CCNP Meetings

Eran Eldar
Department of Cognitive and Brain Sciences
The Hebrew University of Jerusalem

Emotions ubiquitously impact action, learning, and perception, yet their essence and role remain widely debated. The recent emergence of computational cognitive accounts of emotion has the potential to offer greater conceptual precision informed by normative principles and neurobiological data. However, these nascent efforts have so far been mutually inconsistent and limited in scope. In this talk, I will offer an integrative computational account of the human emotional landscape as composed of different emotions, each promoting adaptive behavior by mediating a distinct type of inference concerning oneself and the environment. This account builds on three parsimonious assumptions. First, that the inference each emotion mediates can be deduced from the circumstances that evoke it and the behavior it promotes. Second, emotions that primarily differ in valence reflect similar inferences about positive versus negative experiences. Third, emotional states that primarily differ in scope (timespan and object-specificity) reflect similar inferences about the immediate context versus the general environment. We apply these principles to integrate a large body of empirical research on the causes and consequences of different emotions with the computational and cognitive neuroscience of learning and decision making. The results bring to light an emotional ecosystem composed of multiple interacting elements which together serve to evaluate outcomes (pleasure & pain), learn expected values (happiness & sadness), adjust behavior (content & anger), and plan in order to realize (desire & hope) or avoid (fear & anxiety) uncertain future prospects.

View a recording of this session here.

Joo-Hyun Song
Department of Cognitive, Linguistic & Psychological Sciences
Brown University

Our daily experience can be thought of as a sequence of acquiring perceptual input to make decisions, then planning and executing appropriate actions. Hence, examining the influence of perception on action flows logically. Investigating the inverse may seem unusual. However, in a series of studies, we have accumulated evidence supporting co-dependence between action and perception. First, we demonstrated that simultaneous easy-action preparation or even prior action training can enhance sensitivity to an action-relevant low-level visual property, such as orientation. This newly-observed modulation of visual perception by action fluency cannot be explained by the traditional sequence of information processing stages. In addition, we discovered that improvement of motor timing enhances the sensitivity of time perception, even for implicit timing patterns inherent to a complex motor task. We interpret this as evidence for a shared temporal mechanism between perception and movement, regardless of the rhythmicity or complexity of the motor tasks. Furthermore, we found that learning a visuomotor rotation, but not actions without a rotation component, facilitated response time on a subsequent mental rotation task. This result suggests that visuomotor learning can enhance mental processes through common components. Taken together, our work supports a close interplay between the action system and perception, which highlights the necessity of an integrated approach to understand our adaptive behavior in a complex environment. The integrated approach would allow us to investigate a range of broader questions that would have not been possible by studying the motor system alone or vision alone.

View a recording of this session here.

David Zald
Department of Psychiatry
Robert Wood Johnson Medical School

This is a continuation of our Spring 2022 workshops on using computational models. In this third part of the workshop, we will discuss the integration of computational models into functional MRI studies. The workshop will cover some basics of experimental design and analysis of fMRI data, as well as specific considerations when using fMRI in the context of computational neuropsychiatry.

View a recording of this session here.

Avram Holmes
Department of Neuroscience
Yale University

Research in my laboratory focuses on discovering the fundamental organization of large-scale human brain networks. A core motivation that drives this work is the search for specific network-level signatures or “fingerprints” that co-vary with heritable behavioral variation in the general population and mark vulnerability for psychiatric illness onset. To date, research on the biological origins of psychopathology has largely focused on discrete illness categories. Although patient groups within this diagnostic system are treated as distinct entities, there are often murky boundaries between health and disease and across the disorders themselves. To establish the etiology of these complex syndromes, we must account for diagnostic heterogeneity, both relatively selective and disorder-spanning symptoms, and the dimensional nature of genetic risk.

In this talk, I will present two converging lines of research from my laboratory that aim to identify neurobiological markers of psychiatric illness. First, I will discuss a recent effort to link individual variability across the collective set of functional brain connections with the nature and severity of symptom profiles across unipolar depression, bipolar depression, and schizophrenia. Second, I will present ongoing research that uses measures of post-mortem gene expression to examine the relative influence of inhibitory interneuron subtypes on brain function, cortical specialization, and human behavior. In doing so, I will highlight how this information can be leveraged to understand individual variability in the diverse processing capabilities of the human brain and associated vulnerability for psychiatric illness onset.

View a recording of this session here.

Matt Nassar
Department of Neuroscience
Brown University

People flexibly adjust their use of information according to context. The same piece of information, for example the unexpected outcome of an action, might be highly influential on future behavior in one situation — but utterly ignored in another one. Bayesian models have provided insight into why people display this sort of behavior, and even identified potential neural mechanisms that link to behavior in specific tasks and environments, but to date have fallen short of providing broader mechanistic insights that generalize across tasks or statistical environments. Here I’ll examine the possibility that such broader insights might be gained through careful consideration of task structure. I’ll show that we can think about a large number of sequential tasks as requiring the same inference problem — that is to infer the latent states of the world and the parameters of those latent states — with the primary distinctions within the class defined by transition structure. Then I’ll talk about how a neural network that updates latent states according to a known transition structure and learns “parameters” of the world for each latent state can explain adaptive learning behavior across environments and provide the first insights into neural correlates of adaptive learning across environments. This model generates internal signals that identify the need for latent state updating, which maps onto previous observations made in pupil dilations and P300 responses across different task environments. I will also discuss an experiment that we are currently setting up to test the idea that these signals might reflect a latent state update signal, with a focus on relationships to learning and perception. Finally, I will briefly mention some theoretical work examining how latent states could be used to shape noise correlations in neural populations in order to speed learning.

View a recording of this session here.

Links to relevant work:
https://pubmed.ncbi.nlm.nih.gov/34144114/
https://pubmed.ncbi.nlm.nih.gov/35105677/
https://pubmed.ncbi.nlm.nih.gov/34193556/

Woo-Young Ahn
Department of Psychology
Seoul National University

Machine learning has the potential to facilitate the development of computational methods that improve the measurement of cognitive and mental functioning, and adaptive design optimization (ADO) is a promising machine-learning method that might lead to rapid, precise, and reliable markers of individual differences. In this talk, I will first discuss the importance of reliability of (bio)markers. Then, I will present a series of studies that utilized ADO in the area of decision-making and for the development of ADO-based digital phenotypes for addiction and related behaviors. Lastly, I will discuss other promising approaches that might allow us to develop (bio)markers with clinical utility.

View a recording of this session here.

Yael Niv
Princeton Neuroscience Institute
Princeton University

This is a continuation of the model fitting workshop from February 23rd. In this second part of the workshop, we will discuss how to compare models and use data to determine which of several alternatives is best.

View slides from this session here.

View a recording of this session here.

Jessica Hamilton
Department of Psychology
Rutgers University

Rates of depression and suicide among adolescents have increased in the past 10 years, and suicide is now the second leading cause of death among individuals aged 10-24. To address this major public health problem, Dr. Jessica Hamilton’s program of research focuses on identifying and modifying developmentally-informed risk and protective factors for youth suicide, particularly aimed at reducing disparities in suicide. Specifically, Dr. Hamilton will describe her research examining sleep and social media as both potential risk factors and opportunities for prevention, and how her research harnesses advancing technology to better understand and prevent adolescent suicide risk.

View a recording of this session here.

Travis Baker
Center for Molecular and Behavioral Neuroscience
Rutgers University

In this talk, I will present two recent studies using TMS to manipulate electrophysiological and computational correlates of decision-making, and augmented reality (AR) to manipulate real-world environments in real-time during goal-directed navigation. Because TMS offers a powerful tool for investigating causal brain-behavior relations, and AR can alter one’s ongoing perception of the real-world, such experimental applications, when paired with computational modeling, may help reconcile or dispute theories and models of decision-making, help increase the ecological validity of human decision-making studies, and advance our understanding and treatment of mental disorders, with a special focus on addictions.

View a recording of this session here.

Candace Raio
Department of Psychiatry
NYU Grossman School of Medicine
New York University

The ability to effectively deploy goal-directed control during learning and decision-making is essential for adaptive behavior but is often compromised under stress. Using experimental paradigms that draw upon learning models and behavioral economics, I will present research that examines how stress changes the use of goal-directed control strategies and suggest that these changes may stem from the increased cognitive cost of exercising control. Further, I will show that the subjective cost of control can be measured using a novel economic decision-making approach and that these costs are highly sensitive to changes in affective state. Finally, I will argue that stress can also confer adaptive benefits for survival and demonstrate how this may emerge in decision contexts marked by uncertainty.

View a recording of this session here.