Individual Differences in Brain Activity Related to Human Intelligence

Post by Elisa Guma

What's the science?

One of the characteristics of humans is the ability to perform cognitively challenging tasks in varied situations. This adaptability is typically used as a measure of general intelligence. However, there is still debate over whether general intelligence should be studied as one general ability, or as a mixture of many distinct psychological processes. For more than three decades, neuroscientists have tried to tie this question to the brain's functional architecture. It has been proposed that to meet task demands, specific brain regions become transiently active and that these brain regions are part of heavily overlapping dynamic functional networks. This week in Nature Communications, Soreq and colleagues sought to investigate how different cognitive tasks and performance on those cognitive tasks relate to dynamic brain activity.

How did they do it?

The authors relied on an established intelligence test which included twelve cognitive tasks. The data they used had online test responses from over 60,000 participants. They also obtained functional magnetic resonance imaging (fMRI) data from healthy young adults as they performed the exact same test. These data enabled them to identify the brain regions found to be consistently active for all 12 tasks ('domain-general regions’). To identify these brain regions, the authors relied on a classic yet often underused conjunction analysis method. These regions are considered task agnostic since they are active for all tasks and contain very little or no task-specific information. Areas outside the 'domain general' mask include the regions in which task-specific information is stored. The authors then compared how similar tasks were using either a cortical mask (a selection of particular brain regions) containing general and specific regions or the domain-general mask. By applying the same approach to the online behavioural data, they could compare brain and psychometric (the behavioural test data) task similarities.

Next, they wanted to see how distinct tasks were. In other words, they asked: Can we tell that a person is doing a particular task only from their neural information? The authors examined the relationships between neuronal information based on brain activity (the relative local metabolic requirements of each brain region) and dynamic functional connectivity (how different brain regions interact in specific time windows). To quantify brain activity, the authors assessed all voxels contained within the ‘domain general’ mask, and to quantify dynamic functional connectivity the authors parcelled the 'domain general' mask into distinct regions For the cortical mask, the authors used an unbiased parcellation set based on resting-state fMRI data. Using these mined features (i.e. brain activity and connectivity), the authors performed a machine-learning-based classification analysis with the 12 cognitive tasks. 

In response to the reviewers' comments, the authors were asked to confirm that their ability to correctly classify between the different tasks was not influenced by the motor or visual differences between the tasks but rather was a function of the different cognitive abilities required of each task. The authors accomplished this by aggregating the various tasks into three meta-labels (i.e. cognitive, visual, and motor) and conducting another classification analysis, this time comparing the actual task meta-label assignment to permuted ones. Furthermore, following the reviewers' criticism, the authors examined whether neuronal information can predict how well an individual will perform on the intelligence test. Specifically, they calculated each individual's task performance principal component analysis (PCA) and overall classification accuracy from the brain (i.e., how well we can correctly predict from your brain what tasks you are performing).

What did they find?

The authors first replicated the existence of a task-active (i.e., domain-general) network that showed activity across all 12 tasks. This network consisted of areas including the parietal, visual, and motor areas. The authors then demonstrated that cortical and domain-general similarities are highly correlated with psychometric similarity. This suggests that they all rely on the same mixtures of the underlying physiological and neurobiological systems that actually perform the tasks. In addition, the slight difference between the domain-general mask and the cortical mask suggests that the domain-general mask is not entirely homogeneous.

Classification of tasks based on either brain activity or connectivity from domain-general regions yielded impressive model performance with 37% and 43.1% accuracies which are 5 times better than chance (8%) for a balanced 12-tasks classification problem. The same learning algorithm, however, using cortical information, produced dramatic improvement, achieving 49% and 69% accuracy. Connectivity and activity complemented each other - the model trained on both measurements outperformed independent models (with 74%). 

Then the authors showed that the actual assignment of tasks into different cognitive and motor meta-labels significantly outperformed permuted assignments that maintained the same imbalance. It is possible that we decode these tasks by a mixture of networks that carry out both cognitive and motor functions. Additionally, they demonstrated a link between the accuracy of the models at the individual level and the performance of the tasks being classified. Tasks from higher-performing individuals were classified more accurately. Finally, they demonstrated that they could accurately predict individual performance and that this ability was mainly due to connectivity between different prominent resting state networks, such as default mode and dorsal attention.

What's the impact?

This study used a combination of machine learning techniques applied to fMRI and psychometric data to show that performance on cognitive tasks could accurately predict dynamic brain connectivity networks and that more similar cognitive tasks activated more overlapping networks. Further, better performance on cognitive tasks was related to the ability to activate more specific dynamic networks and to flexibly switch between them. The data presented here provide interesting information about how human intelligence and brain activity may be related. Finally, this framework may be applied to clinical data in hopes of identifying markers for quantifying disease pathologies.

Soreq E al. Neuroimaging evidence for a network sampling theory of individual differences in human intelligence test performance. Nature Communications (2021). The original scientific publication here.

Mapping Gene Transcription and Neurocognition Across Human Neocortex

Post by Andrew Vo

What's the science?

As neuroscientists, we study the brain at different levels: from genes and neurons at the microscale to cognition and behavior at the macroscale. We know intuitively that these two scales are linked, building from gene expression to protein interactions, to neuronal wiring and firing, to complex psychological processes. However, this link has yet to be demonstrated in a comprehensive, data-driven, and multivariate framework. This week in Nature Human Behaviour, Hansen and colleagues bridge these two scales by spatially mapping gene expression and functional activation patterns across the human cortex.

How did they do it?

The authors began by performing partial least-squares (PLS) analysis on two open source datasets: (1) the Allen Human Brain Atlas (AHBA) that maps the expression of different genes across the brain and (2) Neurosynth that meta-analytically assigns psychological terms to brain regions they are commonly associated with. The resulting latent variable represented a covarying pattern of gene expression and functional activation, which they referred to as a gene-cognition gradient. In other words, they generated a scale that captured how much gene expression was related to functional activation of different brain regions. Next, they determined which specific sets of genes and psychological processes were related to one another by computing their loadings (i.e., the strength of their contributions) on the gene-cognition gradient. They further examined the biological processes and specific cell types associated with the uncovered gene sets.

To test whether their gene-cognition gradient followed the brain’s structural organization—an intermediate step proposed to link gene expression to functional activation—they compared it to several other previously reported brain patterns. These patterns described microstructural, laminar (referring to the brain’s layers) and functional organizations of the brain. Finally, they tested whether the gene-cognition gradient evolves across neurodevelopment by examining this pattern in the BrainSpan dataset, which provides gene expression estimates across varying stages of human development.

What did they find?

Multivariate PLS analysis revealed a pattern of gene expression and functional activation that spatially covaried across the brain in a ventromedial-dorsolateral gradient. This pattern separated gene sets that were related to affective (emotion-related) processes, neurogenesis, and differentiation, and support cell (e.g., astrocytes, microglia) expression from those gene sets associated with perception and attention, synaptic signaling, and inhibitory/excitatory neurons. Taken together, these findings demonstrate a link between gene expression and functional brain processes.

The authors also found that the gene-cognition gradient reflected other previously described brain organizations, hierarchies based on microstructural, laminar, or functional attributes. This suggests that the link between gene expression and functional activation is likely mediated through brain structure. Examining changes in the gene-cognition gradient across different stages of neurodevelopment, they found that the pattern strengthened over time and was most pronounced in adolescence and adulthood.

What's the impact?

In summary, this study identified a gene-cognition gradient that directly couples genetic expression to functional activation across the human cortex. This gradient delineates sets of genes, biological processes, and specific cell types related to emotional versus perceptual processes. The organization of this gene-cognition coupling follows the brain’s structural and functional hierarchies and matures through neurodevelopment. The study builds on previous literature that focused on single genes, brain regions, or cognitive functions by analyzing high-dimensional genetic and psychological data in a multivariate framework to offer a broader, more comprehensive view of the link between genes and cognition. This framework opens doors to new hypotheses about the genes involved in specific psychological processes, and vice-versa. It may also allow thorough characterization of brain alterations related to different psychiatric disorders across multiple scales, from transcription to cognition.

Hansen et al. Mapping gene transcription and neurocognition across human neocortex. Nature Human Behaviour (2021). Access the original scientific publication here.

Adolescent Cannabis Use and Outcomes in Young-Adulthood

Post by Leigh Christopher

What's the science?

One common concern with cannabis legalization is the possibility that cannabis use negatively impacts brain development during youth. Many studies have shown links between cannabis use and negative outcomes like mental health problems, cognitive problems, and a reduced ability to obtain education and income later in life. Understanding whether cannabis use is actually the cause of negative outcomes in adolescents is challenging from an experimental perspective. Other genetic and environmental factors might contribute to vulnerability to negative outcomes in response to cannabis use, making it difficult to disentangle which factors are causal. Many studies to date have been limited as they have either 1) examined the impact of cannabis use at one point in time (not over time), or 2)  they have not accounted for genetic factors that could influence vulnerability to negative outcomes. This week in PNAS, Schaefer and colleagues used three longitudinal twin studies which fully account for shared genetic and environmental factors, to examine the effects of cannabis use on cognitive, psychiatric, and socioeconomic outcomes in young adults.

How did they do it?

The authors looked for associations between cannabis use in adolescents and negative outcomes in young adulthood using a large sample (3762 participants) that included data from 3 longitudinal twin studies. Analyses conducted in monozygotic (identical) twins account for shared genetic and environmental contributions to the outcome measure of interest since these twins have identical genes and come from the same families. Therefore, twin studies act as a much stronger indicator of causality - a finding that twins who use more cannabis and show more negative outcomes would indicate that the negative outcomes are not due to any confounding genetic or shared environmental vulnerability, but rather are due to the cannabis use itself. Having said that, there are always other potential twin-specific confounders that could differ between a set of twins such as exposure to other drugs. The authors created an adolescent cannabis use index to examine the participants' cannabis use prior to and during adolescence. They examined whether individuals who used more cannabis also experienced more negative cognitive, psychiatric, and socioeconomic outcomes in young adulthood.

What did they find?

Broadly speaking, cannabis use was associated with a number of psychiatric, cognitive, and socioeconomic outcomes such as anxiety, depression, and lower educational attainment. The authors then looked at whether this association still held true after accounting for shared genetic and environmental variability by examining monozygotic twin pairs who differed in terms of their cannabis use. When looking at these twins, the association between cannabis use and cognitive or psychiatric outcomes was no longer significant, suggesting that this association is due to genetic predisposition or other family background factors. The association between cannabis use and socioeconomic outcomes, however, remained significant, indicating that the link between cannabis use and worse socioeconomic outcomes (housing, income, education, occupational status) is not confounded by shared genetic or environmental vulnerability. Since twins may also differ in some environmental factors like exposure to substance abuse in adolescence, the authors performed a follow-up analysis to account for exposure to alcohol and tobacco. They found that the results did not differ when taking these factors into account. Lastly, the authors conducted a follow-up analysis to examine the pathways through which cannabis use might influence socioeconomic outcomes. They found that cannabis use was predictive of worse academic performance, motivation, and problems in school after accounting for shared genetic or environmental vulnerability. 

What's the impact?

This study was the first to look at the impact of adolescent cannabis use on multiple adult outcomes using a large, longitudinal sample of twins with repeated assessments of cannabis use administered during the teenage years. These findings suggest that cannabis use does not cause negative cognitive or psychiatric outcomes in adolescents and that these outcomes are more likely driven by shared genetic or environmental vulnerability. However, this study did show that cannabis use is linked to worse socioeconomic outcomes after controlling for genetic factors and that cannabis use likely impacts academic performance leading to worse outcomes in young adulthood.

Schaefer et al. Associations between adolescent cannabis use and young-adult functioning in three longitudinal twin studies. PNAS (2021). Access the original scientific publication here.