Abstract
Intelligence describes the general cognitive ability level of a person. It is one of the most fundamental concepts in psychological science and is crucial for effective adaption of behavior to varying environmental demands. Changing external task demands have been shown to induce reconfiguration of functional brain networks. However, whether neural reconfiguration between different tasks is associated with intelligence has not yet been investigated. We used fMRI data from 812 subjects to show that higher scores of general intelligence are related to less brain network reconfiguration between resting state and seven different tasks as well as to network reconfiguration between tasks. This association holds for all functional brain networks except the motor system, and replicates in two independent samples (N = 138, N = 184). Our findings suggest that the intrinsic network architecture of individuals with higher general intelligence scores is closer to the network architecture as required by various cognitive demands. Multi-task brain network reconfiguration may, therefore, reflect the neural equivalent of the behavioral positive manifold – the essence of the concept of general intelligence. Finally, our results support neural efficiency theories of cognitive ability and reveal insights into human intelligence as an emergent property from a distributed multi-task brain network.
Introduction
Intelligence captures the general cognitive ability level of a person. It is critically involved in learning from experiences and a prerequisite for effective adaption to changing environmental demands (Sternberg, 1997). People who score high on tests of general intelligence perform better in multiple different cognitive tasks – an observation that is called the positive manifold of general intelligence (Spearman, 1904). Although scientists started to investigate the biological underpinning of intelligence many decades ago, and correlates have been identified in brain structure (Gregory et al., 2016), brain function (Neubauer & Fink, 2009), and in intrinsic brain connectivity (Basten et al., 2015; Hilger et al., 2017; Jung & Haier, 2007), it remains an open question whether there exists an equivalent of the positive manifold of general intelligence within the human brain, i.e., a ‘neuro-g’ (Haier, 2017).
Intrinsic brain networks can be assessed in the absence of task demands during the so-called resting state (Biswal et al., 1995; Van den Heuvel & Hulshoff Pol, 2010). Their topology has been suggested as reflection of information transfer between different brain regions and various topological network attributes have been related to differences in cognitive ability (Dubois et al., 2018; Hilger et al., 2020). Recently, the focus has broadened to include functional brain network interactions measured during active cognition (Braun et al., 2015; Cohen & D’Esposito, 2016). Introducing such task demands to the investigation of functional connectivity has been shown to amplify relations between phenotypes and their neural basis, suggesting task-based connectivity as promising marker of general intelligence (Greene et al., 2018, 2020).
Brain network reconfiguration, defined as changes in fMRI-derived functional brain connectivity in adaption to different cognitive states, has previously been studied by comparing resting-with task-state fMRI assessments (Schultz & Cole, 2016). Task-evoked changes in functional connectivity seem to be crucial for shifting neural processing (Cole et al., 2021), and a pioneering study revealed a significant (negative) association between a global estimate of brain network reconfiguration and general intelligence (Schultz & Cole, 2016). However, as the exact nature of changes has been shown to depend on the kind of the cognitive task (Braun et al., 2015; Cohen & D’Esposito, 2016; Soreq et al., 2021) as well as on the intensity level of the cognitive challenge (Hearne et al., 2017; Shine et al., 2016), considering brain network reconfiguration as task-general phenomenon may only provide limited insights into underlying processes. Mechanistic insights into general intelligence, i.e., into implicated cognitive processes and into a potential neural equivalent of the positive manifold, would therefore require the investigation of reconfiguration between different cognitive tasks. Such multi-task brain network reconfiguration has been demonstrated to capture meaningful variations between persons (Salehi et al., 2020) but has not yet been related to general intelligence. Furthermore, it has not yet been tested whether the association between brain network reconfiguration and general intelligence is driven by specific functional systems or whether it represents a whole-brain phenomenon. This would allow for additional insights about intelligence-relevant processes and how these processes are implemented on a neural level.
Here, we use fMRI data from a large sample of healthy adults (N = 812) assessed during resting state and seven different tasks to test the hypothesis that higher levels of general intelligence relate to less brain network reconfiguration. Specifically, we expected this association to manifest in reaction to different cognitive demands and on various spatial scales. We used a straight-forward operationalization of brain network reconfiguration and implemented our analyses on a whole-brain level as well as on the level of seven and 17 canonical functional brain networks. The results confirm our hypotheses and suggest that functional brain networks of more intelligent people may require less adaption when switching between different cognitive states, thus, pointing towards the existence of an advantageous intrinsic brain network architecture. Further, we show that although the different cognitive states were induced by different demanding tasks, their relative contribution to the observed effect was nearly identical, a finding that supports the assumption of a general neural correlate, i.e., a neural positive manifold. Finally, the involvement of multiple brain networks suggest intelligence as an emergent property of a widely distributed multi-task brain network.
Method
Participants
Main analyses were conducted on data from the HCP Young Adult Sample S1200 including 1200 Subjects of age 22-37 years (656 female, 1089 right-handed, mean age = 28.8 years). All study procedures were approved by the Washington University Institutional Review Board, and informed consent in accordance with the declaration of Helsinki was obtained from all participants (for details see Van Essen et al., 2013). Subjects with a Mini-Mental State Examination (MMSE) score ≤ 26 (serious cognitive impairment) or missing cognitive data needed for calculating a general intelligence factor were excluded. Cognitive measures of the remaining 1186 subjects were used as input for factor analysis to estimate a latent factor of general intelligence (see next section). After additional exclusion due to missing fMRI data and excessive head motion (see below), the final sample consists of 812 subjects (422 female, 739 right-handed, 22-37 years, mean age = 28.6 years).
General intelligence g
To estimate a latent factor of general intelligence (g-factor), bi-factor analysis based on the Schmid-Leiman transformation (Schmid & Leiman, 1957) was conducted in accordance to Dubois et al. (2018) for 12 cognitive measures (Table S1) of 1186 subjects.
Data acquisition and preprocessing
We used fMRI data from all four resting-state runs and data acquired during seven tasks (two runs each) capturing information from eight different external demands. Resting-state runs comprise 14.55 min data (1,200 time points), while task runs vary between 2.27 min (176 time points) and 5.02 min (405 time points) lengths. See Van Essen et al. (2013) for details of imaging parameters, Smith et al. (2013) for details of the resting-state acquisition, and Barch et al. (2013) for additional information about tasks. We used the minimally preprocessed HCP fMRI data (Glasser et al., 2013) and implemented further preprocessing comprising a nuisance regression strategy with 24 head motion parameters, eight mean signals from white matter and cerebrospinal fluid, and four global signals (Parkes et al., 2018). For task data, basis-set task regressors (Cole et al., 2019) were used simultaneously with the nuisance regressors to remove mean task-evoked activations. Finally, timeseries of neural activation were extracted from 200 nodes covering the entire cortex (Schaefer et al., 2018). In-scanner head motion was measured by framewise displacement (FD, Jenkinson et al., 2002). As recommended in Parkes et al. (2018), subjects were only included if mean FD < .2 mm, proportion of spikes (FD > .25 mm) < 20%, and no spikes above 5 mm were observed.
Functional connectivity
Subject-specific weighted functional connectivity matrices (FC) were computed using Fisher z-transformed Pearson correlations between time series of neural activation from 200 cortical regions. For each state (rest, tasks), FC was first computed for RL and LR phase directions separately and averaged afterwards. Functional connections were then filtered based on their correlation with intelligence (p < .1, Finn et al., 2015). Connections inconsistently correlated with intelligence across states (positive in one, negative in another or vice versa) were excluded. To rule out potential confoundations of the reconfiguration measures, the filtering step was cross-validated: The sample was divided into ten subsamples (by ensuring absence of family relations and equal distributions of intelligence scores via stratified folds) and intelligence-relevant connections were selected in nine subsamples only. This selection of connections was then applied to the withheld subsample. Reconfiguration measures were calculated on a whole-brain level, as well as within and between pairs of networks based on the Yeo 7/17 canonical systems (Yeo et al., 2011). Note that for the analysis on the level of 17 functional networks, the p-threshold was increased to p < .2 to ensure a sufficient number of remaining connections (see Fig. 1 for a schematic illustration of the general workflow, Fig. S1 for the filtering procedure).
Schematic illustration of global analysis workflow. (a) Brain activity was assessed with fMRI during eight different cognitive states (resting state and seven tasks). (b) For each state, functional brain connectivity matrices (FCs, d) were computed by correlating the time series of 200 nodes with each other. For noise reduction, FCs were filtered based on their correlation with intelligence (see Fig. S1 and Methods for details). Brain network reconfiguration was calculated for all state combinations as cosine distances (Δ) between the filtered FCs (e). To assess the relationship between brain network reconfiguration and intelligence, reconfiguration values were correlated (Spearman correlations, controlled for age, sex, handedness and in-scanner head motion) with a latent g-factor derived from 12 cognitive scores using a bi-factor analysis model (c). BOLD, Blood oxygen level dependent (signal); t, time; c, cognitive score.
Brain network reconfiguration
Reconfiguration of functional connectivity was operationalized as cosine distance dcos between the filtered FCs of two states:
with connection weights vectors a and b and total number of connections n. Note, that cosine distance captures changes in orientation between two vectors and thus, indexes changes in the architecture (structure) of functional connectivity rather than changes in the strengths of connections (as captured with, e.g., Manhattan distance, Euclidean distance).
Association between reconfiguration and intelligence
Relations between reconfigurations and intelligence were assessed with Spearman rank-order partial correlations by controlling for age, sex, handedness, and in-scanner head motion (mean FD over all scans and mean of percentage of spikes > .25 mm over all scans). For multiple comparisons, p-values were FDR corrected (α = .05). Reconfiguration was computed with different foci: A) Whole-brain average reconfiguration: The average reconfiguration between resting state and all tasks and the average reconfiguration between all pairs of tasks. B) Whole-brain state combination-specific reconfiguration: Reconfigurations for each pair of rest-task and task-task combinations (one resting state and seven tasks), resulting in 28 reconfiguration scores. C) Whole-brain state-specific reconfigurations: Reconfiguration averaged over all state combinations a specific state was involved in. Note that for task states, only combinations with different tasks (no rest) were included. D) Network- and state combination-specific reconfiguration: Reconfiguration of all network combinations (within- and between network connectivity) for each state combination. These latter reconfiguration measures (case D) were further averaged to a) network combination-specific reconfigurations: reconfiguration averaged over all state combinations, b) network-specific reconfigurations: reconfiguration averaged over all state combinations, and averaged over all network combinations the respective network was involved in, c) state-specific reconfigurations on the network level: reconfiguration averaged over all state combinations the respective state was involved in, and averaged over all network combinations the respective network was involved in.
External replication
For testing the robustness of our findings against varying measures of intelligence, varying cognitive demands induced by different tasks, and sample dependence, all analyses were repeated in two independent data sets (PIOP1, PIOP2) from The Amsterdam Open MRI Collection (AOMIC, Snoek et al., 2021). All study procedures were approved by the faculty’s ethical committee before data collection started (PIOP1 EC number: 2015-EXT-4366, PIOP2 EC number: 2017-EXT-7568) and informed consent in accordance with the declaration of Helsinki was obtained from all participants (for more details see Snoek et al., 2021). PIOP1 includes fMRI data of 216 subjects collected from six cognitive states (resting state, five tasks), while PIOP2 contains fMRI data of 226 subjects from four states (resting state, three tasks). The Raven’s Advanced Progressive Matrices Test (36 item version - set II, Raven et al., 1998) was used in both samples for measuring intelligence. After excluding subjects with missing descriptive and behavioral data and after applying motion exclusion criteria (see above), 138 subjects (PIOP1) and 184 subjects (PIOP2) remained for analyses. The fMRI data was downloaded in the minimal preprocessed form, using an alternative preprocessing pipeline (fMRIprep v1.4.1, Esteban et al., 2019). Further preprocessing to extract nuisance regressed time series followed the same steps as specified above. As the PIOP samples are relatively small compared to the main sample and brain-behavior relationships are less reliable in small samples (Assem et al., 2020; Marek et al., 2020), no p-threshold was used for the selection of functional connections here. Instead, and to increase the robustness of this analyses, a filter mask was computed from the larger main sample (containing connections correlating only either positively or negatively with intelligence p < .01 in at least one of the filtered FCs of intersecting state combinations) and only connections located in these main sample filter mask and correlating with intelligence in the same direction in the replication samples were used in analyses.
Results
Intelligence
General intelligence was operationalized as latent g-factor from 12 cognitive measures (Table S1) computed with bi-factor analysis (Dubois et al., 2018) using data from 1186 subjects of the Human Connectome Project (Van Essen et al., 2013). As per model fit criteria of Hu & Bentler (1999), the 4-bi-factor model fits the data well (CFI = 0.979, RMSEA = 0.0395, SRMR = 0.0213). The statistical model and the g-factor distribution in contrast to the PMAT-score (brief assessment of intelligence provided by the HCP) is shown in Fig. S2.
Less brain network reconfiguration is associated with higher intelligence
Brain network reconfiguration was operationalized as cosine distance between filtered functional connectivity matrices (FCs) of two out of eight different cognitive states (see Methods, Fig. 1 for a schematic illustration of the analyses workflow, Fig. S1 for details about the FC filtering procedure). Averaged across all rest-task and task-task state combinations, less brain network reconfiguration was associated with higher intelligence scores (rest-task: rho = -.23, p < .001; task-task: rho = -.23, p < .001; Fig. 2a). This effect also holds when using stricter thresholds for the cross-validated filtering approach, e.g., p < .01 (Table S2) or when using alternative mathematical operationalizations of reconfiguration (Pearson correlation between Fisher z-transformed FCs: rest-task: rho = .23, p < .001, task-task: rho = .23, p < .001; Manhattan distance between bi-partitioned FCs: rho = -.19, p < .001, task-task: rho = -.24, p < .001).
Less brain network reconfiguration is associated with higher intelligence. (a) Scatterplots illustrating the association (partial Spearman correlation, rho) between a latent g-factor of intelligence (derived from 12 cognitive tasks, x-axis) and the standardized residuals resulting from linear regression of age, sex, handedness, and in-scanner head motion on the variable of interest, that is brain network reconfiguration. Brain network reconfiguration was operationalized as cosine distance between functional connectivity matrices averaged over all possible rest-task combinations (y-axis, left panel), and all possible task-task combinations (y-axis, right panel) respectively. Note that only in this subfigure data of one subject was excluded due to visualization purposes. (b) Association between intelligence and brain network reconfiguration for all possible state combinations (FDR-corrected p-values, α = .05, all correlations are significant). The strengths of correlations are depicted in different colors (see color bar). (c) Associations between intelligence and a total measure of state-specific reconfiguration. i.e., reconfiguration values were averaged over all state combinations the respective state was involved in (FDR-corrected p-values, α = .05, all correlations are significant). Note that for task states, only combinations with different tasks (no rest) were included. Significant differences in correlation values (p < .05, marked with an asterisk) were only observed for the comparisons between the language task (LAN) and the social cognition, the relational processing, and the emotion processing task (SOC, REL, EMO) respectively. (d) Associations between intelligence and brain network- and state combination-specific reconfiguration values. Brain networks were derived from the Yeo atlas (Yeo et al., 2011, seven network partition used here) and network combinations refer to all within and between network connectivity combinations (columns). The strengths of correlations are depicted in different colors (see color bar). Note that NaN (not a number) values exist if in a specific network-state combination no single brain connection passes the filtering procedure (see Fig. S1 and Methods). For details about the assignment of the correlation values to the specific state and network combinations, see Fig. S5. Std. res., standardized residuals; FDR, false discovery rate; RES, resting state; WM, working memory task; GAM, gambling task; MOT, motor task; LAN, language processing task; SOC, social cognition task; REL, relational processing task; EMO, emotion processing task.
Higher intelligence is related to less reconfiguration across different cognitive demands
Significant associations between higher intelligence and less brain network reconfiguration were observed for all rest-task and task-task state combinations (Fig. 2b). The correlations between reconfiguration and intelligence ranged from rho = -.10 (p = .006) for reconfiguration between resting state and social recognition task to rho = -.23 (p < .001) for reconfiguration between working memory and motor task. Again, similar associations were observed when using alternative reconfiguration metrics (Fig. S3). For evaluating the total influence of each individual state, reconfiguration values were averaged across all rest-task combinations (for resting state) and separately over all task-task combinations a respective task was involved in (for each task state). The total influence of the language task was significantly stronger (p < .05) than the influence of the social recognition task, the relational processing, and the emotion processing task, while all other states did not differ significantly in their total influence on the observed effect (Fig. 2c and Fig. S4).
The relation between reconfiguration and intelligence depends on different functional brain systems rather than on specific cognitive demands
By parcellating the brain into seven functional networks (Yeo et al., 2011), and by considering all possible network and state combinations, we observed that the variance of the effect between different state combinations was significantly smaller than the variance of the effect between different network combinations (Wilcoxon rank sum test, W = 441, p < .001, Fig. 2d, Fig. S5). This suggests prior importance of the differentiation between different brain systems rather than between different external demands.
Higher intelligence is related to less reconfiguration across different spatial scales
Next, we analyzed the relative contribution of seven and 17 functional brain networks to the observed effect. Overall, higher intelligence scores were associated with less reconfiguration of within and between network connectivity in multiple brain networks. Dorsal and ventral attention systems, the control network, the default mode network, and limbic areas showed consistent negative associations, while in the visual and somatomotor networks the effect was weaker and the pattern more heterogeneous (Fig. 3a). To derive a more global measure of total network-specific reconfiguration, we then aggregated reconfiguration scores across all network-combinations a respective network was involved in. Higher intelligence was significantly associated with less connectivity reconfiguration in respect to all networks, except the motor system (Fig. 3b). Similar relations were observed within and between 17 functional brain networks (Fig. 3c,d).
Brain network-specific associations between general intelligence and brain network reconfiguration. Partial spearman correlation (rho) between intelligence (g-factor derived from 12 cognitive tasks) and brain network-specific reconfiguration values (averaged cosine distance between functional connectivity matrices of eight different cognitive states) for seven and 17 separate functional brain networks (Yeo et al., 2011). Network-specific correlations were projected onto the surface of the brain. The strengths of correlations are depicted in different colors (see color bar). All correlations were controlled for influences of age, sex, handedness, and in-scanner head motion. All significant correlations (FDR-corrected p-values, α = .05) are marked with asterisks. Note that NaN (not a number) values exist if in a specific network combination no single brain connection passes the filtering procedure (see Fig. S1 and Methods). (a) Associations between intelligence and brain network combination-specific reconfiguration scores for seven functional brain networks. (b) Associations between intelligence and reconfiguration scores for seven functional brain networks (averaged across all within- and between network combinations a respective network is involved in). (c) Associations between intelligence and brain network combination-specific reconfiguration scores for 17 functional brain networks. (d) Associations between intelligence and reconfiguration scores for 17 functional brain networks (averaged across all within- and between network combinations a respective network is involved in). FDR, false discovery rate; VIS, visual network; SMN, somatomotor network; DAN, dorsal attention network; VAN, salience/ventral attention network; LIM, limbic network; CON, control network; DMN, default mode network; TEMP, temporal parietal network.
Network-specific reconfigurations in response to varying external demands
Finally, we investigated network-specific contributions on the association between intelligence and brain network reconfiguration for each cognitive state. To this end, network-specific reconfiguration scores were aggregated across all rest-task combinations (for resting state) or task-task combinations a respective task was involved in (for each task state). As illustrated in Fig. 4, network-specific associations between reconfiguration and intelligence were relatively stable across all cognitive states. Again, similar relations were observed at the level of 17 functional brain networks (Fig. S6).
Associations between intelligence and brain network reconfiguration for each cognitive state. The strengths of partial Spearman correlations (rho) between intelligence (g-factor derived from 12 cognitive tasks) and brain network reconfiguration (cosine distance between functional connectivity matrices of different states) are illustrated in different colors (see color bar). All associations were controlled for influences of age, sex, handedness, and in-scanner head motion. Seven functional brain networks were derived from the Yeo atlas (Yeo et al., 2011). For calculating state-specific associations, cosine distances were averaged over all state combinations a respective state was involved in (for task states, only combinations with different tasks were included), and averaged over all network combinations (within- and between network connectivity) the respective network was involved in.
External replication: Generalization to different measures of intelligence and different cognitive demands
To evaluate the robustness of our findings against different measures of intelligence and against varying cognitive demands induced by different tasks, all analyses were repeated in two independent samples (The Amsterdam Open MRI Collection AOMIC, Snoek et al., 2021, PIOP1: N = 138, PIOP2: N = 184, see Methods). In line with our main analyses, less brain network reconfiguration was associated with higher intelligence. This effect holds for rest-task and task-task reconfiguration (PIOP1 rest-task: rho = -.32, p < .001, task-task: rho = -.26, p = .003; PIOP2 rest-task: rho = -.23, p = .002, task-task: rho = -.26, p < .001, Table S2) and became visible across most state combinations (Fig. S4). In PIOP1, nine out of 15 rest-task and task-task combinations showed a significant negative association (range: -.12 ≤ rho ≤ -.35, .001 < p ≤ .18), while in PIOP2, five out of six rest-task and task-task combinations showed the respective effect (range: -.13 ≤ rho ≤ -.30, .001 < p ≤ .09). Aggregating across all state combinations in which a respective state was involved in demonstrated that only in the PIOP2 sample the total influence of the stop signal task was significantly stronger (p < .05) than the total influence of the emotional matching task. All other states did not differ significantly in their total influence on the observed effect (Fig. S4). Finally, results from network-specific analysis were also similar to the results from the main sample (Fig. S7, S8). In sum, the results of the replication analyses support the robustness of our findings and suggest that the association between higher intelligence scores and less brain network reconfiguration generalizes to different cohorts, imaging acquisition parameters, operationalizations of intelligence, and to different cognitive demands.
Robustness control analyses
Although the adopted procedure for filtering out noise-contaminated functional brain connections was thoroughly cross validated (see Methods), to rule out the possibility that results are biased by this step, all whole-brain analyses were repeated by a) considering all possible functional brain connections (no filter) and b) implementing a filter based on the pure overlap and ignoring the sign of the association between connectivity and intelligence (different filter). The same association between higher intelligence and less brain network reconfiguration were observed in both cases (no filter: rest-task: rho = -.12, p < .001; task-task: rho = -.12, p < .001; different filter: rest-task: rho = -.21, p < .001; task-task: rho = -.21, p < .001). Without filtering, state-specific effects were overall smaller with not all state combinations reaching the significance threshold, while state-specific results based on the different filtering procedure were nearly identical (see Table S2, Fig. S9). Together, these analyses suggest that our filtering procedure successfully reduced noise and, most importantly, demonstrate that the observed association between higher intelligence and less brain network reconfiguration does not represent a spurious result of the filter.
Discussion
We showed that general intelligence is associated with less brain network reconfiguration expressed by higher similarity between functional brain connectivity linked to various cognitive states. In line with our initial hypothesis, this effect was not only observed for reconfiguration between rest and task but also for reconfiguration between different tasks each associated with a specific cognitive demand. Multiple control analyses and replication in two independent samples demonstrate the robustness of our findings and suggest generalizability of this effect to different measures of intelligence and to additional cognitive demands. Finally, multiple functional brain systems contributed to this effect suggesting that intelligence is an emergent whole-brain phenomenon.
Our finding that less reconfiguration of functional brain connectivity is related to higher intelligence suggests that more intelligent people may have an intrinsic brain network architecture that is better suited to fulfil many cognitive demands. In this regard, intrinsic brain connectivity, as assessed during rest, can be understood as baseline or inherent architecture that undergoes task-specific adaptations to optimally support upcoming external demands (Cole et al., 2021). The observation that higher intelligence is not only associated with less reconfiguration between rest and task but also with less reconfiguration between different tasks implies that the intelligence-associated advantage in network adaption primarily refers to task-general (in contrast to task-specific) adaptions, which have been shown to build the major proportion of functional connectivity adaptions when external demands are induced (Cole et al., 2014). An intrinsic network architecture which is closer to such a general task-supporting functional brain network structure may allow more intelligent people to switch faster and more efficiently (in terms of energy consumption) between rest and task as well as between different tasks associated with varying cognitive demands. Less required network adaption could thus potentially contribute to lower reaction times (Jensen, 2006) or smaller latencies of event-related brain potentials in reactions to upcoming task stimuli (Schubert & Frischkorn, 2020), both of which have been associated with higher intelligence. Our findings add therefore empirical support to the Neural Efficiency Hypothesis of intelligence that globally suggests that more intelligent people make use of their brain in a more efficient way (Neubauer & Fink, 2009).
The observation that all tasks seem to contribute to the observed effect with almost equal strength, further supports the assumption that task-general adaptation has a greater importance for intelligence-associated processing advantages than task-specific adaptation. The overlap of network architectures as required by different cognitive demands can thus be interpreted as reflection of the positive manifold of general intelligence on the neural level (Kovacs & Conway, 2016; Spearman, 1904). In that, our study reinforces one of the oldest theories of human intelligence (g-factor, Spearman, 1904) and provides at least a preliminary answer to the question about the existence of a ‘neuro-g‘ (Haier, 2017).
We further analyzed the impact of variations between different functional brain systems on the observed effect and showed that the relation between reconfiguration and intelligence depends more on variations between different brain networks than on variations between different tasks. Specifically, reconfiguration in all brain systems except in the somatomotor network was significantly related to intelligence. The here proposed neural positive manifold (the overlap of network architectures as required by different demands) may thus include most but not all functional brain systems and may support a wide variety of different cognitive demands. However, although our replication supports the generalizability to additional cognitive states, the selection of tasks was limited in the current study, and it therefore requires further investigation to test whether this effect is universal for a broader range of cognitive demands.
Overall, the involvement of multiple functional brain systems supports neurocognitive models (P-FIT, MD, see also Barbey, 2018) and meta-analytic findings (Basten et al., 2015) that reveal individual differences in intelligence to be not only associated with variations in structural or functional characteristics of a single brain region, but rather more to properties of a distributed network with major implications of neural systems associated with attentional control (Hilger et al., 2017, 2020), executive functioning (Unsworth et al., 2009), and the default-mode of brain function (Basten et al., 2013). Although limbic brain systems have long been neglected by most of intelligence research, recent evidence supports these brain regions implication also in cognition (Catani et al., 2013) and our study can further contribute to this accumulating evidence.
Several limitations need to be mentioned. First, although we applied in-sample cross-validation strategies to increase the generalizability of the functional connectivity filter, we cannot completely rule out any remaining influences of sample-specific characteristics on the connection selection procedure. This could impact functional connectivity results especially as we observed that filters became instable in smaller samples (replication samples). To address this issue, we conducted a conservative approach and applied the filter mask of the main sample (HCP) to both replication samples which weakens the overall effect but increases its robustness. Future studies may take sample size into strong consideration and draw specific attention to construct robust and across-sample generalized functional connection selection strategies. Second, the sample of our study was restricted to subjects between 22-37 years of age; thus, future studies should address the question whether results generalize to a broader age range. Third, although our replication analysis shows generalizability of our finding to a different intelligence test and different cognitive demands, future studies may contribute to further broaden the picture to more diverse tasks as well as tasks that are more directly associated with the intelligence test assessment (Soreq et al., 2021) including multiple levels of difficulty (Hearne et al., 2017; Sripada et al., 2020). Such investigations would be valuable for gaining more comprehensive insights into the role different cognitive processes may play within the relationship between brain network reconfiguration and general intelligence. Lastly, the analyses reported here might also be adapted to time-resolved connectivity and the analyses of momentary switches between cognitive states (Greene et al., 2020; Shine et al., 2019).
In sum, our study reveals that greater efficiency in the reconfiguration of functional brain networks in response to various external demands is associated with an increased capacity for cognition and intellectual performance. In general, superior performance may profit from fast and efficient neural processing. The here observed association between general intelligence and less task-induced brain network reconfiguration that holds across a broad variety of different cognitive demands may reflect that the intrinsic brain network architecture of more intelligent people is per se closer to a network configuration as required by various external demands. We conclude that such a network architecture constitutes an optimal foundation for fast and efficient cognitive processing that ultimately contributes to intelligent behavior. Finally, the involvement of multiple brain systems suggests intelligence as an emergent whole-brain phenomenon. Taken together, our study reveals multi-task brain network reconfiguration as promising marker to further understand the mechanisms underlying human cognition.
Transparency
Author Contributions
J.T., J.F, O.S. and K.H. conceived of the study. J.T. analyzed the data. J.F. and O.S. preprocessed the data and provided theoretical input during analyses. J.T. and K.H. wrote the manuscript. K.H. developed the initial idea and acquired funding for the project. All authors were involved in the interpretation of results and provided feedback during manuscript preparation.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Open Practices Statement
All analysis code used in the current study was made available by the authors on GitHub: Preprocessing: https://github.com/faskowit/app-fmri-2-mat; Main analyses: https://github.com/jonasAthiele/BrainReconfiguration_Intelligence, https://doi.org/10.5281/zenodo.5031683. All data used in the current study can be accessed online under: https://www.humanconnectome.org/study/hcp-young-adult (HCP), https://doi.org/10.18112/openneuro.ds002785.v2.0.0 (AOMIC-PIOP1), and https://doi.org/10.18112/openneuro.ds002790.v2.0.0 (AOMIC-PIOP2).
Acknowledgements
The research leading to these results has received funding from the German Research Foundation (Grant No. HI 2185 - 1/1) assigned to K. H. Further, this material is based on the work supported by the National Science Foundation Graduate Research Fellowship under Grant No. 1342962 (J.F.). The authors thank the Human Connectome Project (Van Essen et al., 2013), WU-Minn Consortium (Principal Investigators: David Van Essen and Kamil Ugurbil; 1U54MH091657) funded by the 16 NIH Institutes and Centers that support the NIH Blueprint for Neuroscience Research; and by the McDonnell Center for Systems Neuroscience at Washington University, for providing data of the main sample, and all contributors to The Amsterdam Open MRI Collection (Snoek et al., 2021, Principal Investigator: H. Steven Scholte) for providing data of the replication samples.
References
