Prediction of Individual Differences from Neuroimaging Data

Modular morals: Mapping the organization of the moral brain

Is morality the product of multiple domain-specific psychological mechanisms, or one domain-general mechanism? Previous research suggests that morality consists of a range of solutions to the problems of cooperation recurrent in human social life. This theory of 'morality as cooperation' suggests that there are (at least) seven specific moral domains: family values, group loyalty, reciprocity, heroism, deference, fairness and property rights. However, it is unclear how these types of morality are implemented at the neuroanatomical level. The possibilities are that morality is (1) the product of multiple distinct domain-specific adaptations for cooperation, (2) the product of a single domain-general adaptation which learns a range of moral rules, or (3) the product of some combination of domain-specific and domain-general adaptations. To distinguish between these possibilities, we first conducted an anatomical likelihood estimation meta-analysis of previous studies investigating the relationship between these seven moral domains and neuroanatomy. This meta-analysis provided evidence for a combination of specific and general adaptations. Next, we investigated the relationship between the seven types of morality - as measured by the Morality as Cooperation Questionnaire (Relevance) - and grey matter volume in a large neuroimaging (n = 607) sample. No associations between moral values and grey matter volume survived whole-brain exploratory testing. We conclude that whatever combination of mechanisms are responsible for morality, either they are not neuroanatomically localised, or else their localisation is not manifested in grey matter volume. Future research should employ phylogenetically informed a priori predictions, as well as alternative measures of morality and of brain function.

Large Individual Differences in Functional Connectivity in the Context of Major Depression and Antidepressant Pharmacotherapy

Clinical studies of major depression (MD) generally focus on group effects, yet interindividual differences in brain function are increasingly recognized as important and may even impact effect sizes related to group effects. Here, we examine the magnitude of individual differences in relation to group differences that are commonly investigated (e.g., related to MD diagnosis and treatment response). Functional MRI data from 107 participants (63 female, 44 male) were collected at baseline, 2, and 8 weeks during which patients received pharmacotherapy (escitalopram, N = 68) and controls (N = 39) received no intervention. The unique contributions of different sources of variation were examined by calculating how much variance in functional connectivity was shared across all participants and sessions, within/across groups (patients vs controls, responders vs nonresponders, female vs male participants), recording sessions, and individuals. Individual differences and common connectivity across groups, sessions, and participants contributed most to the explained variance (>95% across analyses). Group differences related to MD diagnosis, treatment response, and biological sex made significant but small contributions (0.3-1.2%). High individual variation was present in cognitive control and attention areas, while low individual variation characterized primary sensorimotor regions. Group differences were much smaller than individual differences in the context of MD and its treatment. These results could be linked to the variable findings and difficulty translating research on MD to clinical practice. Future research should examine brain features with low and high individual variation in relation to psychiatric symptoms and treatment trajectories to explore the clinical relevance of the individual differences identified here.

One size does not fit all: notable individual variation in brain activity correlates of antidepressant treatment response

Introduction

To date, no robust electroencephalography (EEG) markers of antidepressant treatment response have been identified. Variable findings may arise from the use of group analyses, which neglect individual variation. Using a combination of group and single-participant analyses, we explored individual variability in EEG characteristics of treatment response.

Methods

Resting-state EEG data and Montgomery-Åsberg Depression Rating Scale (MADRS) symptom scores were collected from 43 patients with depression before, at 1 and 12 weeks of pharmacotherapy. Partial least squares (PLS) was used to: 1) identify group differences in EEG connectivity (weighted phase lag index) and complexity (multiscale entropy) between eventual medication responders and non-responders, and 2) determine whether group patterns could be identified in individual patients.

Results

Responders showed decreased alpha and increased beta connectivity, and early, widespread decreases in complexity over treatment. Non-responders showed an opposite connectivity pattern, and later, spatially confined decreases in complexity. Thus, as in previous studies, our group analyses identified significant differences between groups of patients with different treatment outcomes. These group-level EEG characteristics were only identified in ~40-60% of individual patients, as assessed quantitatively by correlating the spatiotemporal brain patterns between groups and individual results, and by independent raters through visualization.

Discussion

Our single-participant analyses suggest that substantial individual variation exists, and needs to be considered when investigating characteristics of antidepressant treatment response for potential clinical applicability.

Clinical trial registration

https://clinicaltrials.gov, identifier NCT00519428.

Beyond the brain localization of complex traits: Distributed white matter markers of personality

OBJECTIVE

Extensive work in personality neuroscience has shown mixed results in the ability to localize reliable relationships between personality traits and neuroimaging measures. However, recent work in translational neuroimaging has recognized that multifaceted psychological dispositions are not represented in discrete, highly localized brain areas. As such, standard univariate neuroimaging analyses may not be well-suited for capturing broad personality traits supported by distributed networks.

METHOD

The present study uses an out-of-sample predictive modeling approach to identify multivariate signatures of Big Five personality traits within the structural integrity of white matter pathways using diffusion magnetic resonance imaging. In Study 1 (N = 491), we trained a ridge regression model to predict each of the Big Five traits and tested these models in an independent hold-out subsample.

RESULTS

We found that models for both Neuroticism and Openness were significantly related to predictive accuracy in the hold-out sample. Study 2 (N = 108) applied Study 1's predictive models to an independent set of data collected on a different scanner and using a different Big Five scale. Here, we found that the model for Neuroticism remained a significant predictor of individual difference.

CONCLUSION

Our findings provide evidence that this white matter signature of Neuroticism generalizes across differences in measurement and samples.

Challenges of implementing computer-aided diagnostic models for neuroimages in a clinical setting

Advances in artificial intelligence have cultivated a strong interest in developing and validating the clinical utilities of computer-aided diagnostic models. Machine learning for diagnostic neuroimaging has often been applied to detect psychological and neurological disorders, typically on small-scale datasets or data collected in a research setting. With the collection and collation of an ever-growing number of public datasets that researchers can freely access, much work has been done in adapting machine learning models to classify these neuroimages by diseases such as Alzheimer's, ADHD, autism, bipolar disorder, and so on. These studies often come with the promise of being implemented clinically, but despite intense interest in this topic in the laboratory, limited progress has been made in clinical implementation. In this review, we analyze challenges specific to the clinical implementation of diagnostic AI models for neuroimaging data, looking at the differences between laboratory and clinical settings, the inherent limitations of diagnostic AI, and the different incentives and skill sets between research institutions, technology companies, and hospitals. These complexities need to be recognized in the translation of diagnostic AI for neuroimaging from the laboratory to the clinic.

MRI predictors of pharmacotherapy response in major depressive disorder

Major depressive disorder is among the most prevalent psychiatric disorders, exacting a substantial personal, social, and economic toll. Antidepressant treatment typically involves an individualized trial and error approach with an inconsistent success rate. Despite a pressing need, no reliable biomarkers for predicting treatment outcome have yet been discovered. Brain MRI measures hold promise in this regard, though clinical translation remains elusive. In this review, we summarize structural MRI and functional MRI (fMRI) measures that have been investigated as predictors of treatment outcome. We broadly divide these into five categories including three structural measures: volumetric, white matter burden, and white matter integrity; and two functional measures: resting state fMRI and task fMRI. Currently, larger hippocampal volume is the most widely replicated predictor of successful treatment. Lower white matter hyperintensity burden has shown robustness in late life depression. However, both have modest discriminative power. Higher fractional anisotropy of the cingulum bundle and frontal white matter, amygdala hypoactivation and anterior cingulate cortex hyperactivation in response to negative emotional stimuli, and hyperconnectivity within the default mode network (DMN) and between the DMN and executive control network also show promise as predictors of successful treatment. Such network-focused measures may ultimately provide a higher-dimensional measure of treatment response with closer ties to the underlying neurobiology.

Neuroimaging Biomarkers in Schizophrenia

Schizophrenia is a complex neuropsychiatric syndrome with a heterogeneous genetic, neurobiological, and phenotypic profile. Currently, no objective biological measures-that is, biomarkers-are available to inform diagnostic or treatment decisions. Neuroimaging is well positioned for biomarker development in schizophrenia, as it may capture phenotypic variations in molecular and cellular disease targets, or in brain circuits. These mechanistically based biomarkers may represent a direct measure of the pathophysiological underpinnings of the disease process and thus could serve as true intermediate or surrogate endpoints. Effective biomarkers could validate new treatment targets or pathways, predict response, aid in selection of patients for therapy, determine treatment regimens, and provide a rationale for personalized treatments. In this review, the authors discuss a range of mechanistically plausible neuroimaging biomarker candidates, including dopamine hyperactivity, N-methyl-d-aspartate receptor hypofunction, hippocampal hyperactivity, immune dysregulation, dysconnectivity, and cortical gray matter volume loss. They then focus on the putative neuroimaging biomarkers for disease risk, diagnosis, target engagement, and treatment response in schizophrenia. Finally, they highlight areas of unmet need and discuss strategies to advance biomarker development.

Model-based whole-brain effective connectivity to study distributed cognition in health and disease

Neuroimaging techniques are now widely used to study human cognition. The functional associations between brain areas have become a standard proxy to describe how cognitive processes are distributed across the brain network. Among the many analysis tools available, dynamic models of brain activity have been developed to overcome the limitations of original connectivity measures such as functional connectivity. This goes in line with the many efforts devoted to the assessment of directional interactions between brain areas from the observed neuroimaging activity. This opinion article provides an overview of our model-based whole-brain effective connectivity to analyze fMRI data, while discussing the pros and cons of our approach with respect to other established approaches. Our framework relies on the multivariate Ornstein-Uhlenbeck (MOU) process and is thus referred to as MOU-EC. Once tuned, the model provides a directed connectivity estimate that reflects the dynamical state of BOLD activity, which can be used to explore cognition. We illustrate this approach using two applications on task-evoked fMRI data. First, as a connectivity measure, MOU-EC can be used to extract biomarkers for task-specific brain coordination, understood as the patterns of areas exchanging information. The multivariate nature of connectivity measures raises several challenges for whole-brain analysis, for which machine-learning tools present some advantages over statistical testing. Second, we show how to interpret changes in MOU-EC connections in a collective and model-based manner, bridging with network analysis. Our framework provides a comprehensive set of tools that open exciting perspectives to study distributed cognition, as well as neuropathologies.

The relationship between individual differences in gray matter volume and religiosity and mystical experiences: A preregistered voxel-based morphometry study

The neural substrates of religious belief and experience are an intriguing though contentious topic. Here, we had the unique opportunity to establish the relation between validated measures of religiosity and gray matter volume in a large sample of participants (N = 211). In this registered report, we conducted a confirmatory voxel-based morphometry analysis to test three central hypotheses regarding the relationship between religiosity and mystical experiences and gray matter volume. The preregisterered hypotheses, analysis plan, preprocessing and analysis code and statistical brain maps are all available from online repositories. By using a region-of-interest analysis, we found no evidence that religiosity is associated with a reduced volume of the orbito-frontal cortex and changes in the structure of the bilateral inferior parietal lobes. Neither did we find support for the notion that mystical experiences are associated with a reduced volume of the hippocampus, the right middle temporal gyrus or with the inferior parietal lobes. A whole-brain analysis furthermore indicated that no structural brain differences were found in association with religiosity and mystical experiences. We believe that the search for the neural correlates of religious beliefs and experiences should therefore shift focus from studying structural brain differences to a functional and multivariate approach.

Individual structural features constrain the mouse functional connectome

Whole brain dynamics intuitively depend upon the internal wiring of the brain; but to which extent the individual structural connectome constrains the corresponding functional connectome is unknown, even though its importance is uncontested. After acquiring structural data from individual mice, we virtualized their brain networks and simulated in silico functional MRI data. Theoretical results were validated against empirical awake functional MRI data obtained from the same mice. We demonstrate that individual structural connectomes predict the functional organization of individual brains. Using a virtual mouse brain derived from the Allen Mouse Brain Connectivity Atlas, we further show that the dominant predictors of individual structure-function relations are the asymmetry and the weights of the structural links. Model predictions were validated experimentally using tracer injections, identifying which missing connections (not measurable with diffusion MRI) are important for whole brain dynamics in the mouse. Individual variations thus define a specific structural fingerprint with direct impact upon the functional organization of individual brains, a key feature for personalized medicine.

Extracting orthogonal subject- and condition-specific signatures from fMRI data using whole-brain effective connectivity

The study of brain communication based on fMRI data is often limited because such measurements are a mixture of session-to-session variability with subject- and condition-related information. Disentangling these contributions is crucial for real-life applications, in particular when only a few recording sessions are available. The present study aims to define a reliable standard for the extraction of multiple signatures from fMRI data, while verifying that they do not mix information about the different modalities (e.g., subjects and conditions such as tasks performed by them). In particular, condition-specific signatures should not be contaminated by subject-related information, since they aim to generalize over subjects. Practically, signatures correspond to subnetworks of directed interactions between brain regions (typically 100 covering the whole brain) supporting the subject and condition identification for single fMRI sessions. The key for robust prediction is using effective connectivity instead of functional connectivity. Our method demonstrates excellent generalization capabilities for subject identification in two datasets, using only a few sessions per subject as reference. Using another dataset with resting state and movie viewing, we show that the two signatures related to subjects and tasks correspond to distinct subnetworks, which are thus topologically orthogonal. Our results set solid foundations for applications tailored to individual subjects, such as clinical diagnostic.

Connecting brain responsivity and real-world risk taking: Strengths and limitations of current methodological approaches

In line with the goal of limiting health risk behaviors in adolescence, a growing literature investigates whether individual differences in functional brain responses can be related to vulnerability to engage in risky decision-making. We review this body of work, investigate when and in what way findings converge, and provide best practice recommendations. We identified 23 studies that examined individual differences in brain responsivity and adolescent risk taking. Findings varied widely in terms of the neural regions identified as relating to risky behavior. This heterogeneity is likely due to the abundance of approaches used to assess risk taking, and to the disparity of fMRI tasks. Indeed, brain-behavior correlations were typically found in regions showing a main effect of task. However, results from a test of publication bias suggested that region of interest approaches lacked evidential value. The findings suggest that neural factors differentiating riskier teens are not localized to a single region. Therefore, approaches that utilize data from the entire brain, particularly in predictive analyses, may yield more reliable and applicable results. We discuss several decision points that researchers should consider when designing a study, and emphasize the importance of precise research questions that move beyond a general desire to address adolescent risk taking.

Discriminating cognitive status in Parkinson's disease through functional connectomics and machine learning

There is growing interest in the potential of neuroimaging to help develop non-invasive biomarkers in neurodegenerative diseases. In this study, connection-wise patterns of functional connectivity were used to distinguish Parkinson’s disease patients according to cognitive status using machine learning. Two independent subject samples were assessed with resting-state fMRI. The first (training) sample comprised 38 healthy controls and 70 Parkinson’s disease patients (27 with mild cognitive impairment). The second (validation) sample included 25 patients (8 with mild cognitive impairment). The Brainnetome atlas was used to reconstruct the functional connectomes. Using a support vector machine trained on features selected through randomized logistic regression with leave-one-out cross-validation, a mean accuracy of 82.6% (p < 0.002) was achieved in separating patients with mild cognitive impairment from those without it in the training sample. The model trained on the whole training sample achieved an accuracy of 80.0% when used to classify the validation sample (p = 0.006). Correlation analyses showed that the connectivity level in the edges most consistently selected as features was associated with memory and executive function performance in the patient group. Our results demonstrate that connection-wise patterns of functional connectivity may be useful for discriminating Parkinson’s disease patients according to the presence of cognitive deficits.