Neural dynamics of emotion and cognition: From trajectories to underlying neural geometry

How can we study, characterize, and understand the neural underpinnings of cognitive-emotional behaviors as inherently dynamic processes? In the past 50 years, Stephen Grossberg has developed a research program that embraces the themes of dynamics, decentralized computation, emergence, selection and competition, and autonomy. The present paper discusses how these principles can be heeded by experimental scientists to advance the understanding of the brain basis of behavior. It is suggested that a profitable way forward is to focus on investigating the dynamic multivariate structure of brain data. Accordingly, central research problems involve characterizing "neural trajectories" and the associated geometry of the underlying "neural space." Finally, it is argued that, at a time when the development of neurotechniques has reached a fever pitch, neuroscience needs to redirect its focus and invest comparable energy in the conceptual and theoretical dimensions of its research endeavor. Otherwise we run the risk of being able to measure "every atom" in the brain in a theoretical vacuum.

Representational Organization of Novel Task Sets during Proactive Encoding

Recent multivariate analyses of brain data have boosted our understanding of the organizational principles that shape neural coding. However, most of this progress has focused on perceptual visual regions (Connolly et al., 2012), whereas far less is known about the organization of more abstract, action-oriented representations. In this study, we focused on humans' remarkable ability to turn novel instructions into actions. While previous research shows that instruction encoding is tightly linked to proactive activations in frontoparietal brain regions, little is known about the structure that orchestrates such anticipatory representation. We collected fMRI data while participants (both males and females) followed novel complex verbal rules that varied across control-related variables (integrating within/across stimuli dimensions, response complexity, target category) and reward expectations. Using representational similarity analysis (Kriegeskorte et al., 2008), we explored where in the brain these variables explained the organization of novel task encoding, and whether motivation modulated these representational spaces. Instruction representations in the lateral PFC were structured by the three control-related variables, whereas intraparietal sulcus encoded response complexity and the fusiform gyrus and precuneus organized its activity according to the relevant stimulus category. Reward exerted a general effect, increasing the representational similarity among different instructions, which was robustly correlated with behavioral improvements. Overall, our results highlight the flexibility of proactive task encoding, governed by distinct representational organizations in specific brain regions. They also stress the variability of motivation-control interactions, which appear to be highly dependent on task attributes, such as complexity or novelty.SIGNIFICANCE STATEMENT In comparison with other primates, humans display a remarkable success in novel task contexts thanks to our ability to transform instructions into effective actions. This skill is associated with proactive task-set reconfigurations in frontoparietal cortices. It remains yet unknown, however, how the brain encodes in anticipation the flexible, rich repertoire of novel tasks that we can achieve. Here we explored cognitive control and motivation-related variables that might orchestrate the representational space for novel instructions. Our results showed that different dimensions become relevant for task prospective encoding, depending on the brain region, and that the lateral PFC simultaneously organized task representations following different control-related variables. Motivation exerted a general modulation upon this process, diminishing rather than increasing distances among instruction representations.

Neural architecture of the vertebrate brain: implications for the interaction between emotion and cognition

Cognition is considered a hallmark of the primate brain that requires a high degree of signal integration, such as achieved in the prefrontal cortex. Moreover, it is often assumed that cognitive capabilities imply "superior" computational mechanisms compared to those involved in emotion or motivation. In contrast to these ideas, we review data on the neural architecture across vertebrates that support the concept that association and integration are basic features of the vertebrate brain, which are needed to successfully adapt to a changing world. This property is not restricted to a few isolated brain centers, but rather resides in neuronal networks working collectively in a context-dependent manner. In different vertebrates, we identify shared large-scale connectional systems involving the midbrain, hypothalamus, thalamus, basal ganglia, and amygdala. The high degree of crosstalk and association between these systems at different levels supports the notion that cognition, emotion, and motivation cannot be separated - all of them involve a high degree of signal integration.

An integrative Bayesian approach to matrix-based analysis in neuroimaging

Understanding the correlation structure associated with brain regions is a central goal in neuroscience, as it informs about interregional relationships and network organization. Correlation structure can be conveniently captured in a matrix that indicates the relationships among brain regions, which could involve electroencephalogram sensors, electrophysiology recordings, calcium imaging data, or functional magnetic resonance imaging (FMRI) data-We call this type of analysis matrix-based analysis, or MBA. Although different methods have been developed to summarize such matrices across subjects, including univariate general linear models (GLMs), the available modeling strategies tend to disregard the interrelationships among the regions, leading to "inefficient" statistical inference. Here, we develop a Bayesian multilevel (BML) modeling framework that simultaneously integrates the analyses of all regions, region pairs (RPs), and subjects. In this approach, the intricate relationships across regions as well as across RPs are quantitatively characterized. The adoption of the Bayesian framework allows us to achieve three goals: (a) dissolve the multiple testing issue typically associated with seeking evidence for the effect of each RP under the conventional univariate GLM; (b) make inferences on effects that would be treated as "random" under the conventional linear mixed-effects framework; and (c) estimate the effect of each brain region in a manner that indexes their relative "importance". We demonstrate the BML methodology with an FMRI dataset involving a cognitive-emotional task and compare it to the conventional GLM approach in terms of model efficiency, performance, and inferences. The associated program MBA is available as part of the AFNI suite for general use.

What drives prioritized visual processing? A motivational relevance account

Emotion is fundamental to our being, and an essential aspect guiding behavior when rapid responding is required. This includes whether we approach or avoid a stimulus, and the accompanying physiological responses. A common tenet is that threat-related content drives stimulus processing and biases visual attention, so that rapid responding can be initiated. In this paper, it will be argued instead that prioritization of threatening stimuli should be encompassed within a motivational relevance framework. To more fully understand what is, or is not, prioritized for visual processing one must, however, additionally consider: (i) stimulus ambiguity and perceptual saliency; (ii) task demands, including both perceptual load and cognitive load; and (iii) endogenous/affective states of the individual. Combined with motivational relevance, this then leads to a multifactorial approach to understanding the drivers of prioritized visual processing. This accords with current recognition that the brain basis allowing for visual prioritization is also multifactorial, including transient, dynamic and overlapping networks. Taken together, the paper provides a reconceptualization of how "emotional" information prioritizes visual processing.

Dynamic Threat Processing

During real-life situations, multiple factors interact dynamically to determine threat level. In the current fMRI study involving healthy adult human volunteers, we investigated interactions between proximity, direction (approach vs. retreat), and speed during a dynamic threat-of-shock paradigm. As a measure of threat-evoked physiological arousal, skin conductance responses were recorded during fMRI scanning. Some brain regions tracked individual threat-related factors, and others were also sensitive to combinations of these variables. In particular, signals in the anterior insula tracked the interaction between proximity and direction where approach versus retreat responses were stronger when threat was closer compared with farther. A parallel proximity-by-direction interaction was also observed in physiological skin conductance responses. In the right amygdala, we observed a proximity by direction interaction, but intriguingly in the opposite direction as the anterior insula; retreat versus approach responses were stronger when threat was closer compared with farther. In the right bed nucleus of the stria terminalis, we observed an effect of threat proximity, whereas in the right periaqueductal gray/midbrain we observed an effect of threat direction and a proximity by direction by speed interaction (the latter was detected in exploratory analyses but not in a voxelwise fashion). Together, our study refines our understanding of the brain mechanisms involved during aversive anticipation in the human brain. Importantly, it emphasizes that threat processing should be understood in a manner that is both context-sensitive and dynamic.

Altered segregation between task-positive and task-negative regions in mild traumatic brain injury

Changes in large-scale brain networks that accompany mild traumatic brain injury (mTBI) were investigated using functional magnetic resonance imaging (fMRI) during the N-back working memory task at two cognitive loads (1-back and 2-back). Thirty mTBI patients were examined during the chronic stage of injury and compared to 28 control participants. Demographics and behavioral performance were matched across groups. Due to the diffuse nature of injury, we hypothesized that there would be an imbalance in the communication between task-positive and Default Mode Network (DMN) regions in the context of effortful task execution. Specifically, a graph-theoretic measure of modularity was used to quantify the extent to which groups of brain regions tended to segregate into task-positive and DMN sub-networks. Relative to controls, mTBI patients showed reduced segregation between the DMN and task-positive networks, but increased functional connectivity within the DMN regions during the more cognitively demanding 2-back task. Together, our findings reveal that patients exhibit alterations in the communication between and within neural networks during a cognitively demanding task. These findings reveal altered processes that persist through the chronic stage of injury, highlighting the need for longitudinal research to map the neural recovery of mTBI patients.

Brain dynamics and temporal trajectories during task and naturalistic processing

Human functional Magnetic Resonance Imaging (fMRI) data are acquired while participants engage in diverse perceptual, motor, cognitive, and emotional tasks. Although data are acquired temporally, they are most often treated in a quasi-static manner. Yet, a fuller understanding of the mechanisms that support mental functions necessitates the characterization of dynamic properties. Here, we describe an approach employing a class of recurrent neural networks called reservoir computing, and show the feasibility and potential of using it for the analysis of temporal properties of brain data. We show that reservoirs can be used effectively both for condition classification and for characterizing lower-dimensional "trajectories" of temporal data. Classification accuracy was approximately 90% for short clips of "social interactions" and around 70% for clips extracted from movie segments. Data representations with 12 or fewer dimensions (from an original space with over 300) attained classification accuracy within 5% of the full data. We hypothesize that such low-dimensional trajectories may provide "signatures" that can be associated with tasks and/or mental states. The approach was applied across participants (that is, training in one set of participants, and testing in a separate group), showing that representations generalized well to unseen participants. Taken together, we believe the present approach provides a promising framework to characterize dynamic fMRI information during both tasks and naturalistic conditions.

Diversity in action: exchange of perspectives and reflections on taxonomies of individual differences

Throughout the last 2500 years, the classification of individual differences in healthy people and their extreme expressions in mental disorders has remained one of the most difficult challenges in science that affects our ability to explore individuals' functioning, underlying psychobiological processes and pathways of development. To facilitate analyses of the principles required for studying individual differences, this theme issue brought together prominent scholars from diverse backgrounds of which many bring unique combinations of cross-disciplinary experiences and perspectives that help establish connections and promote exchange across disciplines. This final paper presents brief commentaries of some of our authors and further scholars exchanging perspectives and reflecting on the contributions of this theme issue.This article is part of the theme issue 'Diverse perspectives on diversity: multi-disciplinary approaches to taxonomies of individual differences'.

Embracing integration and complexity: placing emotion within a science of brain and behaviour

The present paper addresses conceptual issues that are central to emotion research. What is emotion? What are its defining characteristics? The field struggles with questions like these almost constantly. I argue that definitions, and deciding what is the proper status of emotion, are not a requirement for scientific progress - in fact, they can hinder it. Therefore, "emotion" researchers should strive to develop a science of complex behaviours, and worry less about their exact nature. But for interesting behaviours, is most of the explaining that is needed present at the level of isolated systems (perception, cognition, etc.) or at the level of interactions between them? I suggest that the level of interactions is where most of the work is needed. Accordingly, I advocate that it is important to embrace integration, and not to strive to necessarily disentangle the multiple contributions underlying behaviours. More generally, it is argued that we need to revise models of causation adopted when reasoning about the mind and brain. Instead, a "complex systems" approach is required where the interactions between multiple components lead to system-level - emergent - properties that cannot be isolated or attributed to more elementary parts.

Author Reply: Placing Emotion Within a Science of Brain and Behavior

In this response, I suggest that the focus of “emotion” researchers should be more on striving to develop a science of brain and behavior than on deciding what is the proper status of emotion. Because structure and function are closely intertwined in biological systems, advancing our understanding of complex behaviors will necessitate researching their brain substrates.

Emotion and the Interactive Brain: Insights From Comparative Neuroanatomy and Complex Systems

Although emotion is closely associated with motivation, and interacts with perception, cognition, and action, many conceptualizations still treat emotion as separate from these domains. Here, a comparative/evolutionary anatomy framework is presented to motivate the idea that long-range, distributed circuits involving the midbrain, thalamus, and forebrain are central to emotional processing. It is proposed that emotion can be understood in terms of large-scale network interactions spanning the neuroaxis that form "functionally integrated systems." At the broadest level, the argument is made that we need to move beyond a Newtonian view of causation to one involving complex systems where bidirectional influences and nonlinearities abound. Therefore, understanding interactions between subsystems and signal integration becomes central to unraveling the organization of the emotional brain.

Potential reward reduces the adverse impact of negative distractor stimuli

Knowledge about interactions between reward and negative processing is rudimentary. Here, we employed functional MRI to probe how potential reward signaled by advance cues alters aversive distractor processing during perception. Behaviorally, the influence of aversive stimuli on task performance was reduced during the reward compared to no-reward condition. In the brain, at the task phase, paralleling the observed behavioral pattern, we observed significant interactions in the anterior insula and dorsal anterior cingulate cortex, such that responses during the negative (vs neutral) condition were reduced during the reward compared to no-reward condition. Notably, negative distractor processing in the amygdala appeared to be independent of the reward manipulation. During the initial cue phase, we observed increased reward-related responses in the ventral striatum/accumbens, which were correlated with behavioral interference scores at the subsequent task phase, revealing that participants with increased reward-related responses exhibited a greater behavioral benefit of reward in reducing the adverse effect of negative images. Furthermore, during processing of reward (vs no-reward) cues, the ventral striatum exhibited stronger functional connectivity with fronto-parietal regions important for attentional control. Together, our findings contribute to the understanding of how potential reward influences attentional control and reduces negative distractor processing in the human brain.

Information about peer choices shapes human risky decision-making

Humans frequently make choices that involve risk for health and well-being. At the same time, information about others’ choices is omnipresent due to new forms of social media and information technology. However, while past research has shown that peers can exert a strong influence on such risky choices, understanding how information about risky decisions of others affects one’s own risky decisions is still lacking. We therefore developed a behavioral task to measure how information about peer choices affects risky decision-making and call it the social Balloon Analogue Risk Task (sBART). We tested this novel paradigm in a sample of 52 college young adults. Here we show that risky decisions were influenced in the direction of the perceived choices of others – riskier choices of others led to riskier behavior whereas safer choices of others led to less risky behavior. These findings indicate that information about peer choices is sufficient to shape one’s own risky behavior.

LEICA: Laplacian eigenmaps for group ICA decomposition of fMRI data

Independent component analysis (ICA) is a data-driven method that has been increasingly used for analyzing functional Magnetic Resonance Imaging (fMRI) data. However, generalizing ICA to multi-subject studies is non-trivial due to the high-dimensionality of the data, the complexity of the underlying neuronal processes, the presence of various noise sources, and inter-subject variability. Current group ICA based approaches typically use several forms of the Principal Component Analysis (PCA) method to extend ICA for generating group inferences. However, linear dimensionality reduction techniques have serious limitations including the fact that the underlying BOLD signal is a complex function of several nonlinear processes. In this paper, we propose an effective non-linear ICA-based model for extracting group-level spatial maps from multi-subject fMRI datasets. We use a non-linear dimensionality reduction algorithm based on Laplacian eigenmaps to identify a manifold subspace common to the group, such that this mapping preserves the correlation among voxels' time series as much as possible. These eigenmaps are modeled as linear mixtures of a set of group-level spatial features, which are then extracted using ICA. The resulting algorithm is called LEICA (Laplacian Eigenmaps for group ICA decomposition). We introduce a number of methods to evaluate LEICA using 100-subject resting state and 100-subject working memory task fMRI datasets from the Human Connectome Project (HCP). The test results show that the extracted spatial maps from LEICA are meaningful functional networks similar to those produced by some of the best known methods. Importantly, relative to state-of-the-art methods, our algorithm compares favorably in terms of the functional cohesiveness of the spatial maps generated, as well as in terms of the reproducibility of the results.

Attentional capture by simultaneous pleasant and unpleasant emotional distractors

Both high-arousal pleasant and unpleasant task-irrelevant stimuli capture attention and divert processing away from the main task leading to impaired behavioral performance in concurrent tasks. Most studies have separately investigated interference effects of unpleasant and pleasant stimuli on behavior. Thus, little is known about how pleasant and unpleasant task-irrelevant stimuli influence behavior simultaneously. In the present study, we investigated this question during a visual-letter search task. We tested two alternative hypotheses about the influence of simultaneous pleasant and unpleasant task-irrelevant stimuli on task performance. If behavior is purely determined by the intensity of the distractor stimuli (independent of valence), then we would expect the interference effect of simultaneous pleasant and unpleasant distractors to be similar to the influence of two pleasant or two unpleasant distractor stimuli. In contrast, because of opponent interactions between appetitive and aversive motivational systems, the interference effect of simultaneous pleasant and unpleasant stimuli might be weakened. We found that the interference effect of a compound pleasant-plus-unpleasant stimulus was greater than that of a neutral-plus-emotional stimulus and similar to that of two pleasant or two unpleasant stimuli. These results suggest that at the level of behavior, the influence of joint pleasant and unpleasant task-irrelevant stimuli during perception is mainly determined by the intensity of the stimuli, and independent of their valence. (PsycINFO Database Record (c) 2018 APA, all rights reserved).

Dynamics of Intersubject Brain Networks during Anxious Anticipation

How do large-scale brain networks reorganize during the waxing and waning of anxious anticipation? Here, threat was dynamically modulated during human functional MRI as two circles slowly meandered on the screen; if they touched, an unpleasant shock was delivered. We employed intersubject correlation analysis, which allowed the investigation of network-level functional connectivity across brains, and sought to determine how network connectivity changed during periods of approach (circles moving closer) and periods of retreat (circles moving apart). Analysis of positive connection weights revealed that dynamic threat altered connectivity within and between the salience, executive, and task-negative networks. For example, dynamic functional connectivity increased within the salience network during approach and decreased during retreat. The opposite pattern was found for the functional connectivity between the salience and task-negative networks: decreases during approach and increases during approach. Functional connections between subcortical regions and the salience network also changed dynamically during approach and retreat periods. Subcortical regions exhibiting such changes included the putative periaqueductal gray, putative habenula, and putative bed nucleus of the stria terminalis. Additional analysis of negative functional connections revealed dynamic changes, too. For example, negative weights within the salience network decreased during approach and increased during retreat, opposite what was found for positive weights. Together, our findings unraveled dynamic features of functional connectivity of large-scale networks and subcortical regions across participants while threat levels varied continuously, and demonstrate the potential of characterizing emotional processing at the level of dynamic networks.

Understanding emotion with brain networks

Emotional processing appears to be interlocked with perception, cognition, motivation, and action. These interactions are supported by the brain's large-scale non-modular anatomical and functional architectures. An important component of this organization involves characterizing the brain in terms of networks. Two aspects of brain networks are discussed: brain networks should be considered as inherently overlapping (not disjoint) and dynamic (not static). Recent work on multivariate pattern analysis shows that affective dimensions can be detected in the activity of distributed neural systems that span cortical and subcortical regions. More broadly, the paper considers how we should think of causation in complex systems like the brain, so as to inform the relationship between emotion and other mental aspects, such as cognition.

Cognitive-motivational interactions: beyond boxes-and-arrows models of the mind-brain

How do motivation and cognitive control interact in brain and behavior? The past decade has witnessed a steady growth in studies investigating both the behavioral and the brain basis of these interactions. In this paper, I describe such interactions in the context of the dual completion model, which proposes that motivational significance influences both perceptual and executive competition. Embracing a research agenda that attempts to understand cognition-motivation interactions highlights considerable challenges faced by investigators. For example, even the standard language utilized, with terms such as "perception," "attention," "cognition," and "motivation," encourages a modular-like conceptualization of the underlying processes and mechanisms. I propose that large-scale interactions involving both task-related and valuation-related networks help understand how motivation shapes executive function. I argue that, ultimately, the mind and brain sciences need to move beyond "boxes and arrows" and fully embrace the richness and complexity of the interactions between motivation and cognition. In the last 10 years, the study in humans of the interactions of motivation with perception and cognition has grown at a fast pace. The growth has included behavioral studies characterizing the processes involved, and neuroimaging studies investigating the regions and circuits underlying the behaviors in question. This literature acknowledges the fact that perception and cognition do not happen in a vacuum but are, instead, situated in contexts that feature value. Although this assertion is uncontroversial, the mind and brain sciences have studied perception and cognition for many decades by largely extricating value from them. Fortunately, this state of affairs has now changed and the field has a newfound vigor in attempting to understand the impact of motivation on these mental functions.

Do Intelligent Robots Need Emotion?

What is the place of emotion in intelligent robots? Researchers have advocated the inclusion of some emotion-related components in the information-processing architecture of autonomous agents. It is argued here that emotion needs to be merged with all aspects of the architecture: cognitive-emotional integration should be a key design principle.

A Network Model of the Emotional Brain

Emotion is often understood in terms of a circumscribed set of cortical and subcortical brain regions. I propose, instead, that emotion should be understood in terms of large-scale network interactions spanning the entire neuroaxis. I describe multiple anatomical and functional principles of brain organization that lead to the concept of 'functionally integrated systems', cortical-subcortical systems that anchor the organization of emotion in the brain. The proposal is illustrated by describing the cortex-amygdala integrated system and how it intersects with systems involving the ventral striatum/accumbens, septum, hippocampus, hypothalamus, and brainstem. The important role of the thalamus is also highlighted. Overall, the model clarifies why the impact of emotion is wide-ranging, and how emotion is interlocked with perception, cognition, motivation, and action.

Dynamic Networks in the Emotional Brain

Research on the emotional brain has often focused on a few structures thought to be central to this type of processing-hypothalamus, amygdala, insula, and so on. Conceptual thinking about emotion has viewed this mental faculty as linked to broader brain circuits, too, including early ideas by Papez and others. In this article, we discuss research that embraces a distributed view of emotion circuits and efforts to unravel the impact on emotional manipulations on the processing of several large-scale brain networks that are chiefly important for mental operations traditionally labeled with terms such as "perception," "action," and "cognition." Furthermore, we describe networks as dynamic processes and how emotion-laden stimuli strongly affect network structure. As networks are not static entities, their organization unfolds temporally, such that specific brain regions affiliate with them in a time-varying fashion. Thus, at a specific moment, brain regions participate more strongly in some networks than others. In this dynamic view of brain function, emotion has broad, distributed effects on processing in a manner that transcends traditional boundaries and inflexible labels, such as "emotion" and "cognition." What matters is the coordinated action that supports behaviors.