Sparse Bayesian modeling of hierarchical independent component analysis: Reliable estimation of individual differences in brain networks.

Independent component analysis (ICA) is one of the leading approaches for studying brain functional networks. There is increasing interest in neuroscience studies to investigate individual differences in brain networks and their association with demographic characteristics and clinical outcomes. In this work, we develop a sparse Bayesian group hierarchical ICA model that offers significant improvements over existing ICA techniques for identifying covariate effects on the brain network. Specifically, we model the population-level ICA source signals for brain networks using a Dirichlet process mixture. To reliably capture individual differences on brain networks, we propose sparse estimation of the covariate effects in the hierarchical ICA model via a horseshoe prior. Through extensive simulation studies, we show that our approach performs considerably better in detecting covariate effects in comparison with the leading group ICA methods. We then perform an ICA decomposition of a between-subject meditation study. Our method is able to identify significant effects related to meditative practice in brain regions that are consistent with previous research into the default mode network, whereas other group ICA approaches find few to no effects.

Ranking brain areas encoding the perceived level of pain from fMRI data.

Pain perception is thought to emerge from the integrated activity of a distributed brain system, but the relative contribution of the different network nodes is still incompletely understood. In the present functional magnetic resonance imaging (fMRI) study, we aimed to identify the more relevant brain regions to explain the time profile of the perceived pain intensity in healthy volunteers, during noxious chemical stimulation (ascorbic acid injection) of the left hand. To this end, we performed multi-way partial least squares regression of fMRI data from twenty-two a-priori defined brain regions of interest (ROI) in each hemisphere, to build a model that could efficiently reproduce the psychophysical pain profiles in the same individuals; moreover, we applied a novel three-way extension of the variable importance in projection (VIP) method to summarize each ROI contribution to the model. Brain regions showing the highest VIP scores included the bilateral mid-cingulate, anterior and posterior insular, and parietal operculum cortices, the contralateral paracentral lobule, bilateral putamen and ipsilateral medial thalamus. Most of these regions, with the exception of medial thalamus, were also identified by a statistical analysis on mean ROI beta values estimated using the time course of the psychophysical rating as a regressor at the voxel level. Our results provide the first rank-ordering of brain regions involved in coding the perceived level of pain. These findings in a model of acute prolonged pain confirm and extend previous data, suggesting that a bilateral array of cortical areas and subcortical structures is involved in pain perception.

Human parietofrontal networks related to action observation detected at rest.

Recent data show a broad correspondence between human resting-state and task-related brain networks. We performed a functional magnetic resonance imaging (fMRI) study to compare, in the same subjects, the spatial independent component analysis (ICA) maps obtained at rest and during the observation of either reaching/grasping hand actions or matching static pictures. Two parietofrontal networks were identified by ICA from action observation task data. One network, specific to reaching/grasping observation, included portions of the anterior intraparietal cortex and of the dorsal and ventral lateral premotor cortices. A second network included more posterior portions of the parietal lobe, the dorsomedial frontal cortex, and more anterior and ventral parts, respectively, of the dorsal and ventral premotor cortices, extending toward Broca's area; this network was more generally related to the observation of hand action and static pictures. A good spatial correspondence was found between the 2 observation-related ICA maps and 2 ICA maps identified from resting-state data. The anatomical connectivity among the identified clusters was tested in the same volunteers, using persistent angular structure-MRI and deterministic tractography. These findings extend available knowledge of human parietofrontal circuits and further support the hypothesis of a persistent coherence within functionally relevant networks during rest.

Resting-state networks and anosognosia in Alzheimer's disease.

Background: Recent evidence suggests that anosognosia or unawareness of cognitive impairment in Alzheimer's Disease (AD) may be explained by a disconnection between brain regions involved in accessing and monitoring information regarding self and others. It has been demonstrated that AD patients with anosognosia have reduced connectivity within the default mode network (DMN) and that anosognosia in people with prodromal AD is positively associated with bilateral anterior cingulate cortex (ACC), suggesting a possible role of this region in mechanisms of awareness in the early phase of disease. We hypothesized that anosognosia in AD is associated with an imbalance between the activity of large-scale resting-state functional magnetic resonance imaging (fMRI) networks, in particular the DMN, the salience network (SN), and the frontoparietal network (FPN). Methods: Sixty patients with MCI and AD dementia underwent fMRI and neuropsychological assessment including the Anosognosia Questionnaire Dementia (AQ-D), a measure of anosognosia based on a discrepancy score between patient's and carer's judgments. After having applied Independent Component Analysis (ICA) to resting fMRI data we performed: (i) correlations between the AQ-D score and functional connectivity in the DMN, SN, and FPN, and (ii) comparisons between aware and unaware patients of the DMN, SN, and FPN functional connectivity. Results: We found that anosognosia was associated with (i) weak functional connectivity within the DMN, in posterior and middle cingulate cortex particularly, (ii) strong functional connectivity within the SN in ACC, and between the SN and basal ganglia, and (iii) a heterogenous effect concerning the functional connectivity of the FPN, with a weak connectivity between the FPN and PCC, and a strong connectivity between the FPN and ACC. The observed effects were controlled for differences in severity of cognitive impairment and age. Conclusion: Anosognosia in the AD continuum is associated with a dysregulation of the functional connectivity of three large-scale networks, namely the DMN, SN, and FPN.

Age-specific prevalence of the different clinical presentations of AD and FTD in young-onset dementia.

BACKGROUND: Studies have shown that the prevalence of all-variants Alzheimer's disease (AD) and frontotemporal dementia (FTD) both increase with age, even before the age of 65. However, it is not known whether their different clinical presentations all increase in prevalence with age in the same way. METHODS: We studied the prevalence of the different clinical presentations of young-onset AD and FTD by 5-year age groups in a population-based study identifying all dementia patients with a diagnosis of AD and FTD and symptoms onset before age 65 in the Modena province, Italy. By using regression models of cumulative occurrences, we also estimated age-specific prevalence and compared the growth curves of the clinical presentations. RESULTS: The prevalence of all-variants AD increased with age, from 18/1,000,000 in the 40-44 age group to 1411/1,000,000 in the 60-64 age group. The prevalence of all-variants FTD also increased with age, from 18/1,000,000 to 866/1,000,000. An estimation of age-specific prevalence functions of each clinical presentation showed that atypical non-amnestic AD and aphasic FTD grew the most in early ages, followed by the behavioural variant of FTD (bvFTD). Then, around the age of 60, amnestic AD took over and its age-specific prevalence continued to increase disproportionally compared to all the other clinical variants of AD and FTD, which, instead, started to decrease in prevalence. CONCLUSIONS: Amnestic AD is the clinical presentation that increases the most with advancing age, followed by bvFTD, suggesting that there is a differential vulnerability to the effect of ageing within the same neurodegenerative disease.

Dynamical properties of BOLD activity from the ventral posteromedial cortex associated with meditation and attentional skills.

Neuroimaging data suggest a link between the spontaneous production of thoughts during wakeful rest and slow fluctuations of activity in the default mode network (DMN), a set of brain regions with high basal metabolism and a major neural hub in the ventral posteromedial cortex (vPMC). Meta-awareness and regulation of mind-wandering are core cognitive components of most contemplative practices and to study their impact on DMN activity, we collected functional MRI (fMRI) data from a cohort of experienced Zen meditators and meditation-naive controls engaging in a basic attention-to-breathing protocol. We observed a significant group difference in the skewness of the fMRI BOLD signal from the vPMC, suggesting that the relative incidence of states of elevated vPMC activity was lower in meditators; furthermore, the same parameter was significantly correlated with performance on a rapid visual information processing (RVIP) test for sustained attention conducted outside the scanner. Finally, a functional connectivity analysis with the vPMC seed revealed a significant association of RVIP performance with the degree of temporal correlation between vPMC and the right temporoparietal junction (TPJ), a region strongly implicated in stimulus-triggered reorienting of attention. Together, these findings suggest that the vPMC BOLD signal skewness and the temporal relationship of vPMC and TPJ activities reflect the dynamic tension between mind-wandering, meta-awareness, and directed attention, and may represent a useful endophenotype for studying individual differences in attentional abilities and the impairment of the latter in specific clinical conditions.

Neural correlates of emotional valence for faces and words.

Stimuli with negative emotional valence are especially apt to influence perception and action because of their crucial role in survival, a property that may not be precisely mirrored by positive emotional stimuli of equal intensity. The aim of this study was to identify the neural circuits differentially coding for positive and negative valence in the implicit processing of facial expressions and words, which are among the main ways human beings use to express emotions. Thirty-six healthy subjects took part in an event-related fMRI experiment. We used an implicit emotional processing task with the visual presentation of negative, positive, and neutral faces and words, as primary stimuli. Dynamic Causal Modeling (DCM) of the fMRI data was used to test effective brain connectivity within two different anatomo-functional models, for the processing of words and faces, respectively. In our models, the only areas showing a significant differential response to negative and positive valence across both face and word stimuli were early visual cortices, with faces eliciting stronger activations. For faces, DCM revealed that this effect was mediated by a facilitation of activity in the amygdala by positive faces and in the fusiform face area by negative faces; for words, the effect was mainly imputable to a facilitation of activity in the primary visual cortex by positive words. These findings support a role of early sensory cortices in discriminating the emotional valence of both faces and words, where the effect may be mediated chiefly by the subcortical/limbic visual route for faces, and rely more on the direct thalamic pathway to primary visual cortex for words.

Altered resting-state effective connectivity of fronto-parietal motor control systems on the primary motor network following stroke.

Previous brain imaging work suggests that stroke alters the effective connectivity (the influence neural regions exert upon each other) of motor execution networks. The present study examines the intrinsic effective connectivity of top-down motor control in stroke survivors (n=13) relative to healthy participants (n=12). Stroke survivors exhibited significant deficits in motor function, as assessed by the Fugl-Meyer Motor Assessment. We used structural equation modeling (SEM) of resting-state fMRI data to investigate the relationship between motor deficits and the intrinsic effective connectivity between brain regions involved in motor control and motor execution. An exploratory adaptation of SEM determined the optimal model of motor execution effective connectivity in healthy participants, and confirmatory SEM assessed stroke survivors' fit to that model. We observed alterations in spontaneous resting-state effective connectivity from fronto-parietal guidance systems to the motor network in stroke survivors. More specifically, diminished connectivity was found in connections from the superior parietal cortex to primary motor cortex and supplementary motor cortex. Furthermore, the paths demonstrated large individual variance in stroke survivors but less variance in healthy participants. These findings suggest that characterizing the deficits in resting-state connectivity of top-down processes in stroke survivors may help optimize cognitive and physical rehabilitation therapies by individually targeting specific neural pathway.

Emergence of associative learning in a neuromorphic inference network.

Objective. In the theoretical framework of predictive coding and active inference, the brain can be viewed as instantiating a rich generative model of the world that predicts incoming sensory data while continuously updating its parameters via minimization of prediction errors. While this theory has been successfully applied to cognitive processes-by modelling the activity of functional neural networks at a mesoscopic scale-the validity of the approach when modelling neurons as an ensemble of inferring agents, in a biologically plausible architecture, remained to be explored.Approach.We modelled a simplified cerebellar circuit with individual neurons acting as Bayesian agents to simulate the classical delayed eyeblink conditioning protocol. Neurons and synapses adjusted their activity to minimize their prediction error, which was used as the network cost function. This cerebellar network was then implemented in hardware by replicating digital neuronal elements via a low-power microcontroller.Main results. Persistent changes of synaptic strength-that mirrored neurophysiological observations-emerged via local (neurocentric) prediction error minimization, leading to the expression of associative learning. The same paradigm was effectively emulated in low-power hardware showing remarkably efficient performance compared to conventional neuromorphic architectures.Significance. These findings show that: (a) an ensemble of free energy minimizing neurons-organized in a biological plausible architecture-can recapitulate functional self-organization observed in nature, such as associative plasticity, and (b) a neuromorphic network of inference units can learn unsupervised tasks without embedding predefined learning rules in the circuit, thus providing a potential avenue to a novel form of brain-inspired artificial intelligence.

From Generative Models to Generative Passages: A Computational Approach to (Neuro) Phenomenology.

This paper presents a version of neurophenomenology based on generative modelling techniques developed in computational neuroscience and biology. Our approach can be described as computational phenomenology because it applies methods originally developed in computational modelling to provide a formal model of the descriptions of lived experience in the phenomenological tradition of philosophy (e.g., the work of Edmund Husserl, Maurice Merleau-Ponty, etc.). The first section presents a brief review of the overall project to naturalize phenomenology. The second section presents and evaluates philosophical objections to that project and situates our version of computational phenomenology with respect to these projects. The third section reviews the generative modelling framework. The final section presents our approach in detail. We conclude by discussing how our approach differs from previous attempts to use generative modelling to help understand consciousness. In summary, we describe a version of computational phenomenology which uses generative modelling to construct a computational model of the inferential or interpretive processes that best explain this or that kind of lived experience.

Bayesian Joint Modeling of Multiple Brain Functional Networks.

Investigating the similarity and changes in brain networks under different mental conditions has become increasingly important in neuroscience research. A standard separate estimation strategy fails to pool information across networks and hence has reduced estimation accuracy and power to detect between-network differences. Motivated by a fMRI Stroop task experiment that involves multiple related tasks, we develop an integrative Bayesian approach for jointly modeling multiple brain networks that provides a systematic inferential framework for network comparisons. The proposed approach explicitly models shared and differential patterns via flexible Dirichlet process-based priors on edge probabilities. Conditional on edges, the connection strengths are modeled via Bayesian spike and slab prior on the precision matrix off-diagonals. Numerical simulations illustrate that the proposed approach has increased power to detect true differential edges while providing adequate control on false positives and achieves greater network estimation accuracy compared to existing methods. The Stroop task data analysis reveals greater connectivity differences between task and fixation that are concentrated in brain regions previously identified as differentially activated in Stroop task, and more nuanced connectivity differences between exertion and relaxed task. In contrast, penalized modeling approaches involving computationally burdensome permutation tests reveal negligible network differences between conditions that seem biologically implausible.

The Impact of Mindfulness Meditation on the Wandering Mind: a Systematic Review.

Through the practice of Mindfulness Meditation (MM), meditators become familiar with the observation of ongoing spontaneous thoughts, while maintaining an attitude of openness and equanimity. The aim of this systematic review is to present a synthesis of available findings of the short and long-term effects of MM on mind wandering (MW). We included studies that considered both first-person and behavioral/physiological measures of MW. The search resulted in 2035 papers, 24 of which were eligible. Reviewed studies revealed a high heterogeneity in designs, outcome measures and interventions. Most of the pre-post intervention studies showed that a protracted practice of MM (at least 2 weeks) reduced MW, limiting its negative effects on different cognitive tasks. Cross-sectional studies highlighted differences between expert meditators and naïve individuals: meditators self-reported less MW and showed decreased Default Mode Network activity, during meditation and resting-state. Further studies are needed to replicate available findings and to more deeply explore how MW is influenced by meditation, also considering its qualitative characteristics that remain largely unexplored.

Modeling Early Phases of COVID-19 Pandemic in Northern Italy and Its Implication for Outbreak Diffusion.

The COVID-19 pandemic has sparked an intense debate about the hidden factors underlying the dynamics of the outbreak. Several computational models have been proposed to inform effective social and healthcare strategies. Crucially, the predictive validity of these models often depends upon incorporating behavioral and social responses to infection. Among these tools, the analytic framework known as "dynamic causal modeling" (DCM) has been applied to the COVID-19 pandemic, shedding new light on the factors underlying the dynamics of the outbreak. We have applied DCM to data from northern Italian regions, the first areas in Europe to contend with the outbreak, and analyzed the predictive validity of the model and also its suitability in highlighting the hidden factors governing the pandemic diffusion. By taking into account data from the beginning of the pandemic, the model could faithfully predict the dynamics of outbreak diffusion varying from region to region. The DCM appears to be a reliable tool to investigate the mechanisms governing the spread of the SARS-CoV-2 to identify the containment and control strategies that could efficiently be used to counteract further waves of infection.

Reward-related brain activity and behavior are associated with peripheral ghrelin levels in obesity.

BACKGROUND/OBJECTIVES: While excessive food consumption represents a key factor in the development of obesity, the underlying mechanisms are still unclear. Ghrelin, a gut-brain hormone involved in the regulation of appetite, is impaired in obesity. In addition to its role in eating behavior, this hormone was shown to affect brain regions controlling reward, including the striatum and prefrontal cortex, and there is strong evidence of impaired reward processing in obesity. The present study investigated the possibility that disrupted reward-related brain activity in obesity relates to ghrelin deficiency. SUBJECTS/METHODS: Fifteen severely obese subjects (BMI > 35 kg/m2) and fifteen healthy non-obese control subjects (BMI < 30 kg/m2) were recruited. A guessing-task paradigm, previously shown to activate the ventral striatum, was used to assess reward-related brain neural activity by functional magnetic resonance imaging (fMRI). Fasting blood samples were collected for the measurement of circulating ghrelin. RESULTS: Significant activations in the ventral striatum, ventromedial prefrontal cortex and extrastriate visual cortex were elicited by the fMRI task in both obese and control subjects. In addition, greater reward-related activations were present in the dorsolateral prefrontal cortex, and precuneus/posterior cingulate of obese subjects compared to controls. Obese subjects exhibited longer choice times after repeated reward and lower circulating ghrelin levels than lean controls. Reduced ghrelin levels significantly predicted slower post-reward choices and reward-related hyperactivity in dorsolateral prefrontal cortices in obese subjects. CONCLUSION: This study provides evidence of association between circulating ghrelin and reward-related brain activity in obesity and encourages further exploration of the role of ghrelin system in altered eating behavior in obesity.

The contemplative exercise through the lenses of predictive processing: A promising approach.

The theory of predictive processing in the comprehensive articulation proposed by Karl Friston is a framework that boasts an impressively wide explanatory power in neurobiology, where processes apparently as diverse as perception, action, attention, and learning unfold, and are coherently orchestrated, according to the single general mandate of free-energy minimization. In the present opinion piece, I argue that the adoption of this theoretical perspective can provide a much needed unitary framework for contemplative research as well, whose explosive growth in terms of the number of published studies and amount of collected data has not been matched yet by a similarly extensive effort to theoretically organize the findings, so that a deeper understanding of meditation-related processes can be attained. After an introduction to the basic notions of predictive processing, a tentative application of the latter to the meditative exercise is discussed, taking as a paradigmatic example the Japanese Zen meditation practice of shikantaza. Finally, I provide a short list of experimental paradigms that seem particularly useful to test the hypotheses born out of the predictive processing approach to contemplative research.

The epistemic and pragmatic value of non-action: a predictive coding perspective on meditation.

The surge of interest about mindfulness meditation is associated with a growing empirical evidence about its impact on the mind and body. Yet, despite promising phenomenological or psychological models of mindfulness, a general mechanistic understanding of meditation steeped in neuroscience is still lacking. In parallel, predictive processing approaches to the mind are rapidly developing in the cognitive sciences with an impressive explanatory power: processes apparently as diverse as perception, action, attention, and learning, can be seen as unfolding and being coherently orchestrated according to the single general mandate of free-energy minimization. Here, we briefly explore the possibility to supplement previous phenomenological models of focused attention meditation by formulating them in terms of active inference. We first argue that this perspective can account for how paying voluntary attention to the body in meditation helps settling the mind by downweighting habitual and automatic trajectories of (pre)motor and autonomic reactions, as well as the pull of distracting spontaneous thought at the same time. Secondly, we discuss a possible relationship between phenomenological notions such as opacity and de-reification, and the deployment of precision-weighting via the voluntary allocation of attention. We propose the adoption of this theoretical framework as a promising strategy for contemplative research. Explicit computational simulations and comparisons with experimental and phenomenological data will be critical to fully develop this approach.

Short-term mindfulness practice attenuates reward prediction errors signals in the brain.

Activity changes in dopaminergic neurons encode the ongoing discrepancy between expected and actual value of a stimulus, providing a teaching signal for a reward prediction process. Previous work comparing a cohort of long-term Zen meditators to controls demonstrated an attenuation of reward prediction signals to appetitive reward in the striatum. Using a cross-commodity design encompassing primary- and secondary-reward conditioning experiments, the present study asks the question of whether reward prediction signals are causally altered by mindfulness training in naïve subjects. Volunteers were randomly assigned to 8 weeks of mindfulness training (MT), active control training (CT), or a one-time mindfulness induction group (MI). We observed a decreased response to positive prediction errors in the putamen in the MT group compared to CT using both a primary and a secondary-reward experiment. Furthermore, the posterior insula showed greater activation to primary rewards, independently of their predictability, in the MT group, relative to CT and MI group. These results support the notion that increased attention to the present moment and its interoceptive features - a core component of mindfulness practice - may reduce predictability effects in reward processing, without dampening (in fact, enhancing) the response to the actual delivery of the stimulus.

A neural basis for social cooperation.

Cooperation based on reciprocal altruism has evolved in only a small number of species, yet it constitutes the core behavioral principle of human social life. The iterated Prisoner's Dilemma Game has been used to model this form of cooperation. We used fMRI to scan 36 women as they played an iterated Prisoner's Dilemma Game with another woman to investigate the neurobiological basis of cooperative social behavior. Mutual cooperation was associated with consistent activation in brain areas that have been linked with reward processing: nucleus accumbens, the caudate nucleus, ventromedial frontal/orbitofrontal cortex, and rostral anterior cingulate cortex. We propose that activation of this neural network positively reinforces reciprocal altruism, thereby motivating subjects to resist the temptation to selfishly accept but not reciprocate favors.

Human striatal responses to monetary reward depend on saliency.

While the striatum has been implicated in reward processing, an alternative view contends that the striatum processes salient events in general. Using fMRI, we investigated human striatal responses to monetary reward while modulating the saliency surrounding its receipt. Money was maximally salient when its receipt depended on a correct response (active) and minimally salient when its receipt was completely independent of the task (passive). The saliency manipulation was confirmed by skin conductance responses and subjective ratings of the stimuli. Significant caudate and nucleus accumbens activations occurred following the active compared to passive money. Such activations were attributed to saliency rather than the motor requirement associated with the active money because striatal activations were not observed when the money was replaced by inconsequential, nonrewarding stimuli. The present study provides evidence that the striatum's role in reward processing is dependent on the saliency associated with reward, rather than value or hedonic feelings.

A unified framework for group independent component analysis for multi-subject fMRI data.

Independent component analysis (ICA) is becoming increasingly popular for analyzing functional magnetic resonance imaging (fMRI) data. While ICA has been successfully applied to single-subject analysis, the extension of ICA to group inferences is not straightforward and remains an active topic of research. Current group ICA models, such as the GIFT [Calhoun, V.D., Adali, T., Pearlson, G.D., Pekar, J.J., 2001. A method for making group inferences from functional MRI data using independent component analysis. Hum. Brain Mapp. 14, 140-151.] and tensor PICA [Beckmann, C.F., Smith, S.M., 2005. Tensorial extensions of independent component analysis for multisubject FMRI analysis. Neuroimage 25, 294-311.], make different assumptions about the underlying structure of the group spatio-temporal processes and are thus estimated using algorithms tailored for the assumed structure, potentially leading to diverging results. To our knowledge, there are currently no methods for assessing the validity of different model structures in real fMRI data and selecting the most appropriate one among various choices. In this paper, we propose a unified framework for estimating and comparing group ICA models with varying spatio-temporal structures. We consider a class of group ICA models that can accommodate different group structures and include existing models, such as the GIFT and tensor PICA, as special cases. We propose a maximum likelihood (ML) approach with a modified Expectation-Maximization (EM) algorithm for the estimation of the proposed class of models. Likelihood ratio tests (LRT) are presented to compare between different group ICA models. The LRT can be used to perform model comparison and selection, to assess the goodness-of-fit of a model in a particular data set, and to test group differences in the fMRI signal time courses between subject subgroups. Simulation studies are conducted to evaluate the performance of the proposed method under varying structures of group spatio-temporal processes. We illustrate our group ICA method using data from an fMRI study that investigates changes in neural processing associated with the regular practice of Zen meditation.