Sonia J. Bishop - US grants

DESCRIPTION (provided by applicant): Two key NIMH objectives pertain to (a) the discovery of 'markers'that can be used to identify individuals at risk of psychiatric disease and (b) the development of preventative treatment measures to pre-empt the onset of disorder in these individuals. The proposed research seeks to further these objectives in the context of anxiety disorders. It investigates the neurocognitive mechanisms through which trait vulnerability to anxiety is conferred and examines whether cognitive and neurocognitive training can be used to restore function in areas of impairment. In addition, it explores whether neural markers can be used to identify those individuals 'at- risk'of anxiety disorders who are most likely to benefit from preventative cognitive training interventions. Functional magnetic resonance imaging (fMRI) will be used to test a dual route model of trait vulnerability to anxiety, the contention being that at least two, largely independent, sources of individual variability confer vulnerability to anxiety disorders. Our first specific aim is to establish whether (i) impoverished recruitment of frontal control mechanisms and (ii) enhanced amygdala responsivity to cues that signal threat are independently associated with trait vulnerability to anxiety. Both attention and fear conditioning fMRI experiments will be conducted in healthy volunteers with varying levels of trait anxiety. We aim to establish whether trait anxiety-related dysregulation of frontal function encompasses both dorsolateral and ventromedial prefrontal regions, with disruption to the former impacting the regulation of attention, and disruption to the latter impacting the down-regulation and extinction of conditioned fear. Our second specific aim is to establish whether trait anxious individuals primarily characterised by frontal hypo-activity will benefit from attention and emotion regulation training to a greater extent than those primarily characterised by amygdala hyper-responsivity. Here, we will also investigate whether the benefits of attention and emotion regulation training can be augmented by provision of neurofeedback from the frontal regions supporting these processes. We will examine whether these training procedures can be used to restore impoverished dorsolateral and ventromedial prefrontal function, and we will investigate the impact of this upon sustained attention and the down-regulation and extinction of conditioned fear. PUBLIC HEALTH RELEVANCE: For nearly 20% of the population, having an Anxiety Disorder causes serious disruption to everyday life. The proposed research uses brain imaging techniques to divide individuals at high risk of Anxiety Disorders into those who show low activity in frontal regions of the brain used to control attention and emotional responses versus those who show heightened activity in the amygdala, a region involved in detecting signs of danger. This will enable us to test the prediction that the former group will benefit to a greater extent from training procedures aimed at improving attention and the regulation of negative emotions.

Multi-feature modeling of the neural representation of emotion- and identity-related information derived from facial and vocal cues. People signal their internal emotional state by a range of cues including facial expressions, non-verbal vocalizations and the tone and content of speech. Much work to date on emotional signaling has focused on a small number of emotions and has relied on stimulus sets with limited numbers of exemplars and poor representation of individuals of different ages and ethnicities. The computer vision literature has long recognized the need to be able to compare models, for example of face recognition, across complex, diverse, and naturalistic stimulus sets. Here, we argue that a parallel approach needs to be applied if we are to achieve a robust and generalizable understanding of how the human brain represents emotion and identity related information derived from facial and vocal cues. Three multi-session functional magnetic resonance imaging (fMRI) experiments are proposed. In the first, participants will view a large corpus (~2000) of faces of individuals varying in age, gender and ethnicity and showing a wide range of emotional expressions. In the second, participants will be presented with a large (~1000) and equally diverse set of emotional vocalizations. The third experiment will make use of an even more complex and naturalistic stimulus set comprising ~1000 video clips of individuals expressing emotions through facial expression, non-verbal vocalizations and emotion-laden speech. This set will be broken into three parts, with as closely matched content as possible. These will be presented to participants in audio only, visual only and audio-visual (bimodal) conditions, with the stimuli allocated to each condition balanced across participants. For each experiment, data collection will comprise separate Model Estimation and Model Validation periods with the majority of stimuli presented at Validation being distinct from those presented at Estimation. A range of models will be fit to the fMRI data acquired during Estimation runs and tested and compared using data acquired during Validation runs. These models contain sets of features that describe the emotional content of the stimuli presented in terms of either dimensional or categorical models of emotion. By comparing the fit of these models we can determine which models capture most variance in voxel response profiles and examine how this varies across brain regions. Across the three experiments, we also seek to establish whether there are regions where voxels show common coding of emotional state regardless of whether information is carried by facial or vocal cues or a combination of both. Further, by contrasting models with and without terms for characteristics such as age, gender and ethnicity we can also investigate the (in)dependence of the representation of emotion and identity-related features. The proposed research has the potential to greatly advance our understanding of how cues to others' emotional state are represented in the human brain and the extent to which this is influenced by the characteristics of the person we are interacting with. In the medium term, we plan to extend this to also model listener/ viewer characteristics (both demographics and predisposition to anxiety or depression) in a hope to advance our understanding of how biases in the interpretation of emotional signals can arise.

PROJECT SUMMARY In our daily lives, we choose between different courses of action with the hope of achieving desired positive outcomes and avoiding feared negative outcomes; this is complicated by various forms of uncertainty that impact the probability that any given action will result in a particular outcome. Within NIMH?s RDoC framework, studies of the mechanisms involved in reward-based action valuation and choice have informed constructs listed under the Positive Valence Systems (PVS) domain. Associated paradigms examine how differences in outcome probability (first order uncertainty) and action-outcome contingency uncertainty (second order uncertainty) impact choice between alternate options. Within the Negative Valence Systems (NVS) domain, the construct of potential threat (anxiety) does not include consideration of the impact of potential threat, its probability and action- outcome contingency uncertainty, upon action valuation or choice; in addition paradigms listed under the NVS domain have no choice (instrumental) element and use physiological indices as dependent measures. These differences between constructs and tasks across the PVS and NVS domains hinder attempts to elucidate whether psychopathology-related deficits in probabilistic decision-making and the factors influencing action valuation and choice are common across both domains or unique to one or the other. Here, we will address this by creating equivalent PVs (reward) and NVS (shock) versions of two probabilistic decision-making tasks. We will use a hierarchical Bayesian computational framework to model behavioral and brain (functional magnetic resonance imaging) data from PVS and NVS versions of each task. This data will be acquired from healthy adult humans with a range of anxiety and depressive symptomatology. In addition to group-level analyses, we will use bifactor analysis to examine the latent factors underlying variance in anxiety and depressive symptomatology across participants and will relate scores on these factors to parameter estimates obtained by modeling of behavioral and brain data. Using this approach, we will examine commonalities and differences in the mechanisms supporting probabilistic decision-making when potential outcomes are aversive versus rewarding and alterations to these mechanisms as a function of anxiety and depressive related symptomatology. We hope that this will advance our understanding of the aspects of decision-making disrupted in anxiety and depression and the potential consequences for daily life. An additional goal of this research is to provide tasks and models that can be used in future clinical studies of probabilistic decision-making across both PVS and NVS domains.

Project Abstract Computational modeling can help us formalize how choice behaviors can be optimally adapted to different situations and investigate the ways in which individuals deviate from optimal behavior. Both anxious and depressed individuals report difficulties with decision-making; these difficulties have consequences for social interactions and occupational function. Understanding whether anxiety and depression are associated with common or unique deficits in decision-making has been hampered by studies focusing on either anxiety or depression alone and overlooking issues of comorbidity. This is important to address to better identify which aspects of decision-making should be targets for intervention in different patient groups. The separate investigation of anxiety- and depression-related deficits in decision-making has also led to a lack of equivalence of tasks and limited use of both reward-related and aversive outcomes within the same study. In the proposed research, we will conduct bifactor analysis of item-level responses to anxiety and depression questionnaires and use participant scores on the dimensions obtained to interrogate whether deficits in decision- making under second-order uncertainty are common to both anxiety and depression or unique to one or the other. We focus upon second-order uncertainty as this characterizes many of the situations we encounter in every-day life but there has been limited investigation of whether anxiety or depression are linked to deficits in adjusting decision-making to second-order uncertainty. Second-order uncertainty arises both when the probability of our actions resulting in certain outcomes changes across time (volatility) and when information needed to estimate how likely a given action is to lead to a given outcome is not fully available (ambiguity). In the proposed studies, we will use volatility and ambiguity manipulations to examine whether deficits in decision-making under second-order uncertainty are common to both anxiety and depression or unique to one or other and whether such deficits are domain general or domain specific (vary by outcome type: aversive, reward gain or reward loss). On-line studies will be used to conduct replication work and to examine if impaired decision-making under second-order uncertainty is primarily linked to internalizing symptomatology or common to a broader range of psychopathology. These online studies will also enable us to test exploratory hypotheses pertaining to other dimensions of psychopathology. Understanding the extent to which alterations in decision- making under second order uncertainty are unique to anxiety or depression, common to both anxiety and depression (i.e. a transdiagnostic marker of Internalizing psychopathology), or associated with psychopathology more broadly is important to clarify so that we can better tailor cognitive and psycho- educational interventions to different patient groups. It may also help clarify whether existing interventions developed in relation to anxiety (e.g. CBT focusing on ambiguity aversion) might valuably be applied to other forms of psychopathology.