown for myelination of brain regions (myelin-weighted T1w/T2w) with the same renderings at the voxel level (top) and in the Glasser parcellation (bottom). c, Scatterplots between the functional hierarchy (Gin, Gout and Gtot) and structural hierarchy (myelination). The linear correlations are shown by the red line (with standard error in shaded gray) overlaid on the scatterplots (see Results for r values). This shows that myelination is highly correlated with Gtot and mainly driven by correlation with Gout. On the other hand, there is a much lower correlation with the incoming flow, that is, an integrative measure of Gin. This means that the static measure of myelination is likely to mostly reflect the driving flow in sensory areas, but provides much less information on integrative areas. d, The same as a but for the DK80 parcellation. e, Myelination in the DK80 parcellation. f, Scatterplots between the functional hierarchy (Gin, Gout and Gtot) and structural hierarchy (myelination) in the DK80 parcellation. The linear correlations are shown by the red line (with standard error in shaded gray) overlaid on the scatterplots (see Results for r values). The results in this coarser-scale DK80 parcellation are fully consistent with the finer-scale Glasser parcellation.CC BY · PubMed Central
Fig. 3 Functional hierarchy across different tasks and rest. a, Cortical renderings of all seven tasks and rest for the incoming (Gin), outgoing (Gout) and total (Gtot) directional information flow computed from the NDTE matrix of 1,003 HCP participants (see Fig. 1 and Methods) in the DK80 parcellation (using 3D views from the side and midline). b, Matrices of the comparison of incoming (Gin), outgoing (Gout) and total (Gtot) directional information flows in the seven tasks and rest. As can be clearly seen, the Gin matrix and the renderings of the incoming flows of information (receivers) are significantly different between tasks and rest. This suggests that different tasks process incoming flows differently. This is in contrast to the Gout matrix and the renderings of the outgoing flows of information (drivers), which are similar. This shows that sensory areas are consistently driving the information flow. Interestingly, as can be seen from the Gtot matrix, the total processing of information flow is more similar within the seven tasks than compared with rest, suggesting the extrinsic, sensory nature of task processing compared with the intrinsic nature of resting-state processing.
Fig. 3 Functional hierarchy across different tasks and rest. a, Cortical renderings of all seven tasks and rest for the incoming (Gin), outgoing (Gout) and total (Gtot) directional information flow computed from the NDTE matrix of 1,003 HCP participants (see Fig. 1 and Methods) in the DK80 parcellation (using 3D views from the side and midline). b, Matrices of the comparison of incoming (Gin), outgoing (Gout) and total (Gtot) directional information flows in the seven tasks and rest. As can be clearly seen, the Gin matrix and the renderings of the incoming flows of information (receivers) are significantly different between tasks and rest. This suggests that different tasks process incoming flows differently. This is in contrast to the Gout matrix and the renderings of the outgoing flows of information (drivers), which are similar. This shows that sensory areas are consistently driving the information flow. Interestingly, as can be seen from the Gtot matrix, the total processing of information flow is more similar within the seven tasks than compared with rest, suggesting the extrinsic, sensory nature of task processing compared with the intrinsic nature of resting-state processing.CC BY · PubMed Central
Fig. 4 Identifying the GW as the intersection of FRICs for rest and seven tasks. a, We computed the functional hierarchical organization of all seven tasks (emotion, gambling, language, motor, relational, social and working memory) and rest for the HCP participants. This allows us to compute the FRIC as the set of regions that define a ‘club’ of functional hubs characterized by a tendency to be more densely functionally connected among themselves than to other brain regions from where they receive integrative information (see Methods). As can be seen, these FRICs are similar but not identical across tasks and rest. b, We compute the regions in the GW as the intersection of the FRIC members across all possible tasks and resting state. Here, we used the maximal amount of tasks available to provide a reliable estimate of the GW. At the bottom of the figure, we show a rendering of the cortical and subcortical regions in the GW. The FRIC regions for all seven tasks and rest defining the GW are the following eight brain regions: left precuneus, left nucleus accumbens, left putamen, left posterior cingulate cortex, right hippocampus, right amygdala and left and right isthmus cingulate. Lowering the threshold for participation to more than six FRICs adds two further regions: right nucleus accumbens and right posterior cingulate (in seven FRICs) and left and right rostral anterior cingulate (in six FRICs). Further lowering the threshold to four FRICs includes another three brain regions: left amygdala and left globus pallidus internus (in five FRICs) and left parahippocampal cortex (in four FRICs). c, These regions fit well with the idea suggested by Dehaene and Changeux that the GW is ideally placed for integrating information from perceptual (PRESENT), long-term memory (PAST), evaluative (VALUE) and attentional (FOCUSING) systems.
Fig. 4 Identifying the GW as the intersection of FRICs for rest and seven tasks. a, We computed the functional hierarchical organization of all seven tasks (emotion, gambling, language, motor, relational, social and working memory) and rest for the HCP participants. This allows us to compute the FRIC as the set of regions that define a ‘club’ of functional hubs characterized by a tendency to be more densely functionally connected among themselves than to other brain regions from where they receive integrative information (see Methods). As can be seen, these FRICs are similar but not identical across tasks and rest. b, We compute the regions in the GW as the intersection of the FRIC members across all possible tasks and resting state. Here, we used the maximal amount of tasks available to provide a reliable estimate of the GW. At the bottom of the figure, we show a rendering of the cortical and subcortical regions in the GW. The FRIC regions for all seven tasks and rest defining the GW are the following eight brain regions: left precuneus, left nucleus accumbens, left putamen, left posterior cingulate cortex, right hippocampus, right amygdala and left and right isthmus cingulate. Lowering the threshold for participation to more than six FRICs adds two further regions: right nucleus accumbens and right posterior cingulate (in seven FRICs) and left and right rostral anterior cingulate (in six FRICs). Further lowering the threshold to four FRICs includes another three brain regions: left amygdala and left globus pallidus internus (in five FRICs) and left parahippocampal cortex (in four FRICs). c, These regions fit well with the idea suggested by Dehaene and Changeux that the GW is ideally placed for integrating information from perceptual (PRESENT), long-term memory (PAST), evaluative (VALUE) and attentional (FOCUSING) systems.CC BY · PubMed Central
Fig. 5 Establishing causal significance by lesioning whole-brain model. a, To establish the causal mechanistic functional relevance of the FRIC brain regions orchestrating the functional hierarchical organization of the resting state, we created a whole-brain model to fit the empirical NDTE flow matrix. b, Lesioning all FRIC compared with non-FRIC members in this whole-brain model causes a significant breakdown in information flow (see Results for significance testing). The correlation between empirical (Emp) and simulated NDTE matrices is maximally affected when FRIC members are lesioned compared both with non-FRIC members that were randomly selected (nFRIC) and with non-FRIC members that were selected using a biased method in which we chose the regions with the top 50% Gin values (nFRICbias). We also compared FRIC versus nFRIC and FRIC versus nFRICbias. c, Similarly, we found significant differences in the number of FRIC members detected when lesioning the original empirical FRIC members compared with empirical non-FRIC members, Emp versus nFRIC and Emp versus nFRICbias. We also compared FRIC versus nFRIC and FRIC versus nFRICbias. This is mechanistic evidence for the significance of FRIC in orchestrating functional organization. Emp, empirical; FRIC, functional rich club; nFRIC, nFRICbias, non-functional rich club members.
Fig. 5 Establishing causal significance by lesioning whole-brain model. a, To establish the causal mechanistic functional relevance of the FRIC brain regions orchestrating the functional hierarchical organization of the resting state, we created a whole-brain model to fit the empirical NDTE flow matrix. b, Lesioning all FRIC compared with non-FRIC members in this whole-brain model causes a significant breakdown in information flow (see Results for significance testing). The correlation between empirical (Emp) and simulated NDTE matrices is maximally affected when FRIC members are lesioned compared both with non-FRIC members that were randomly selected (nFRIC) and with non-FRIC members that were selected using a biased method in which we chose the regions with the top 50% Gin values (nFRICbias). We also compared FRIC versus nFRIC and FRIC versus nFRICbias. c, Similarly, we found significant differences in the number of FRIC members detected when lesioning the original empirical FRIC members compared with empirical non-FRIC members, Emp versus nFRIC and Emp versus nFRICbias. We also compared FRIC versus nFRIC and FRIC versus nFRICbias. This is mechanistic evidence for the significance of FRIC in orchestrating functional organization. Emp, empirical; FRIC, functional rich club; nFRIC, nFRICbias, non-functional rich club members.CC BY · PubMed Central
Extended Data Fig. 1 Functional hierarchy of resting state found using NDTE rendered on Glasser and DK80 parcellations. At the top is shown the full cortical voxelbased renderings of myelin (T1w/T2w). Below is shown renderings for both the Glasser and DK80 parcellations of myelin and the NDTE measures (Gin, Gout and Gtot) in 1003 HCP participants.
Extended Data Fig. 1 Functional hierarchy of resting state found using NDTE rendered on Glasser and DK80 parcellations. At the top is shown the full cortical voxelbased renderings of myelin (T1w/T2w). Below is shown renderings for both the Glasser and DK80 parcellations of myelin and the NDTE measures (Gin, Gout and Gtot) in 1003 HCP participants.CC BY · PubMed Central
Extended Data Fig. 2 Validating NDTE framework using MEG from 89 HCP participants. We used the NDTE framework on MEG resting state data from 89 HCP participants (each having three resting state sessions) and extracted the timeseries from the 62 cortical regions of the DK80 parcellation (see Methods). On the left is shown the scatterplots for Gin and Gout for fMRI resting state data versus myelin (T1w/T2w). On the right is shown the same scatterplots for Gin and Gout for MEG resting state data versus myelin. The Pearson correlation values are r(124)=0.04 for Gin (14-22 Hz, window size 1000 ms, not significant, 95% CI = -0.21–0.29), r(124)=0.48 for Gout (22.5-30.5 Hz, window size 500 ms, P < 0.001, 95% CI = 0.26–0.65). B) The figure shows the scatterplots between Gin and Gout for MEG and fMRI data with correlation values of r(124)=0.40 for Gin (P < 0.001, 95% CI = 0.17–0.59) and r(124)=0.41 for Gout (P < 0.001, 95% CI = 0.18–0.60).
Extended Data Fig. 2 Validating NDTE framework using MEG from 89 HCP participants. We used the NDTE framework on MEG resting state data from 89 HCP participants (each having three resting state sessions) and extracted the timeseries from the 62 cortical regions of the DK80 parcellation (see Methods). On the left is shown the scatterplots for Gin and Gout for fMRI resting state data versus myelin (T1w/T2w). On the right is shown the same scatterplots for Gin and Gout for MEG resting state data versus myelin. The Pearson correlation values are r(124)=0.04 for Gin (14-22 Hz, window size 1000 ms, not significant, 95% CI = -0.21–0.29), r(124)=0.48 for Gout (22.5-30.5 Hz, window size 500 ms, P < 0.001, 95% CI = 0.26–0.65). B) The figure shows the scatterplots between Gin and Gout for MEG and fMRI data with correlation values of r(124)=0.40 for Gin (P < 0.001, 95% CI = 0.17–0.59) and r(124)=0.41 for Gout (P < 0.001, 95% CI = 0.18–0.60).CC BY · PubMed Central
Extended Data Fig. 3 Validation of NDTE framework on test-retest data for task and rest. We validated the NDTE framework on test-retest data and found very high reproducibility (of up to a 0.97 correlation) using only a subsample of 45 participants of the 1003 participants. The correlation values for the rest-retest are the following: EMOTION: r(3160)=0.866, P < 0.001, 95% CI = 0.860–0.872); GAMBLING: r(3160)=0.898, P < 0.001, 95% CI = 0.893–0.903); LANGUAGE: r(3160)=0.907, P < 0.001, 95% CI = 0.902–0.911); MOTOR: r(3160)=0.907, P < 0.001, 95% CI = 0.903–0.912); RELATIONAL: r(3160)=0.866, P < 0.001, 95% CI = 0.860–0.872); SOCIAL: r(3160)=0.924, P < 0.001, 95% CI = 0.920–0.927); WM r(3160)=0.939, P < 0.001, 95% CI = 0.936–0.942); REST: r(3160)=0.970, P < 0.001, 95% CI = 0.968–0.971).
Extended Data Fig. 3 Validation of NDTE framework on test-retest data for task and rest. We validated the NDTE framework on test-retest data and found very high reproducibility (of up to a 0.97 correlation) using only a subsample of 45 participants of the 1003 participants. The correlation values for the rest-retest are the following: EMOTION: r(3160)=0.866, P < 0.001, 95% CI = 0.860–0.872); GAMBLING: r(3160)=0.898, P < 0.001, 95% CI = 0.893–0.903); LANGUAGE: r(3160)=0.907, P < 0.001, 95% CI = 0.902–0.911); MOTOR: r(3160)=0.907, P < 0.001, 95% CI = 0.903–0.912); RELATIONAL: r(3160)=0.866, P < 0.001, 95% CI = 0.860–0.872); SOCIAL: r(3160)=0.924, P < 0.001, 95% CI = 0.920–0.927); WM r(3160)=0.939, P < 0.001, 95% CI = 0.936–0.942); REST: r(3160)=0.970, P < 0.001, 95% CI = 0.968–0.971).CC BY · PubMed Central
Extended Data Fig. 4 Functional hierarchy across different tasks and rest in Brainnetome246 parcellation. a, NDTE matrices for all seven tasks and resting state from all HCP participants in the BN246 parcellation. b, Cortical renderings of the incoming (Gin), outgoing (Gout) and total (Gtot) directional information flow computed from the NDTE matrix using 3D views from the side and midline (similar to Fig. 3). c, Matrices of the comparison of incoming (Gin), outgoing (Gout) and total (Gtot) directional information flow in the seven tasks and rest. Similar to the results for the DK80 and Glasser378 parcellations, the renderings of the incoming flow of information (receivers) are significantly different between tasks and rest.
Extended Data Fig. 4 Functional hierarchy across different tasks and rest in Brainnetome246 parcellation. a, NDTE matrices for all seven tasks and resting state from all HCP participants in the BN246 parcellation. b, Cortical renderings of the incoming (Gin), outgoing (Gout) and total (Gtot) directional information flow computed from the NDTE matrix using 3D views from the side and midline (similar to Fig. 3). c, Matrices of the comparison of incoming (Gin), outgoing (Gout) and total (Gtot) directional information flow in the seven tasks and rest. Similar to the results for the DK80 and Glasser378 parcellations, the renderings of the incoming flow of information (receivers) are significantly different between tasks and rest.CC BY · PubMed Central
Extended Data Fig. 5 Approximation of the Global Workspace in the fine-scale Brainnetome246 parcellation for comparison with formal definition of the Global Workspace in the DK80 parcellation. a, The computational complexity of the FRIC algorithm makes it inappropriate for finer-scale parcellations so we developed an approximation, computing the intersection of the top Gin regions across rest and seven tasks. This is only an approximation since it does not conform fully to our definition of the global workspace, separate from the rest of the brain, which similar to a ‘club’ of functional hubs are characterized by a tendency to be more densely functionally connected among themselves than to other brain regions from where they receive integrative information. b, Nevertheless, this rough approximation shows a remarkable overlap with the Global Workspace derived from our formal definition in the FRIC algorithm applied to the DK80 parcellation (see Results and Fig. 4).
Extended Data Fig. 5 Approximation of the Global Workspace in the fine-scale Brainnetome246 parcellation for comparison with formal definition of the Global Workspace in the DK80 parcellation. a, The computational complexity of the FRIC algorithm makes it inappropriate for finer-scale parcellations so we developed an approximation, computing the intersection of the top Gin regions across rest and seven tasks. This is only an approximation since it does not conform fully to our definition of the global workspace, separate from the rest of the brain, which similar to a ‘club’ of functional hubs are characterized by a tendency to be more densely functionally connected among themselves than to other brain regions from where they receive integrative information. b, Nevertheless, this rough approximation shows a remarkable overlap with the Global Workspace derived from our formal definition in the FRIC algorithm applied to the DK80 parcellation (see Results and Fig. 4).CC BY · PubMed Central
Extended Data Fig. 6 Comparing the anatomical rich club with the functional rich club (FRIC) for tasks and their intersection (Global Workspace). a, Using the rich club algorithm, we computed the anatomical rich club on the structural connectivity matrix for DK80. The figure shows a rendering of the ten brain regions in the rich club: left and right precuneus, left and right thalamus, left and right superior frontal, left superior parietal, left insula and right isthmus cingulate cortices. b, A comparison with the global workspace (shown here as well in Fig. 4 and described in details in the results) shows that they overlap in the left precuneus and the right isthmus cingulate cortex. Given that the anatomical rich club cannot change over time with functional demands, the lack of further overlap is not surprising.
Extended Data Fig. 6 Comparing the anatomical rich club with the functional rich club (FRIC) for tasks and their intersection (Global Workspace). a, Using the rich club algorithm, we computed the anatomical rich club on the structural connectivity matrix for DK80. The figure shows a rendering of the ten brain regions in the rich club: left and right precuneus, left and right thalamus, left and right superior frontal, left superior parietal, left insula and right isthmus cingulate cortices. b, A comparison with the global workspace (shown here as well in Fig. 4 and described in details in the results) shows that they overlap in the left precuneus and the right isthmus cingulate cortex. Given that the anatomical rich club cannot change over time with functional demands, the lack of further overlap is not surprising.CC BY · PubMed Central
Extended Data Fig. 7 Estimation of structural connectivity (SC) matrices for DK80 using standard and special HCP datasets. a, The ‘standard SC matrix’ was extract from the standard HCP multi-shell dMRI data from 985 HCP participants using state-of-the-art tractography methods (see Methods). b, The ‘special SC matrix’ was estimated using an even better state-of-the-art multi-shell dMRI protocol taking 89 minutes for each of 32 HCP participants at the MGH centre. Please note the increased number of inter-hemispheric connections (upper right and lower left quadrants of the special SC matrix compared to the standard SC matrix). c, The two SC matrices are highly correlated as shown in the scatterplot (r(3160)=0.85, P < 0.001, 95% CI = 0.84–0.86).
Extended Data Fig. 7 Estimation of structural connectivity (SC) matrices for DK80 using standard and special HCP datasets. a, The ‘standard SC matrix’ was extract from the standard HCP multi-shell dMRI data from 985 HCP participants using state-of-the-art tractography methods (see Methods). b, The ‘special SC matrix’ was estimated using an even better state-of-the-art multi-shell dMRI protocol taking 89 minutes for each of 32 HCP participants at the MGH centre. Please note the increased number of inter-hemispheric connections (upper right and lower left quadrants of the special SC matrix compared to the standard SC matrix). c, The two SC matrices are highly correlated as shown in the scatterplot (r(3160)=0.85, P < 0.001, 95% CI = 0.84–0.86).CC BY · PubMed Central
Extended Data Fig. 8 Alternative measure of hierarchy. Inspired by the neuroanatomical research by Markov and colleagues, we computed a similar measure of hierarchy as the fraction of feedforward and feedback (FF) organisation. We computed this for the seven tasks and for resting state (top row). As can be seen from the correlation matrices, this FF Hierarchy measure is similar to Gout for all seven tasks (mean correlation is 0.81 and the mean p-values are p < 0.001). On the other hand FF Hierarchy is only non-significantly correlated with Gin, the integrative measure of incoming information flow (mean correlation is −0.02 and the mean p-value is p = 0.243).
Extended Data Fig. 8 Alternative measure of hierarchy. Inspired by the neuroanatomical research by Markov and colleagues, we computed a similar measure of hierarchy as the fraction of feedforward and feedback (FF) organisation. We computed this for the seven tasks and for resting state (top row). As can be seen from the correlation matrices, this FF Hierarchy measure is similar to Gout for all seven tasks (mean correlation is 0.81 and the mean p-values are p < 0.001). On the other hand FF Hierarchy is only non-significantly correlated with Gin, the integrative measure of incoming information flow (mean correlation is −0.02 and the mean p-value is p = 0.243).CC BY · PubMed Central
Extended Data Fig. 9 Validation of NDTE framework. Following the work of Seth, Chorley and Barrett1, we constructed a directional network with five nodes and simulated BOLD timeseries by using a Hopf model. b, This network is represented by the simple directional coupling 5×5 matrix. The generative Hopf model was run 10 times with 1200 timepoints sampled at TR = 0.72 s. We then validated the NDTE framework by applying this to the simulated data (using the same lag of 10 and 100 surrogates as in the analysis of the empirical data). c, The figure shows the recovered NDTE uncorrected matrix without using surrogates and aggregation of p-values. d, In contrast, the figure shows the results of using surrogates and aggregation of p-values and corrected for multiple comparisons. The NDTE framework is clearly able to recover the original couplings.
Extended Data Fig. 9 Validation of NDTE framework. Following the work of Seth, Chorley and Barrett1, we constructed a directional network with five nodes and simulated BOLD timeseries by using a Hopf model. b, This network is represented by the simple directional coupling 5×5 matrix. The generative Hopf model was run 10 times with 1200 timepoints sampled at TR = 0.72 s. We then validated the NDTE framework by applying this to the simulated data (using the same lag of 10 and 100 surrogates as in the analysis of the empirical data). c, The figure shows the recovered NDTE uncorrected matrix without using surrogates and aggregation of p-values. d, In contrast, the figure shows the results of using surrogates and aggregation of p-values and corrected for multiple comparisons. The NDTE framework is clearly able to recover the original couplings.CC BY · PubMed Central
Extended Data Fig. 10 The estimation of the NDTE matrix does not depend on the filtering. a, We computed the NDTE matrix in the DK80 parcellation from 100 unrelated HCP participants using a standard filter of 0.01-0.2 Hz on the underlying BOLD data. b, Similarly, we computed the NDTE matrix for the same participants with no filter. c, The correlation between the two NDTE matrices for the filtered and non-filtered version is almost identical (r(3160)=0.972, P < 0.001, 95% CI = 0.971–0.973).
Extended Data Fig. 10 The estimation of the NDTE matrix does not depend on the filtering. a, We computed the NDTE matrix in the DK80 parcellation from 100 unrelated HCP participants using a standard filter of 0.01-0.2 Hz on the underlying BOLD data. b, Similarly, we computed the NDTE matrix for the same participants with no filter. c, The correlation between the two NDTE matrices for the filtered and non-filtered version is almost identical (r(3160)=0.972, P < 0.001, 95% CI = 0.971–0.973).CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 144 references

Research works printed but not linked: 7

  1. Norman, D. A. & Shallice, T. in Consciousness and Self-Regulation (eds Davidson, R. J., Schwartz, G. E. & Shapiro, D.) 1–18 (Plenum Press, 1980).
  2. Stouffer, S. A., Suchman, E. A., Devinney, L. C., Star, S. A. & Williams, R. M. Jr The American Soldier: Adjustment during Army Life (Studies in Social Psychology in World War II) 1 (Princeton Univ. Press, 1949).
  3. Kantz, H. & Schreiber, T. Nonlinear Time Series Analysis (Cambridge Univ. Press, 1997).
  4. Fornito, A., Zalesky, A. & Bullmore, E. Fundamentals of Brain Network Analysis (Academic Press, 2016).
  5. Kuznetsov, Y. A. Elements of Applied Bifurcation Theory (Springer, 1998).
  6. Deco, G. et al. Single or multiple frequency generators in on-going brain activity: a mechanistic whole-brain model of empirical MEG data. NeuroImage 152, 538–550 (2017). · DOI · Paper page
  7. Deco, G., Kringelbach, M. L., Jirsa, V. K. & Ritter, P. The dynamics of resting fluctuations in the brain: metastability and its dynamical cortical core. Sci. Rep. 7, 3095 (2017). · DOI · Paper page

Cited by

See all 217

Metadata

Authors
Deco, G., Vidaurre, D., & Kringelbach, M. L.
Affiliations
  • Institució Catalana de Recerca i Estudis Avançats (ES)
  • Aarhus University (DK)
Corresponding author
  • Deco, Gustavo
  • Kringelbach, Morten L.
Venue
Nature Human Behaviour
Published
2021-01-04
Volume
5
Issue
4
Pages
497-511
ISSN
2397-3374 (ISSN-L)
OpenAlex
W3118861474
Cited by
162
References
141
Citations by year
2026: 34 · 2025: 31 · 2024: 34 · 2023: 25 · 2022: 16 · 2021: 19
FWCI
12.36
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Human Behaviour (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
  • MPG.PuRe (Max Planck Society) (submittedVersion, other-oa)
  • Repositori digital de la UPF (Universitat Pompeu Fabra) (submittedVersion, cc-by)
Funders
  • Ministerio de Ciencia, Innovación y Universidades (PID2019)
  • National Research Foundation (DNRF117)
  • Danmarks Grundforskningsfond (615539, DNRF117)
  • Novo Nordisk Fonden (NNF19OC-0054895)
  • Horizon 2020 Framework Programme (HBP SGA3), #945539, EU‐H2020, PID2019)
  • Agència de Gestió d'Ajuts Universitaris i de Recerca (945539, 2017/SGR/1545, 2017SGR, SGR 1545)
  • Agencia Estatal de Investigación (945539, H2020, PID2019)
Topics
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Functional Brain Connectivity Studies
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Neural dynamics and brain function
  • Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Advanced Neuroimaging Techniques and Applications
Keywords
global workspace, hierarchical brain organization, whole-brain model, multimodal neuroimaging, long-term memory, attention, perception, resting state
MeSH terms
Adult, Brain, Brain Mapping, Cerebral Cortex, Cognition, Female, Humans, Male, Models, Neurological, Nerve Net, Neural Pathways, Neurosciences, Work, Neural Networks, Computer

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Revisiting the global workspace orchestrating the hierarchical organization of the human brain. https://app.cochat.ai/papers/10.1038/s41562-020-01003-6
The paper
Deco, G., Vidaurre, D., & Kringelbach, M. L. (2021). Revisiting the global workspace orchestrating the hierarchical organization of the human brain. Nature Human Behaviour. https://doi.org/10.1038/s41562-020-01003-6

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.