A Precuneal Causal Loop Mediates External and Internal Information Integration in the Human Brain

TL;DR: This study provides evidence that the medial posterior part of the DMN may drive interactions between large-scale networks, potentially allowing access to stored representations for moment to moment interpretation of an ever-changing environment.

Abstract: Human brains interpret external stimuli based on internal representations. One untested hypothesis is that the default-mode network (DMN) while responsible for internally oriented cognition can also encode externally oriented information. The unique neuroanatomical and functional fingerprint of the posterior part of the DMN supports a prominent role for the precuneus in this process. By utilising imaging data during two tasks from 100 participants, we found that the precuneus is functionally divided into dorsal and ventral subdivisions, each one differentially connecting to internally and externally oriented networks. The strength and direction of their connectivity is modulated by task difficulty in a manner dictated by the balance of internal versus external cognitive demands. Our study provides evidence that the medial posterior part of the DMN may drive interactions between large-scale networks, potentially allowing access to stored representations for moment to moment interpretation of an ever-changing environment.

Summary

The DMN during tasks: A meta-analytic perspective

• The authors first validated that the DMN is indeed involved during goal-directed tasks, as findings regarding DMN's involvement during tasks are still disputed (Fransson, 2006) .
• CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
• For the easy versus difficult contrast, wide-spread deactivations were found in the DMN.
• Forward inference shows that DMN subregions are active in the tasks.

Differential connectivity between v/dPCu and ICNs

• To explore further the differential response of the dorsal and ventral aspects of the PCu to task demands and the role they may serve in processing internal and external information, the authors next focused on the d/vPCu's connectivity, including structural, resting-state, task-state and effective connectivity.
• The authors demonstrated that there is a disparity in the v/dPCu's structural connectivity (SC) and resting-state (rs) FC (connectivity results for each seed are presented in Supplementary Materials) and the difference follows the pattern of internally and externally oriented cognitive function.
• The authors achieved this by computing the v/dPCu's endogenous taskstate FC (tsFC), i.e., the correlation of timeseries throughout the course of the two tasks, after regressing out the event-related haemodynamic response function (HRF).
• Again, contrary to the well-known DMN "anti-correlation" argument, the authors found a large cluster of positive tsFC with the d/vPCu, centring at the seeds themselves, and covering regions within and beyond the DMN (See supplementary materials for the composite of network domains of the v/dPCu's tsFC).

Cognitive demands modulate the effective coupling between the v/dPCu and internally/externally oriented networks

• The connectivity profiles of the d/vPCu as established so far suggest that the PCu overall has the structural framework necessary and may functionally serve as a platform connecting internal and externally related information.
• CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
• Using PPI the authors found that as cognitive demand increased, eFC increased between the vPCu and IoN and between the dPCu and EoN.
• On the contrary, the vPCu was more connected to the rest of the DMN and interoceptive regions in difficult (vs. easy) conditions, and more connected with visual and primary sensory networks in easy (vs. difficult) conditions .
• Among the "one-state", "deterministic" DCMs, the third model wins over others with consistently higher posterior probability and higher exceedance of model evidence.

Model

• The "exchange" model, which specifies an effective connectivity (EC) modulation of dPCu à vPCu in the easy condition and vPCu à dPCu in the difficult condition; (2) the "forward" model, which specified the EC of dPCu à vPCu to be modulated in both difficult and easy conditions.
• The magnitude of its model evidence did not allow us to draw a safe conclusion of favouring it over the other model (K. E. Stephan et al., 2010) .

DISCUSSION

• The present study investigated the functional differentiation of the DMN and proposed a role for the precuneus (PCu) in mediating external and internal information binding.
• CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
• Braga and Buckner (2017) proposed the DMN encompasses several networks, the spatial pattern of which differs between individuals.
• Based on previous work and their current study, the authors believe the PCu is central for linking the function of the DMN subdivisions, thus playing a key role in integrating the brain's information flow from all sources.

NeuroSynth Meta-analysis of fMRI studies:

• When this work was carried out, the latest Neurosynth database contained 14,371 neuroimaging studies (https://github.com/neurosynth/neurosynth-data), associated with more than 3,200 text-based features and over 410,000 activation peaks that span a wide range of published neuroimaging studies.
• Since their focus is DMN functionality during tasks the authors searched for activation coordinates associated with attentional and executive tasks.
• On the other hand, reverse inference relies on the probability of the specified term being frequently discussed alongside a specific activation [i.e., P(Term|Activation)], thus showing the regions that are selectively associated with the term.
• Key parameters were based on the default values in the publicly available NeuroSynth toolbox (https://github.com/neurosynth/neurosynth).
• A frequency cut-off of 0.001 for article words was used to determine if a study used the term incidentally or purposely.

Tasks selected:

• The authors selected the Relational Processing (RP) task and the N-back working memory (N-back) task from the human connectome project (HCP), as these have two .
• CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
• The copyright holder for this preprint this version posted August 21, 2020.
• Analyses for the two tasks were conducted independently first, and then results were averaged across the two tasks to emphasise domain-general brain activity.
• In the main text, domain-specific terms indicating a certain level of specific cognitive demand (such as 0-back and 2-back in the N-back task) were substituted with the general terms of "easy" and "difficult" to indicate the load of general cognitive abilities.

Image preprocessing:

• The ready-to-use HCP data has already been minimally preprocessed and quality-checked by the distributors (Glasser et al., 2013) and the authors carried out extra preprocessing steps with SPM12 (http//www.fil.ion.ucl.ac.uk/spm).
• Specifically, the data was smoothed with a Gaussian kernel of 6 mm FWHM (full-width half maximum) and no low-pass filtering was used as it might reduce signal strength and sensitivity.
• No global signal regression was used, for it may cause anti-correlation artefacts by shifting the distribution of FC towards negative values (Chai et al., 2012; Murphy et al., 2009) .
• To reduce the influence of non-neuronal confounds, eigenvalues were extracted from BOLD (Blood Oxygen Level Dependent) signals within cerebrospinal fluid (CSF) and white matter template masks and were regressed out from target signals during statistical testing.

Activation studies:

• Whole-brain activation was estimated using the standard SPM GLM approach.
• Only the trials with correct responses were explicitly modelled.
• Individual-level GLMs were modelled with difficulty level (2 levels) as the main effect, i.e. difficult (2-back or relational) & easy (0-back or match) conditions, with neuronal nuisance (i.e. CSF, white matter signals), session effects and 6 movement regressors as covariates.
• Group-level GLMs were then constructed to test the significance of the activation in the population level, with age and sex as covariates.

Selection of Regions of Interest (ROI)

• The authors were interested in the functionality of the PCu during attentional demanding goal-directed tasks; therefore, ROIs were selected based on their activation results from the N-back and RP tasks.
• Both activated and deactivated PCu regions associated with increased level of difficulty of the tasks were considered.
• To ensure the PCu region that the authors are considering is actually part of the DMN, they also superimposed it on the DMN canonical mask as defined by the Conn network atlas.
• The seed regions were selected by computing the spatial intersection of their task-related (de)activation clusters (exceeding the cluster-level family-wise-error/FWE-corrected P of 0.05), the PCu as defined by the Conn atlas and the DMN spatial localisation identified by the Conn atlas (Whitfield-Gabrieli & Nieto-Castanon, 2012) .
• Timeseries were extracted from seed regions, and their FC during the course of the experiment was calculated for the two tasks separately.

Functional connectivity studies:

• . CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
• The copyright holder for this preprint this version posted August 21, 2020.

Resting-state Seed-based FC of vPCu & dPCu

• Using SPM12, the authors calculated the FC as the partial correlation of the timeseries extracted from the seed regions (v/dPCu) with that from the rest of the brain, after controlling for the effect of non-neuronal confounds (estimated from white matter and CSF) and head movements.
• Baseline FC and differences between the FC of the (v)dPCu at the group level were estimated with one-sample and paired T-tests.

Structural Connectivity of vPCu & dPCu

• The acquisition, preprocessing the diffusion MRI (dMRI) images from the HCP and the generation of diffusion tensor maps has been detailed in published articles (Sotiropoulos et al., 2013) .
• Based on the diffusion tensor images the authors built probabilistic tractography in FSL5 (https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/FDT).
• The structural connectivity matrix for each individual was obtained using vPCu or dPCu as the seed and the grey matter as the termination mask.
• This matrix was then projected to a standard MNI brain space for further statistical parametric mapping analyses in SPM12.
• Similar to FC analyses, baseline and difference of grey-matter connections of the dPCu and vPCu at the group level, were calculated with one-sample.

Psycho-physiological interactions (PPI)

• PPI was used to evaluate by what amount the cognitive variables during tasks upregulate or downregulate the FC between the seed region and its functionallycoupled regions .
• The implementation of the PPI was similar to the above seed-based FC during tasks, but additionally the GLM model included an interaction term, i.e. a new variable created by dot multiplying the task-related BOLD response and the seed's timeseries.
• Individual-level PPIs were estimated separately for the N-back and the RP task, but the final results were averaged across tasks because the authors wanted to emphasize the common emerging patterns.
• To emphasise that common pattern, the authors then averaged the individual-level modulatory effects between the two tasks, and conducted another group inference based on the individual averaged beta values.
• Results of task-specific PPIs and taskaveraged PPI are both provided.

Anatomical labels and ICN identification based on significant clusters

• To make inferences about the cognitive function of significant regions, the authors used the ly defined ICN-BM network atlas (https://www.nitrc.org/projects/icn_atlas/) for identifying the intrinsic connectivity networks (ICN)s involved in the two tasks.
• The advantage of using the ICN-BM atlas was the nomenclature used not only corresponds to the well-known canonical resting-state connectivity networks, but also to task-based co-activation networks which were generated from a meta-analyses using the BrainMap (BM) dataset (Cole et al., 2016; Smith et al., 2009) .
• When reporting the PPI result, the authors also applied the The spatial overlap reported in the main article was calculated as the ratio of the number of activated voxels over the region/ICN volume which the voxels belong to (Kozák et al., 2017) .
• Other ways of representing the spatial overlap, such as calculating spatial overlaps by taking into account the effect size of each significant voxel, were found to generate similar results and are presented in the Supplementary Materials.
• For identifying the anatomical regions based on the coordinates, the authors adopted the toolboxes of GingerALE (http://brainmap.org/ale/) and Talairach Daemon (http://www.talairach.org/daemon.html) where the Brodmann areas were identified from.

Group analysis and multiple-comparison correction

• For all statistical parametric mapping analyses, random factor effects (RFX) were used (random effect being the intercept of within-subject GLM fitting) and inferences .
• The copyright holder for this preprint this version posted August 21, 2020.
• More stringent voxel-level thresholds were sometimes used because the statistical power in this study was very high and the conventional cluster-forming threshold of P-uncorrected = 0.001 resulted, on some occasions, on clusters that were too extensive to be anatomically meaningful.
• The family-wise error rate was controlled at the cluster level, and a threshold of P-corrected < .05 was used to determine significance among clusters.

Dynamic Causal Modelling (DCM) specification:

• Dynamic causal modelling (DCM) is a generative model in a Bayesian framework for inferring hidden neuronal states from observed fMRI measurements (K. E. Stephan et al., 2010) .
• CC-BY 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
• The copyright holder for this preprint this version posted August 21, 2020.
• ; https://doi.org/10.1101/2020.08.20.259846 doi: bioRxiv preprint 33 DCM model specified the endogenous connectivity between vPCu and dPCu to be bidirectional; and on top of that the authors modelled all possible configurations of how the task difficulty might influence the endogenous connectivity (Table 1 ).
• Based on that, the authors tried both the one-state, deterministic (the default) and the two-state, stochastic DCM class for modelling local neural dynamics.

DCM estimation:

• To determine the most likely model structure, the authors applied a fixed factor effect (FFX) Bayesian model selection (BMS) procedure to all 11 models estimated across all participants independently for the N-back and the RP task.
• The FFX was used as opposed to a random factor effect (RFX) because the authors hypothesised the mechanism to be general across all subjects.
• Finally, the model with the highest model evidence was.

Figures

Figure 5: PPI analyses for the two tasks revealed an inverse pattern of task modulated FC of the two seeds (v/dPCu’s), which is common to both tasks. (a) FC overviews for each task are shown in the circular connectogram plots. These were constructed by segmenting T-score PPI maps to anatomical regions as defined by the Harvard-Oxford cortical and subcortical structural atlases in the CONN toolbox (https://web.conn-toolbox.org/). Cerebellar regions are not

Figure 1: Meta-analysis and univariate analysis shows DMN engagement during the two tasks. (a) and (b) demonstrate the NeuroSynth meta-analysis results, using “attention*” or “execut*” as keywords to search for goal-directed tasks that require attention and executive function. (a) Forward inference shows that DMN subregions are active in the tasks. The forward inference map is produced by calculating the convergence of brain regions most consistently activated by certain cognitive processes. (b) Reverse inference shows that not many DMN regions are specifically associated with these tasks. The reverse inference map is calculated as the likelihood of a search term being used in a study given the presence of reported activation, and it reflects the brain activation specific to a certain cognitive process (Yarkoni et al., 2011). (c) and (d) show that activation of posterior DMN regions is associated with the N-back and RP task (from the HCP database). Warm/cold regions in the brain heatmap indicate higher/lower activity in the difficult condition compared to the easy condition. For panels (a) and (b) standardised Z scores and for panels (c) and (d) T-scores indicating activation strength are provided with colour scales. For highlighting activation in relation to the DMN, 3D renderings of the DMN are shown in shaded grey, on which our activated regions are superimposed. This was constructed by superimposing the Z-score maps (with the cut-off of 3) and T-scores (of

Figure 8: A conceptual predictive model centring at the medial posterior DMN. The medial posterior DMN behaves like a predictive system, with its dorsal part associated with an attentional system (comprising FPCN regions such as the IPS, middle and lateral frontal cortices) and its ventral part associate with a reactive system (comprising regions such as anterior insula) (Tops et al., 2014). The connectivity from each module is derived from our studies, with the pink colour indicating being stronger in difficult conditions, and blue colour indicating being stronger in easy conditions. The (directed) connectivity pattern from our result suggests that external and internal information is exchanged by the reciprocal loop between the vPCu and dPCu. This exchange of information might support the predictive coding theory which emphasises that external information is always inferred (rather than simply being represented from a bottom-up direction) based on an internal belief of the world (prediction).

Figure 7. Illustration of the model structure of the winning DCM model.

Table 1: The N-back and the RP task resulted in the same pattern of PCu fragmentation. The dorsal PCu (dPCu) and ventral PCu (vPCu) were identified as the regions responsive to different cognitive load: the dPCu was activated while the vPCu was deactivated by an increased cognitive demand in difficult versus easy conditions. The N-back task provided a statistically stronger demonstration (larger significant clusters) of this, possibly because the task had more trials i.e. bigger sample size (80 trials in the N-back, vs. 27 in the RP task).

Table 4: Model comparison result suggests that the causal influence from the vPCu to dPCu is modulated in easier conditions and then swapped from the dPCu to vPCu in more difficult conditions. The first 4 rows/aspects of the model structure specified 11 DCM models. The first four DCMs that adopted “one-state”, “deterministic” neural mass model won over the “twostate”, “stochastic” ones, according to the posterior probability at the group level (assuming a fixed factor effect). Among the “one-state”, “deterministic” DCMs, the third model wins over others with consistently higher posterior probability and higher exceedance of model evidence.

Figure 3: (a), (b) and (c) indicate whole-brain contrasts of the connectivity between dPCu and vPCu. Structural connectivity is in (a), resting-state FC in (b) and task-state FC in (c). Task-state FC was calculated as a partial correlation, controlling for the effect of event-related BOLD signals (i.e., disregarding the apparent correlation caused by stimulus-driven activity). T scores for the statistical effect of the connectivity difference are mapped out on the 3D brain reconstructions. Hot colours represent areas that demonstrate stronger connectivity with the dPCu than with the vPCu, and vice versa for the cold colours. Circular wedge plots to the left and right are a representation of ICN spatial involvement i.e. show voxel overlap between canonical ICNs and the connectivity results. The canonical ICNs are defined by the BrainMap/BM ICN atlases from the ICN_atlas toolbox (Kozák et al., 2017). Cognitive domains and descriptions of the BM/ICN atlases for the ICNs are also provided. The same colour scheme for the (ICN-) BM atlases was used throughout the paper.

Figure 2. Seed derivation for the connectivity analyses. Task-specific seeds for functional and effective connectivity during tasks were derived by overlapping activation maps from either the N-back or RP task, and the DMN and PCu masks from the Conn atlas. Analysis for each task was carried out independently. For task-independent structural and resting-state functional connectivity, the PCu seeds (d/vPCu) were defined as the spatial intersection of the activated or deactivated clusters in both tasks, the DMN and the PCu masks.

Figure 6: Task modulation of the d/vPCu’s FC (i.e. eFC) from the PPI analysis predicts reaction time of correct responses. The stronger the dPCu’s FC with bilateral frontoparietal regions, the quicker the participant responds correctly in the difficult condition (using their RT in easy condition as a baseline). Plots show the linear relationships between eFC and (relative) reaction time to correct responses at the global statistical maxima of the two analyses. BM-ICN involvement of the significant regions is demonstrated using the ICN_atlas toolbox (also see figure 3).

Full text

1
A PRECUNEAL CAUSAL LOOP MEDIATES EXTERNAL AND
INTERNAL INFORMATION INTEGRATION IN THE HUMAN BRAIN
D. Lu
1,2,3
, I. Pappas
4
, D. K. Menon
1,2,3
& E. A. Stamatakis
1,2,3
Affiliations:
1
Department of Clinical Neurosciences, University of Cambridge, Addenbrooke’s Hospital, Hills Rd,
CB2 0SP Cambridge, UK.
2
University Division of Anaesthesia, University of Cambridge, Addenbrooke’s Hospital, Hills Rd, CB2
0SP Cambridge.
3
Wolfson Brain Imaging Centre, University of Cambridge, Cambridge Biomedical Campus (Box 65),
CB2, 0QQ, Cambridge, UK.
4
Helen Wills Neuroscience Institute, University of California, Berkeley, CA, 94720, USA.
Corresponding authors. Email addresses: dl577@cam.ac.uk, eas46@cam.ac.uk.
ABSTRACT
Human brains interpret external stimuli based on internal representations. One untested
hypothesis is that the default-mode network (DMN) while responsible for internally
oriented cognition can also encode externally oriented information. The unique
neuroanatomical and functional fingerprint of the posterior part of the DMN supports a
prominent role for the precuneus in this process. By utilising imaging data during two
tasks from 100 participants, we found that the precuneus is functionally divided into
dorsal and ventral subdivisions, each one differentially connecting to internally and
externally oriented networks. The strength and direction of their connectivity is
modulated by task difficulty in a manner dictated by the balance of internal versus
external cognitive demands. Our study provides evidence that the medial posterior part
of the DMN may drive interactions between large-scale networks, potentially allowing
.CC-BY 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2020. ; https://doi.org/10.1101/2020.08.20.259846doi: bioRxiv preprint

2
access to stored representations for moment to moment interpretation of an ever-
changing environment.
INTRODUCTION
The intrinsic coupling between regions in the human brain (or functional
connectivity/FC) is not random but forms consistent spatial patterns known as Intrinsic
(functional) Connectivity Networks (ICN) (Seeley et al., 2007). Each ICN’s relevance to
cognitive function has been established from activation studies, from which we can infer
what information features are encoded in different networks (Cole et al., 2014).
Interactions between ICNs are thought to be a form of information exchange
serving cognitive demands, although the precise functional role of those interactions
remains an active area of research (Bressler & Menon, 2010). A case in point is the
interaction between the default mode network (DMN) and cognitive control networks
which have been reported to be anti-correlated (Fox et al., 2005) and contraposed in
their cognitive function (Weissman et al., 2006). This view however is increasingly
challenged by accumulating findings. While the majority of DMN studies focus on
resting state, i.e. in the absence of external stimulation, emerging evidence shows that
the DMN is indeed engaged during goal-directed tasks (Elton & Gao, 2015; Spreng et
al., 2014; D. Vatansever et al., 2015; Deniz Vatansever et al., 2015, 2018), as opposed
to being a “resting-state” network.
A conciliatory model that can account for both the internal and external role of
DMN is that of predictive coding. The basic assumption is that conscious processing,
whether externally oriented or not, always entails some form of internal processing.
.CC-BY 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2020. ; https://doi.org/10.1101/2020.08.20.259846doi: bioRxiv preprint

3
Thus internal and external representations are not dissociated but rather constantly
update each other in a Bayesian fashion (Barrett & Simmons, 2015; Friston, 2012).
Given its multifaceted role in external and internal processing and motivated by
pharmacological and brain injury studies that highlight the DMN’s prominent role in
conscious processing (Liu et al., 2015; Perri et al., 2016; Vanhaudenhuyse et al., 2010),
we hypothesised that the DMN is fundamental when considering the neural instantiation
of such an account.
The DMN, which comprises medial frontal and medial posterior parietal cortices
as well as angular gyri and hippocampus, is neither anatomically nor functionally
homogeneous (Braga et al., 2013; Braga & Buckner, 2017; Kernbach et al., 2018).
Among the DMN regions, the precuneus (PCu/PCC) has attracted significant interest
due to its complex neuroanatomical, metabolic and functional fingerprint (de Pasquale
et al., 2012; Raichle et al., 2001; Spreng et al., 2009; Utevsky et al., 2014). The PCu
has been characterised as the nexus of the DMN’s integrative fingerprint (Utevsky et al.,
2014). It is engaged in a broad range of cognitive tasks, including both internal
representation and externally-oriented, goal-directed tasks (Cavanna & Trimble, 2006;
Fletcher et al., 1995) and its activity/connectivity can discriminate between
conscious/unconscious states (Utevsky et al., 2014). It has exceptionally high metabolic
rate (Raichle et al., 2001) and is suggested to have undergone the biggest brain
morphological expansion through human evolution (Bruner & Iriki, 2016). The PCu is
also a major connectivity hub from a graph-theoretic perspective (P van den Heuvel &
Sporns, 2011; Tomasi & Volkow, 2011; van den Heuvel & Sporns, 2013). Based on its
.CC-BY 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2020. ; https://doi.org/10.1101/2020.08.20.259846doi: bioRxiv preprint

4
functional features we hypothesised that the PCu may play a key role in integrating
external information with internal representations.
To investigate this claim, we first ascertained the DMN’s involvement in cognitive
tasks by employing a NeuroSynth meta-analytic framework. We then used MRI datasets
(diffusion MRI, resting-state & task-state fMRI) from the Human Connectome Project to
investigate activity as well as functional and structural connectivity of the PCu with
whole-brain ICNs. We found that the dorsal and ventral PCu subdivisions had distinct
activity and connectivity patterns modulated by task difficulty, a factor that influenced a
mirrored interplay between the PCu and internally/externally oriented networks. This
feature of the PCu suggests that it might support the integration between internally and
externally oriented cognitive processes. Moreover, Dynamic Causal Modelling (DCM)
provided evidence for directed couplings between the two PCu subdivisions, hinting at a
combinatorial processing mode where incoming information is associated with internal
representations.
RESULTS
The DMN during tasks: A meta-analytic perspective
We first validated that the DMN is indeed involved during goal-directed tasks, as
findings regarding DMN’s involvement during tasks are still disputed (Fransson, 2006).
We conducted a meta-analysis with the latest NeuroSynth database, using a text-based
filter (“attention* or execut*”) to search for tasks that require either attention or executive
processing or both (since the two cognitive processes are hardly separable in practice),
which yielded 1219 fMRI studies. Using the forward-inference estimation, we found that
.CC-BY 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2020. ; https://doi.org/10.1101/2020.08.20.259846doi: bioRxiv preprint

5
a large portion of the DMN (1667 voxels) was associated with these tasks (Figure 1a),
with significant clusters (Z > 4) located at the PCu and angular gyrus (AnG). In contrast,
the reverse-inference estimation only revealed 68 voxels in DMN regions (Figure 1b).
We acknowledge that NeuroSynth meta-analyses do not differentiate between
activation and deactivation but based on previous literature which has shown PCu and
AnG involvement in cognitive processing, we believe what we see here is that the tasks
employ posterior DMN regions, yet their activation is so pervasive among all kinds of
tasks that cannot be exclusively associated with the specified goal-directed tasks. We
propose that the meta-analysis indicates a domain-general role for posterior DMN areas
which may serve cognitive demands by providing contextual information.
To investigate the relationship between DMN and other ICNs in more detail
during task execution, we selected two fMRI paradigms (100 young healthy participants)
from the HCP dataset. The tasks we chose were the N-back working memory task and
the Relational Processing (RP) task, which have attention-demanding, goal-directed
features, and engage similar cognitive domains at two difficulty levels (N-back: 2-back >
0-back conditions, RP: relational processing > matching conditions); therefore, the two
tasks afforded us the possibility of inferring the brain’s response to varying cognitive
demands regardless of the specific cognitive content. The merit of using two tasks is to
demonstrate that the DMN’s function is generalisable. As a sanity check for both tasks,
the accuracy rate (N-back task: t = 2.44, p = 0.016; RP task: t = 8.80, p = 1.012e-15)
was significantly higher in the easy than difficult conditions, and reaction times (RT)
were significantly shorter (N-back task: t = 10.58, p < 2.2e-16; RP task: t = 10.02, p <
.CC-BY 4.0 International licenseavailable under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprintthis version posted August 21, 2020. ; https://doi.org/10.1101/2020.08.20.259846doi: bioRxiv preprint