learning

Structural and functional characterization of autoimmune antibodies against NMDAR

Project Summary. The goal of this project is to understand the origins and molecular mechanisms underlying the anti-cancer autoimmune response against the N-methyl-D-aspartate receptor (NMDAR) and its correlation with anti-N-methyl-D-aspartate receptor autoimmune encephalitis (NMDARAE). While anti-cancer immune responses can promote tumor elimination, they may also lead to the production of self-reactive antibodies that trigger autoimmune diseases. NMDARAE is the most common form of immune-mediated encephalitis, which results in prominent neuropsychiatric symptoms, including seizures, psychosis, and memory deficits. NMDARs belong to a family of ligand-gated ion channels expressed exclusively in the central nervous system. They are involved in various aspects of brain development and function, including learning and memory. They respond to the neurotransmitter glutamate and a co-agonist, glycine or D-serine, to mediate excitatory neurotransmission, which plays a central role in synaptic plasticity. NMDARAE is associated with ovarian teratomas, where aberrant NMDAR expression is believed to trigger an autoimmune response. In NMDARAE, anti-NMDAR antibodies, as well as B cells and antibody-secreting cells, cross the blood-brain barrier via unknown mechanisms, resulting in the presence of anti-NMDAR antibodies at high titers within the brain and cerebrospinal fluid (CSF). These antibodies target NMDARs, modulating their function and contributing to disease pathology. Emerging evidence, supported by our preliminary data, suggests that NMDARs are also expressed in triple-negative breast cancer (TNBC), extending the relevance of anti-NMDAR autoimmunity beyond ovarian teratomas. In our TNBC mouse model, which ectopically expresses NMDARs (TNBC-NMDAR), we observed the onset of anti-NMDAR autoimmunity, where the produced antibodies cause both anti-tumor activity and symptoms such as lowered seizure threshold, mirroring key features of NMDARAE. Here, we will establish this TNBC mouse model as we develop molecular methods to characterize it. Aim 1 will focus on establishing and characterizing the TNBC- NMDAR mouse model. We will develop a detection method utilizing the intact tetrameric NMDAR channel proteins and a method to isolate B cells expressing B cell receptors against NMDAR from biological samples by using fluorescently labeled intact NMDAR proteins, followed by single-cell RNA sequencing. Aim 2 will utilize single-particle cryo-electron microscopy (cryo-EM) to investigate the interactions between NMDAR and the cloned antibodies, providing insights into epitope recognition, NMDAR subtype specificity, and conformational changes induced by antibody binding. Aim 3 will assess the impact of the cloned antibodies on NMDAR channel activity using electrophysiology. We will also assess anti-tumor activity and NMDARAE onset by each antibody clone. Together, the proposed research will gain insights into the link between anti-cancer anti-NMDAR autoimmunity and NMDARAE. It will also elucidate which functional properties of the cloned antibodies promote anti-tumor activity while contributing to NMDARAE, thereby informing potential therapeutic strategies.

Multimodal computational models for early prediction of peritoneal recurrence in gastric cancer

ABSTRACT Gastric cancer represents a significant disease burden and is a leading cause of cancer-related deaths in the United States and globally. Approximately 80% of gastric cancer patients are diagnosed at an advanced stage, with the peritoneum being the most common site of relapse (peritoneal recurrence) after radical surgery. Nearly 50% of patients with advanced-stage gastric cancer develop peritoneal recurrence post-surgery, resulting in a median survival of only 3–6 months and a markedly reduced quality of life. Early peritoneal recurrence is primarily characterized by micro-metastasis, which traditional imaging techniques struggle to detect due to the small size of metastatic nodules. Predicting the likelihood and timing of peritoneal recurrence is crucial for identifying at- risk patients, enabling timely interventions that could improve survival rates and quality of life. Unfortunately, reliable predictive biomarkers and models for peritoneal recurrence in gastric cancer are lacking in clinical practice, highlighting an urgent need for innovative predictive tools. This proposal aims to develop and validate novel predictive models for early peritoneal recurrence in gastric cancer, leveraging advanced deep learning techniques and multimodal integration of clinical, radiological (CT), and histopathological (hematoxylin and eosin, H&E) data. In Aim 1, we will develop a rational approach for predicting peritoneal recurrence by creating a novel deep learning multimodal method guided by genomics knowledge. Additionally, we will integrate both deep learning-extracted features and traditional hand-crafted radiomics features with clinical data to improve prediction accuracy. Aim 2 focuses on developing a robust prediction model of peritoneal recurrence utilizing a pre-trained foundation model from large-scale H&E image data. Aim 3 will combine CT, H&E, and clinical data to further enhance predictive capabilities, employing an innovative cross-modal collaborative optimization approach for multimodal data integration. All models will be trained and internally validated using a retrospective cohort from Atrium Health Wake Forest Baptist Comprehensive Cancer Center and externally validated in two independent cohorts from additional institutions to ensure robustness across populations and imaging protocols. Additionally, we will compare our models with existing methods, including clinical staging and alternative fusion strategies. If successful, these models will enhance risk stratification and prediction of peritoneal recurrence in gastric cancer patients, significantly improving survival rates and quality of life by identifying those likely to develop peritoneal recurrence post-surgery and facilitating timely intervention. Furthermore, they can help avoid the risk of complications and extra medical costs associated with overtreatment. Since the information is derived from routinely examined CT, H&E and clinical data, they could be seamlessly integrated into current clinical workflows. The AI technology developed through this project has the potential to benefit underserved populations in low- resource settings and reduce healthcare disparities in the U.S.

Mechanisms of antigen-specific T cell activation in MOGAD

PROJECT SUMMARY / ABSTRACT The overarching goal of this application is to train Dr. Carson E. Moseley, MD, PhD, who is a clinical neurologist and a research immunologist, to become an independent investigator studying and treating neuroimmunologic disorders. Myelin oligodendrocyte glycoprotein (MOG) antibody-associated disease (MOGAD) is a recently described, severe, neuroinflammatory syndrome of the central nervous system (CNS) with no approved therapies. Although MOG-specific antibodies helped define the disease, MOG antibodies alone are not clearly pathogenic and our understanding of MOGAD immunopathology is limited. CD4+ T cells are a dominant lymphocyte population in MOGAD lesions, yet the targets of T cell responses to MOG and how T and B cells interact to drive pathogenic immune response in MOGAD are unknown. This proposal uses a complementary approach of human and mouse immunology along with new technologies in T cell repertoire mapping and genome editing to dissect MOG-specific CD4+ T cell responses in MOGAD. Additionally, it will use new models to investigate how B cells promote pathogenic T cell differentiation and select pathogenic T cell receptors. The proposed training plan involves mentored training, seminars, formal learning, and advising to ensure completion of the proposed research and Dr. Moseley’s career development. He will train at UCSF, which is an outstanding institute for research and environment for physician-scientists. He will receive training in human immunology and CRISPR-based gene editing technologies. He will be mentored by Dr. Scott Zamvil, a leader in identifying antigen-specific T cell responses in neuroimmunologic disorders, and co-mentored by Dr. Alexander Marson, an expert in CRISPR gene editing to understand lymphocyte function. This application will provide Dr. Moseley with the long-term skills needed to become an independent investigator leading efforts to study and treat neuroimmunologic disorders.

Factors Driving Wear and Implant Failure in Total Shoulder Arthroplasty

Polyethylene (PE) wear and implant-related failure remain leading causes of revision in total shoulder arthroplasty (TSA), a procedure which now surpasses the growth rate of hip and knee arthroplasty. Both anatomic (aTSA) and reverse (rTSA) TSA outcomes are heavily influenced by complex interactions between rotator cuff function, scapular motion, implant design, and patient-specific loading—factors not adequately captured in current preclinical implant testing standards. Emerging evidence suggests that PE wear progression in TSA is highly dependent on shoulder kinematics, joint loading, implant positioning, and individual patient factors. Nonetheless, data on in vivo motion and load profiles remain sparse, and few tools exist to link these profiles to clinically relevant wear patterns or associated periprosthetic inflammatory tissue responses. Accordingly, the primary objective of this project is to develop validated, patient-specific models that predict PE wear in TSA and identify modifiable surgical, design, and rehabilitation targets to improve implant longevity and restore patient mobility. Additionally, we will establish histopathological hallmarks that indicate TSA failure caused by PE wear debris. Our central hypothesis is that specific shoulder kinematics and joint loading drive distinct PE wear patterns in TSA associated with mechanical failure or inflammatory-mediated osteolysis, depending on implant design and positioning. To achieve the overall objective of this work, shoulder motions and muscle excitations across 25 activities of daily living will be collected at pre-op and post-op (>6 months) in both aTSA and rTSA patients, with long-term follow-up of patient-reported outcomes via validated surveys (5 years). Unsupervised machine learning will categorize patients into movement-based phenotypes, which will then inform a multi-scale modeling framework to estimate in vivo shoulder joint loads and implant wear across the varying movement strategies. Predicted wear patterns will be validated using state-of-the-art preclinical wear simulators. Simultaneously, we will quantify how patient, surgical, and implant factors contribute to wear in retrieved TSA components (>400 samples), correlating imaging-based wear patterns with clinical outcomes, patient-reported function, inflammatory tissue responses, and radiographic indications of loosening. For that purpose, we will establish benchmarks of TSA wear rates and introduce a new histopathological approach augmented by infrared spectroscopic imaging. This work is innovative because we are linking patient-specific movement patterns following TSA with multi-scale computational models to predict PE wear, breaking the current approaches of using generic motions and loads in existing testing standards. This work will produce the first integrated, publicly available database of TSA kinematics, joint loading, and PE wear patterns and rates, along with validated computational tools to inform implant design, surgical planning, rehabilitation strategies, and personalized risk assessment. Ultimately, these advances will improve functional outcomes and long-term success for TSA patients and enable better preclinical testing methods and standards.

Neural circuits for disinhibition in the cerebellum

ABSTRACT Our long-term goal is to understand how the cerebellum adapts and improves movements in response to motor errors. A critical component of this process is signaling from olivary climbing fibers that, by providing strong excitatory drive onto Purkinje cells, induces long-term synaptic plasticity to instantiate corrective adjustments in motor behavior. However, this signaling process is tightly regulated by molecular layer interneurons (MLIs). By strongly inhibiting Purkinje cells, MLIs oppose climbing fiber-driven excitation and gate the induction of corrective plasticity. Thus, for error-driven climbing fiber-induced plasticity and learning to occur effectively, Purkinje cells must undergo disinhibition through the suppression of MLI-mediated input. Notably, MLI ensembles are composed of several subtypes and have a highly structured interconnectivity and are responsive to convergent climbing fiber inputs, suggesting that climbing fiber synchrony- whose functional significance is poorly understood- can selectively engage MLI networks to alter the state of Purkinje cell inhibition. This engagement may balance inhibition and excitation of Purkinje cells during motor errors, creating a circuit mechanism conducive for the acquisition of adaptive learning. The objective of this proposal is to determine how distinct MLI circuits are organized to modulate Purkinje cell excitability through disinhibition in a context-dependent manner, enabling plasticity and learning in response to motor errors. We will employ functional recordings, circuit-targeted activity manipulations, and behavioral analysis to reveal how error-driven instructive signaling emerges from these circuits. In the first aim, we will use in vivo high-density electrophysiology to map functional interactions among MLIs, climbing fibers, and Purkinje cells in the flocculus during the vestibulo-ocular reflex. We will test whether, during motor errors, climbing fibers synchronize their firing to selectively engage disinhibition of Purkinje cells through MLI subtypes in adapting versus non-adapting contexts. In the second aim, we will combine acute slice recordings and molecular anatomy to define direct versus spillover climbing fiber synapses onto MLI subtypes. We will identify synaptic markers and measure climbing-fiber-evoked currents in MLI subtypes, revealing how structural connectivity supports rapid, subtype-specific circuit engagement. In the third aim, we will determine how long-range inputs to the inferior olive, specifically inhibitory projections from the vestibular nuclei, dynamically tune climbing fiber synchrony in vivo and thereby learning through differential engagement of disinhibitory MLI networks. Using functional recording and optogenetic manipulation during the vestibulo- ocular reflex performance, we will establish causal links between climbing fiber synchrony, MLI network state, and adaptive behavior. By fully understanding the logic of instructive signaling, emergent from cerebellar circuit organization and behavioral engagement, we will advance our knowledge of cerebellum-dependent learning processes and provide broader insights into the neural mechanisms of learning and adaptation more generally.

AI-enabled methods for de novo design of functional peptides

PROJECT SUMMARY Macrocyclic peptides offer unique therapeutic potential, particularly for targeting intracellular protein-protein interactions considered ‘undruggable’ with traditional therapeutic modalities. Additionally, peptides can combine the benefits and bridge the gap between conventional small molecule therapeutics and large biologics. However, developing new peptide-based therapeutics using traditional approaches, such as natural product discovery or high-throughput library screening, has remained slow and challenging. Moreover, these conventional approaches cover a small fraction of the chemical and structural space, are restricted to a few starting peptide scaffolds, and typically fail to optimize for multiple therapeutic properties simultaneously. Our central hypothesis is that structure-guided deep learning methods can rapidly explore the chemical and structural space beyond natural products and enable precise, rapid, and custom design of functional peptides simultaneously optimized for target binding, selectivity, and membrane permeability. In our recent work, we developed physics-based methods for designing constrained peptides and macrocycles and, more recently, introduced deep learning methods for structure prediction, sequence redesign, and de novo design of peptide monomers and targeted binders. Here, we propose to develop a new generation of structure-guided deep learning (DL) tools to address the current limitations of computational and experimental methods and enable accurate, accessible, and broadly applicable design of macrocycles. Specifically, we will pursue the projects focused on: (i) leveraging DL methods to systematically enumerate the chemical and structural space of constrained peptides and membrane-traversing peptides to develop scaffolds and core design principles for functional peptide design; (ii) high-throughput design and data collection to improve design selection, filtering metrics, and sequence design algorithms; (iii) developing generative DL methods that expand beyond current capabilities and allow sequence and structure design with vast chemical space of non-canonical amino acids; and (iv) use those new generative methods to design macrocyclic binders against different therapeutically-relevant targets, including the critical fusion and attachment proteins from viruses of pandemic concern. Our preliminary work in these proposed areas demonstrates the feasibility of this approach. The proposed computational tools, scaffold sets, and designed peptides will significantly advance therapeutic design beyond the state-of-the-art and enable rapid and custom design of drug- like peptides tailored for addressing complex therapeutic, diagnostic and research challenges.

Examining the foundations of reading comprehension: a longitudinal study of brain and behavior starting in infancy

SUMMARY Reading comprehension (RC) is one of the most complex skills that we utilize daily and is crucial for functioning in modern society, but despite its significance for academic achievement, employment prospects, and mental health, many children and adults do not exhibit proficient RC abilities. New theoretical models aiming to explain variability in RC suggest a dynamic interplay and co-development among ‘precursor’ foundational and cognitive- linguistic skills, interacting with environmental and socio-ecological factors across the developmental timeline of learning to read. Behavioral and neuroimaging studies in school-age children have demonstrated critical mechanistic support for these multifactorial RC models by identifying the developmental trajectories of precursor skills and further showing that brain areas, tracts, and networks typically underlying language and cognitive skills are also involved in RC. Nevertheless, the precursor skills that support RC start developing in infancy and the brain correlates underlying these precursors begin to develop in utero, which suggests that typical and atypical RC developmental trajectories could diverge long before school age. As such, examining RC development using a multifactorial, longitudinal approach that includes brain and behavior starting in infancy is critical for developing theoretical frameworks that can inform early preventative and intervention strategies. Here, we propose a comprehensive longitudinal study of RC development in which we examine direct and indirect effects on RC from brain, behavioral, familial risk, and environmental data from infancy to adolescence. To achieve this goal, we will combine two existing longitudinal cohorts, one ranging from infancy to late childhood (n = 174) and the other from preschool to early adolescence (n = 137). By applying state-of-the-art pediatric neuroimaging analyses, multiple indicator growth model structural equation models, and an innovative behavior- brain co-development measurement index to this unique, combined dataset, we will be able to identify brain and behavioral measures in infancy that directly and indirectly support subsequent RC development (Aim1). We will further characterize how longitudinal trajectories of behavioral measures as well as brain structure, function, and white matter organization contribute to RC development and how familial risk and environmental factors shape these trajectories (Aim 2). Finally, we will examine how the co-development of brain and behavior, as measured with an innovative co-development index, relates to subsequent RC (Aim 3). If successful, we will contribute the first multifactorial longitudinal model of RC development comprising direct and indirect effects from brain, behavior, brain-behavior co-development, familial risk, and environmental measures beginning in infancy. Understanding RC development using a multifactorial longitudinal lens will be crucial for building theoretical models and developing experimental designs focused on early preventative and intervention approaches long before the start of formal schooling.

Mechanisms and consequences of cerebrovascular dysfunction in preeclampsia

PROJECT SUMMARY/ABSTRACT Preeclampsia (PE) is a common hypertensive disorder of pregnancy that causes significant maternal and fetal morbidity and mortality worldwide. PE women are at a high risk of stroke, including intracerebral hemorrhage, during the peripartum period, suggesting the sequelae of PE adversely impacts the cerebral circulation to promote hemorrhage. In addition, women with severe early-onset PE are at an 85-fold increased risk of death from intracerebral hemorrhage, importantly suggesting severity of disease promotes greater vulnerability of the cerebral circulation to degradation and rupture. However, the consequences of PE extend far beyond pregnancy and are associated with excessive cardiovascular and cerebrovascular disease risk later in life. Women with previous pregnancy complicated by PE can develop cognitive impairment as early as in their 30’s and 40’s, suggesting PE predisposes the brain to early-onset cognitive impairment. Studies have shown that formerly PE women have changes in gray matter volume and increased white matter lesion burden that occurs as a function of time from pregnancy, suggesting that PE continues to progressively damage the brain long after the affected pregnancy. Thus, our overall goal is to elucidate mechanisms by which women with PE are at risk of intracerebral hemorrhage in pregnancy and cognitive decline later in life. Our preliminary studies found greater vascular degradation, hematoma and cerebral edema in a model of severe PE that was associated with vascular inflammation and microglia activation (neuroinflammation). In addition, we found endothelial dysfunction and diminished neurovascular coupling in PE rats that persisted 5 months postpartum. Impaired neurovascular coupling is well-recognized as an underlying contributor to cognitive decline. These effects in postpartum animals with previous exposure to PE were associated with memory impairment that was not present in the pregnant state, suggesting neurovascular dysfunction precedes cognitive decline. Our central hypothesis is that the sequela of PE accelerates hypertension-induced cerebrovascular dysfunction that predisposes to intracerebral hemorrhage during pregnancy and its persistence postpartum results in early-onset cognitive decline. We will therefore elucidate mechanisms by which PE accelerates vascular degradation and worsens outcome from hemorrhagic stroke, probing pathways involved in oxidative degradative processes using multi-omics and multivariate analysis (Aim 1). We will also determine underlying molecular mechanisms that cause persistent cerebral microvascular dysfunction and cognitive decline postpartum, including oxidative stress-induced BBB leakage and persistent neuroinflammation that drives potassium channel dysfunction, reduced neurovascular coupling and neurovascular uncoupling (Aim 2). We will also use machine learning approaches together with multi-omics and outcome measures to identify factors and cellular pathways that are most impactful for prediction of intracerebral hemorrhage and cognitive impairment. The ability to predict and prevent devasting neurovascular disorders associated with PE has the potential to have long-lasting impacts on the lives of women with PE.

Bridging Local and System-Wide Autoreactive, Extrafollicular B Cell Signatures in a TLR7-Driven Model

Project Summary A substantial body of literature has described the development of autoreactive humoral responses in the context of autoimmune disease and recently discerned an exciting new avenue for investigation. While early work focused on canonical mechanisms of activation through the germinal center (GC) response, recent studies have found GC infrastructure to be dispensable for the onset of chronic autoimmunity. It has become clear that an alternative pathway of B cell activation, the extrafollicular (EF) pathway, can drive the onset of new autoreactivity in multiple human disorders including rheumatoid arthritis and systemic lupus erythematosus (SLE). In comparison to the GC pathway, the EF pathway represents a less stringent method for B cell activation, leads to accelerated antibody-secreting cell (ASC) formation, and thus has a higher propensity for the production of autoreactive B cell effectors and ASCs. Recently, our group has identified a similar skew toward the EF response in the context of severe viral infection, tied to acute tolerance loss, increased disease severity, and complicated recovery from infection. These findings highlight how further study of the EF response is crucial to our understanding of autoimmune induction across multiple areas of disease. Toll-like receptor 7 (TLR7) stimulation has been identified as a key contributor to EF B cell development in SLE, and several studies have now linked TLR7 overstimulation to chronic autoimmune disease. While EF effector B cell populations have now been identified in both murine models and humans, substantial gaps in our knowledge remain to be answered concerning i) the origins of these cells and ii) the system-wide and microenvironmental signaling and organization that drive this differentiation pathway. We propose to address these gaps, here, by utilizing a TLR7 agonist (R848) in a murine model to characterize the autoreactive response within the blood and draining lymph node through innovative high-throughput analytical techniques. Systemic shifts in proteomic signatures and immune cell phenotype will be monitored in the blood throughout the induction of autoreactivity, using novel applications of machine-learning based classification. These signatures will then be connected to developing inflammatory microenvironments identified within the draining lymph node by applying a customized set of software tools to spatial transcriptomic data. This work will deepen our understanding of the immunologic mechanisms by which the EF pathway can lead to “run-away” autoreactive B cell development, with the added potential for identification of early blood-based biomarkers for this developing autoreactivity. The above proposed work will provide an ideal training opportunity for the candidate to develop experience with advanced immunologic laboratory techniques, rigorous bioinformatic analysis, a systems-level view of immunology, and scientific communication. The Woodruff and Sanz Labs are highly experienced within the autoimmune disease space with extensive experience with the required techniques and established routes for clinical collaboration to act on these findings.

Role of Two Medial Prefrontal Long-Range Recurrent Networks in Behavior Initiation and Inhibition

Abstract The medial prefrontal cortex (mPFC) is critical for executive function, yet how its dorsal (dmPFC) and ventral (vmPFC) motor-projecting (MP) neurons coordinate behavioral initiation, inhibition, and cognitive flexibility remains poorly understood. This R21 leverages four translational behavioral paradigms (head-fixed Persistent Licking/Shock-Escape; freely moving FED3-based Reversal Learning/Stop-Signal), high-density neural recordings, circuit manipulations, and Brian2 spiking neural network modeling to test our central hypothesis: dmPFC MP neurons drive action initiation and adaptive switching, while vmPFC MP neurons suppress impulsivity and perseveration. In Aim 1a, we quantify behavior using kinematic analyses (jerk, velocity, z-scored) aligned with human executive dysfunction metrics (Action Latency [AL], Reversal Accuracy [RA], Perseveration Errors [PE], Stop-Signal Reaction Time [SSRT]), combined with optogenetic (stGtACR2/ChR2) and chemogenetic (PSAM/varenicline) perturbations. Aim 1b employs optotagging and population analyses (PCA, SVM, Total Spiking Probability Edges) to decode dmPFC/vmPFC MP dynamics across tasks, resolving specialized versus mixed functional roles. Aim 1c integrates these datasets into Brian2 spiking network models to predict neural-behavioral correlations, validated through cross-validation. Exploratory analyses will link murine kinematic signatures to human stop-signal/reversal learning metrics. By elucidating strain-specific (C57BL/6 vs. CD1) circuit mechanisms and delivering translatable biomarkers (AL, RA, PE, SSRT, kinematics), this work addresses a critical gap in understanding neuropsychiatric disorders like ADHD (impulsivity) and schizophrenia (perseveration). The study’s innovative combination of recurrent neural network theory, FED3-based assays, and New Approach Methodology (NAM)-compliant computational modeling pioneers high-risk, high-reward tools for circuit dissection, fully aligning with NIH’s 2025 priorities.

Overcoming Treatment Resistance by Targeting Polyploid Breast Cancer Cells with AI assisted Single-Cell Analysis

Therapy resistance remains a formidable challenge in breast cancer treatment, with emerging evidence identifying polyploid giant cancer cells (PGCCs) as key drivers. These cells, arising through whole-genome doubling (WGD) events, exhibit enhanced resistance to therapies, contributing to disease relapse. PGCCs are characterized by enlarged cell and nuclear sizes, increased DNA content, and greater resilience compared to non-PGCCs. Their prevalence escalates with disease progression and therapeutic stress, underscoring their critical role in treatment resistance. As such, we hypothesize that inhibiting polyploid cancer cells can effectively reduce therapeutic resistance. Despite this, effective strategies targeting PGCCs are limited, hindered by the lack of high-throughput methods to assess PGCC viability and abundance. Traditional screening assays lack the sensitivity to detect the elimination of small populations of PGCCs, while current detection methods, such as visual inspection and flow cytometry, are not suited for high-throughput compound screening. Our preliminary work has established a high-throughput single-cell morphological analysis pipeline capable of quantifying PGCCs, and we successfully screened 2,726 compounds for their efficacy on PGCCs. Based on the preliminary success, we aim to further improve its robustness and accuracy under diverse staining and imaging conditions, ensuring consistent performance across multiple labs for widespread use in PGCC/WGD studies, with deep learning to accelerate the discovery of therapeutic strategies targeting PGCCs. In addition to empirical screening, our scRNA-Seq analysis of PGCCs has revealed altered gene expression, particularly in genes associated with FOXM1, a transcription factor critical in cell cycle regulation and linked to poor outcomes in various cancers. PGCCs also show altered ferroptosis regulators and elevated reactive oxygen species (ROS), indicating susceptibility to ferroptosis. Here, we propose two independent and complementary aims. Aim 1: We will develop and validate a robust deep learning–based single-cell morphological analysis pipeline for accurate PGCC/non-PGCC discrimination across variable staining, imaging, and lab settings. The model will be benchmarked on independent datasets from external labs and released as open-source, version-controlled software with full documentation to support reproducibility and broad adoption in PGCC/WGD research. Aim 2: Leveraging our screen of 2,726 FDA-approved compounds and mechanistic studies of FOXM1 and ferroptosis, we will prioritize and validate therapies that eradicate PGCCs and reduce treatment resistance. Using patient- derived cells, 3D spheroids, and syngeneic/xenograft models, we will rigorously assess top candidates as monotherapy and in combination with standard-of-care agents. Successful completion of this project will accelerate PGCC/WGD research, advance therapeutic strategies to overcome breast cancer resistance, and especially deliver benefits to patients with high PGCC burden. Given the prevalence of WGD across solid tumors and its induction by standard therapies, our approach holds broad clinical relevance and translational impact.

Decoding stress vulnerability

Although stress can be considered as an ongoing process that helps an organism to cope with present and future challenges, when it is too intense or uncontrollable, it can lead to adverse consequences for physical and mental health. Social stress specifically, is a highly prevalent traumatic experience, present in multiple contexts, such as war, bullying and interpersonal violence, and it has been linked with increased risk for major depression and anxiety disorders. Nevertheless, not all individuals exposed to strong stressful events develop psychopathology, with the mechanisms of resilience and vulnerability being still under investigation. During this talk, I will identify key gaps in our knowledge about stress vulnerability and I will present our recent data from our contextual fear learning protocol based on social defeat stress in mice.

Predictive Coding Light

Current machine learning systems consume vastly more energy than biological brains. Neuromorphic systems aim to overcome this difference by mimicking the brain’s information coding via discrete voltage spikes. However, it remains unclear how both artificial and natural networks of spiking neurons can learn energy-efficient information processing strategies. Here we propose Predictive Coding Light (PCL), a recurrent hierarchical spiking neural network for unsupervised representation learning. In contrast to previous predictive coding approaches, PCL does not transmit prediction errors to higher processing stages. Instead, it suppresses the most predictable spikes and transmits a compressed representation of the input. Using only biologically plausible spike-timing based learning rules, PCL reproduces a wealth of findings on information processing in visual cortex and permits strong performance in downstream classification tasks. Overall, PCL offers a new approach to predictive coding and its implementation in natural and artificial spiking neural networks

Computational Mechanisms of Predictive Processing in Brains and Machines

Predictive processing offers a unifying view of neural computation, proposing that brains continuously anticipate sensory input and update internal models based on prediction errors. In this talk, I will present converging evidence for the computational mechanisms underlying this framework across human neuroscience and deep neural networks. I will begin with recent work showing that large-scale distributed prediction-error encoding in the human brain directly predicts how sensory representations reorganize through predictive learning. I will then turn to PredNet, a popular predictive coding inspired deep network that has been widely used to model real-world biological vision systems. Using dynamic stimuli generated with our Spatiotemporal Style Transfer algorithm, we demonstrate that PredNet relies primarily on low-level spatiotemporal structure and remains insensitive to high-level content, revealing limits in its generalization capacity. Finally, I will discuss new recurrent vision models that integrate top-down feedback connections with intrinsic neural variability, uncovering a dual mechanism for robust sensory coding in which neural variability decorrelates unit responses, while top-down feedback stabilizes network dynamics. Together, these results outline how prediction error signaling and top-down feedback pathways shape adaptive sensory processing in biological and artificial systems.

AutoMIND: Deep inverse models for revealing neural circuit invariances

Memory Decoding Journal Club: Distinct synaptic plasticity rules operate across dendritic compartments in vivo during learning

Distinct synaptic plasticity rules operate across dendritic compartments in vivo during learning

Understanding reward-guided learning using large-scale datasets

Understanding the neural mechanisms of reward-guided learning is a long-standing goal of computational neuroscience. Recent methodological innovations enable us to collect ever larger neural and behavioral datasets. This presents opportunities to achieve greater understanding of learning in the brain at scale, as well as methodological challenges. In the first part of the talk, I will discuss our recent insights into the mechanisms by which zebra finch songbirds learn to sing. Dopamine has been long thought to guide reward-based trial-and-error learning by encoding reward prediction errors. However, it is unknown whether the learning of natural behaviours, such as developmental vocal learning, occurs through dopamine-based reinforcement. Longitudinal recordings of dopamine and bird songs reveal that dopamine activity is indeed consistent with encoding a reward prediction error during naturalistic learning. In the second part of the talk, I will talk about recent work we are doing at DeepMind to develop tools for automatically discovering interpretable models of behavior directly from animal choice data. Our method, dubbed CogFunSearch, uses LLMs within an evolutionary search process in order to "discover" novel models in the form of Python programs that excel at accurately predicting animal behavior during reward-guided learning. The discovered programs reveal novel patterns of learning and choice behavior that update our understanding of how the brain solves reinforcement learning problems.

“Brain theory, what is it or what should it be?”

n the neurosciences the need for some 'overarching' theory is sometimes expressed, but it is not always obvious what is meant by this. One can perhaps agree that in modern science observation and experimentation is normally complemented by 'theory', i.e. the development of theoretical concepts that help guiding and evaluating experiments and measurements. A deeper discussion of 'brain theory' will require the clarification of some further distictions, in particular: theory vs. model and brain research (and its theory) vs. neuroscience. Other questions are: Does a theory require mathematics? Or even differential equations? Today it is often taken for granted that the whole universe including everything in it, for example humans, animals, and plants, can be adequately treated by physics and therefore theoretical physics is the overarching theory. Even if this is the case, it has turned out that in some particular parts of physics (the historical example is thermodynamics) it may be useful to simplify the theory by introducing additional theoretical concepts that can in principle be 'reduced' to more complex descriptions on the 'microscopic' level of basic physical particals and forces. In this sense, brain theory may be regarded as part of theoretical neuroscience, which is inside biophysics and therefore inside physics, or theoretical physics. Still, in neuroscience and brain research, additional concepts are typically used to describe results and help guiding experimentation that are 'outside' physics, beginning with neurons and synapses, names of brain parts and areas, up to concepts like 'learning', 'motivation', 'attention'. Certainly, we do not yet have one theory that includes all these concepts. So 'brain theory' is still in a 'pre-newtonian' state. However, it may still be useful to understand in general the relations between a larger theory and its 'parts', or between microscopic and macroscopic theories, or between theories at different 'levels' of description. This is what I plan to do.

From Spiking Predictive Coding to Learning Abstract Object Representation

In a first part of the talk, I will present Predictive Coding Light (PCL), a novel unsupervised learning architecture for spiking neural networks. In contrast to conventional predictive coding approaches, which only transmit prediction errors to higher processing stages, PCL learns inhibitory lateral and top-down connectivity to suppress the most predictable spikes and passes a compressed representation of the input to higher processing stages. We show that PCL reproduces a range of biological findings and exhibits a favorable tradeoff between energy consumption and downstream classification performance on challenging benchmarks. A second part of the talk will feature our lab’s efforts to explain how infants and toddlers might learn abstract object representations without supervision. I will present deep learning models that exploit the temporal and multimodal structure of their sensory inputs to learn representations of individual objects, object categories, or abstract super-categories such as „kitchen object“ in a fully unsupervised fashion. These models offer a parsimonious account of how abstract semantic knowledge may be rooted in children's embodied first-person experiences.

“Development and application of gaze control models for active perception”

Gaze shifts in humans serve to direct high-resolution vision provided by the fovea towards areas in the environment. Gaze can be considered a proxy for attention or indicator of the relative importance of different parts of the environment. In this talk, we discuss the development of generative models of human gaze in response to visual input. We discuss how such models can be learned, both using supervised learning and using implicit feedback as an agent interacts with the environment, the latter being more plausible in biological agents. We also discuss two ways such models can be used. First, they can be used to improve the performance of artificial autonomous systems, in applications such as autonomous navigation. Second, because these models are contingent on the human’s task, goals, and/or state in the context of the environment, observations of gaze can be used to infer information about user intent. This information can be used to improve human-machine and human robot interaction, by making interfaces more anticipative. We discuss example applications in gaze-typing, robotic tele-operation and human-robot interaction.

Neurobiological constraints on learning: bug or feature?

Understanding how brains learn requires bridging evidence across scales—from behaviour and neural circuits to cells, synapses, and molecules. In our work, we use computational modelling and data analysis to explore how the physical properties of neurons and neural circuits constrain learning. These include limits imposed by brain wiring, energy availability, molecular noise, and the 3D structure of dendritic spines. In this talk I will describe one such project testing if wiring motifs from fly brain connectomes can improve performance of reservoir computers, a type of recurrent neural network. The hope is that these insights into brain learning will lead to improved learning algorithms for artificial systems.

Restoring Sight to the Blind: Effects of Structural and Functional Plasticity

Visual restoration after decades of blindness is now becoming possible by means of retinal and cortical prostheses, as well as emerging stem cell and gene therapeutic approaches. After restoring visual perception, however, a key question remains. Are there optimal means and methods for retraining the visual cortex to process visual inputs, and for learning or relearning to “see”? Up to this point, it has been largely assumed that if the sensory loss is visual, then the rehabilitation focus should also be primarily visual. However, the other senses play a key role in visual rehabilitation due to the plastic repurposing of visual cortex during blindness by audition and somatosensation, and also to the reintegration of restored vision with the other senses. I will present multisensory neuroimaging results, cortical thickness changes, as well as behavioral outcomes for patients with Retinitis Pigmentosa (RP), which causes blindness by destroying photoreceptors in the retina. These patients have had their vision partially restored by the implantation of a retinal prosthesis, which electrically stimulates still viable retinal ganglion cells in the eye. Our multisensory and structural neuroimaging and behavioral results suggest a new, holistic concept of visual rehabilitation that leverages rather than neglects audition, somatosensation, and other sensory modalities.

Understanding reward-guided learning using large-scale datasets

Understanding the neural mechanisms of reward-guided learning is a long-standing goal of computational neuroscience. Recent methodological innovations enable us to collect ever larger neural and behavioral datasets. This presents opportunities to achieve greater understanding of learning in the brain at scale, as well as methodological challenges. In the first part of the talk, I will discuss our recent insights into the mechanisms by which zebra finch songbirds learn to sing. Dopamine has been long thought to guide reward-based trial-and-error learning by encoding reward prediction errors. However, it is unknown whether the learning of natural behaviours, such as developmental vocal learning, occurs through dopamine-based reinforcement. Longitudinal recordings of dopamine and bird songs reveal that dopamine activity is indeed consistent with encoding a reward prediction error during naturalistic learning. In the second part of the talk, I will talk about recent work we are doing at DeepMind to develop tools for automatically discovering interpretable models of behavior directly from animal choice data. Our method, dubbed CogFunSearch, uses LLMs within an evolutionary search process in order to "discover" novel models in the form of Python programs that excel at accurately predicting animal behavior during reward-guided learning. The discovered programs reveal novel patterns of learning and choice behavior that update our understanding of how the brain solves reinforcement learning problems.

Harnessing Big Data in Neuroscience: From Mapping Brain Connectivity to Predicting Traumatic Brain Injury

Neuroscience is experiencing unprecedented growth in dataset size both within individual brains and across populations. Large-scale, multimodal datasets are transforming our understanding of brain structure and function, creating opportunities to address previously unexplored questions. However, managing this increasing data volume requires new training and technology approaches. Modern data technologies are reshaping neuroscience by enabling researchers to tackle complex questions within a Ph.D. or postdoctoral timeframe. I will discuss cloud-based platforms such as brainlife.io, that provide scalable, reproducible, and accessible computational infrastructure. Modern data technology can democratize neuroscience, accelerate discovery and foster scientific transparency and collaboration. Concrete examples will illustrate how these technologies can be applied to mapping brain connectivity, studying human learning and development, and developing predictive models for traumatic brain injury (TBI). By integrating cloud computing and scalable data-sharing frameworks, neuroscience can become more impactful, inclusive, and data-driven..

Motor learning selectively strengthens cortical and striatal synapses of motor engram neurons

Join Us for the Memory Decoding Journal Club! A collaboration of the Carboncopies Foundation and BPF Aspirational Neuroscience. This time, we’re diving into a groundbreaking paper: "Motor learning selectively strengthens cortical and striatal synapses of motor engram neurons

Fear learning induces synaptic potentiation between engram neurons in the rat lateral amygdala

Fear learning induces synaptic potentiation between engram neurons in the rat lateral amygdala. This study by Marios Abatis et al. demonstrates how fear conditioning strengthens synaptic connections between engram cells in the lateral amygdala, revealed through optogenetic identification of neuronal ensembles and electrophysiological measurements. The work provides crucial insights into memory formation mechanisms at the synaptic level, with implications for understanding anxiety disorders and developing targeted interventions. Presented by Dr. Kenneth Hayworth, this journal club will explore the paper's methodology linking engram cell reactivation with synaptic plasticity measurements, and discuss implications for memory decoding research.

Cognitive maps as expectations learned across episodes – a model of the two dentate gyrus blades

How can the hippocampal system transition from episodic one-shot learning to a multi-shot learning regime and what is the utility of the resultant neural representations? This talk will explore the role of the dentate gyrus (DG) anatomy in this context. The canonical DG model suggests it performs pattern separation. More recent experimental results challenge this standard model, suggesting DG function is more complex and also supports the precise binding of objects and events to space and the integration of information across episodes. Very recent studies attribute pattern separation and pattern integration to anatomically distinct parts of the DG (the suprapyramidal blade vs the infrapyramidal blade). We propose a computational model that investigates this distinction. In the model the two processing streams (potentially localized in separate blades) contribute to the storage of distinct episodic memories, and the integration of information across episodes, respectively. The latter forms generalized expectations across episodes, eventually forming a cognitive map. We train the model with two data sets, MNIST and plausible entorhinal cortex inputs. The comparison between the two streams allows for the calculation of a prediction error, which can drive the storage of poorly predicted memories and the forgetting of well-predicted memories. We suggest that differential processing across the DG aids in the iterative construction of spatial cognitive maps to serve the generation of location-dependent expectations, while at the same time preserving episodic memory traces of idiosyncratic events.

Learning Representations of Complex Meaning in the Human Brain

Circuit Mechanisms of Remote Memory

Memories of emotionally-salient events are long-lasting, guiding behavior from minutes to years after learning. The prelimbic cortex (PL) is required for fear memory retrieval across time and is densely interconnected with many subcortical and cortical areas involved in recent and remote memory recall, including the temporal association area (TeA). While the behavioral expression of a memory may remain constant over time, the neural activity mediating memory-guided behavior is dynamic. In PL, different neurons underlie recent and remote memory retrieval and remote memory-encoding neurons have preferential functional connectivity with cortical association areas, including TeA. TeA plays a preferential role in remote compared to recent memory retrieval, yet how TeA circuits drive remote memory retrieval remains poorly understood. Here we used a combination of activity-dependent neuronal tagging, viral circuit mapping and miniscope imaging to investigate the role of the PL-TeA circuit in fear memory retrieval across time in mice. We show that PL memory ensembles recruit PL-TeA neurons across time, and that PL-TeA neurons have enhanced encoding of salient cues and behaviors at remote timepoints. This recruitment depends upon ongoing synaptic activity in the learning-activated PL ensemble. Our results reveal a novel circuit encoding remote memory and provide insight into the principles of memory circuit reorganization across time.

Dimensionality reduction beyond neural subspaces

Over the past decade, neural representations have been studied from the lens of low-dimensional subspaces defined by the co-activation of neurons. However, this view has overlooked other forms of covarying structure in neural activity, including i) condition-specific high-dimensional neural sequences, and ii) representations that change over time due to learning or drift. In this talk, I will present a new framework that extends the classic view towards additional types of covariability that are not constrained to a fixed, low-dimensional subspace. In addition, I will present sliceTCA, a new tensor decomposition that captures and demixes these different types of covariability to reveal task-relevant structure in neural activity. Finally, I will close with some thoughts regarding the circuit mechanisms that could generate mixed covariability. Together this work points to a need to consider new possibilities for how neural populations encode sensory, cognitive, and behavioral variables beyond neural subspaces.

The Neurobiology of the Addicted Brain

Screen Savers : Protecting adolescent mental health in a digital world

In our rapidly evolving digital world, there is increasing concern about the impact of digital technologies and social media on the mental health of young people. Policymakers and the public are nervous. Psychologists are facing mounting pressures to deliver evidence that can inform policies and practices to safeguard both young people and society at large. However, research progress is slow while technological change is accelerating.My talk will reflect on this, both as a question of psychological science and metascience. Digital companies have designed highly popular environments that differ in important ways from traditional offline spaces. By revisiting the foundations of psychology (e.g. development and cognition) and considering digital changes' impact on theories and findings, we gain deeper insights into questions such as the following. (1) How do digital environments exacerbate developmental vulnerabilities that predispose young people to mental health conditions? (2) How do digital designs interact with cognitive and learning processes, formalised through computational approaches such as reinforcement learning or Bayesian modelling?However, we also need to face deeper questions about what it means to do science about new technologies and the challenge of keeping pace with technological advancements. Therefore, I discuss the concept of ‘fast science’, where, during crises, scientists might lower their standards of evidence to come to conclusions quicker. Might psychologists want to take this approach in the face of technological change and looming concerns? The talk concludes with a discussion of such strategies for 21st-century psychology research in the era of digitalization.

Decision and Behavior

This webinar addressed computational perspectives on how animals and humans make decisions, spanning normative, descriptive, and mechanistic models. Sam Gershman (Harvard) presented a capacity-limited reinforcement learning framework in which policies are compressed under an information bottleneck constraint. This approach predicts pervasive perseveration, stimulus‐independent “default” actions, and trade-offs between complexity and reward. Such policy compression reconciles observed action stochasticity and response time patterns with an optimal balance between learning capacity and performance. Jonathan Pillow (Princeton) discussed flexible descriptive models for tracking time-varying policies in animals. He introduced dynamic Generalized Linear Models (Sidetrack) and hidden Markov models (GLM-HMMs) that capture day-to-day and trial-to-trial fluctuations in choice behavior, including abrupt switches between “engaged” and “disengaged” states. These models provide new insights into how animals’ strategies evolve under learning. Finally, Kenji Doya (OIST) highlighted the importance of unifying reinforcement learning with Bayesian inference, exploring how cortical-basal ganglia networks might implement model-based and model-free strategies. He also described Japan’s Brain/MINDS 2.0 and Digital Brain initiatives, aiming to integrate multimodal data and computational principles into cohesive “digital brains.”

Learning and Memory

This webinar on learning and memory features three experts—Nicolas Brunel, Ashok Litwin-Kumar, and Julijana Gjorgieva—who present theoretical and computational approaches to understanding how neural circuits acquire and store information across different scales. Brunel discusses calcium-based plasticity and how standard “Hebbian-like” plasticity rules inferred from in vitro or in vivo datasets constrain synaptic dynamics, aligning with classical observations (e.g., STDP) and explaining how synaptic connectivity shapes memory. Litwin-Kumar explores insights from the fruit fly connectome, emphasizing how the mushroom body—a key site for associative learning—implements a high-dimensional, random representation of sensory features. Convergent dopaminergic inputs gate plasticity, reflecting a high-dimensional “critic” that refines behavior. Feedback loops within the mushroom body further reveal sophisticated interactions between learning signals and action selection. Gjorgieva examines how activity-dependent plasticity rules shape circuitry from the subcellular (e.g., synaptic clustering on dendrites) to the cortical network level. She demonstrates how spontaneous activity during development, Hebbian competition, and inhibitory-excitatory balance collectively establish connectivity motifs responsible for key computations such as response normalization.

Brain circuits for spatial navigation

In this webinar on spatial navigation circuits, three researchers—Ann Hermundstad, Ila Fiete, and Barbara Webb—discussed how diverse species solve navigation problems using specialized yet evolutionarily conserved brain structures. Hermundstad illustrated the fruit fly’s central complex, focusing on how hardwired circuit motifs (e.g., sinusoidal steering curves) enable rapid, flexible learning of goal-directed navigation. This framework combines internal heading representations with modifiable goal signals, leveraging activity-dependent plasticity to adapt to new environments. Fiete explored the mammalian head-direction system, demonstrating how population recordings reveal a one-dimensional ring attractor underlying continuous integration of angular velocity. She showed that key theoretical predictions—low-dimensional manifold structure, isometry, uniform stability—are experimentally validated, underscoring parallels to insect circuits. Finally, Webb described honeybee navigation, featuring path integration, vector memories, route optimization, and the famous waggle dance. She proposed that allocentric velocity signals and vector manipulation within the central complex can encode and transmit distances and directions, enabling both sophisticated foraging and inter-bee communication via dance-based cues.

Understanding the complex behaviors of the ‘simple’ cerebellar circuit

Every movement we make requires us to precisely coordinate muscle activity across our body in space and time. In this talk I will describe our efforts to understand how the brain generates flexible, coordinated movement. We have taken a behavior-centric approach to this problem, starting with the development of quantitative frameworks for mouse locomotion (LocoMouse; Machado et al., eLife 2015, 2020) and locomotor learning, in which mice adapt their locomotor symmetry in response to environmental perturbations (Darmohray et al., Neuron 2019). Combined with genetic circuit dissection, these studies reveal specific, cerebellum-dependent features of these complex, whole-body behaviors. This provides a key entry point for understanding how neural computations within the highly stereotyped cerebellar circuit support the precise coordination of muscle activity in space and time. Finally, I will present recent unpublished data that provide surprising insights into how cerebellar circuits flexibly coordinate whole-body movements in dynamic environments.

Brain-Wide Compositionality and Learning Dynamics in Biological Agents

Biological agents continually reconcile the internal states of their brain circuits with incoming sensory and environmental evidence to evaluate when and how to act. The brains of biological agents, including animals and humans, exploit many evolutionary innovations, chiefly modularity—observable at the level of anatomically-defined brain regions, cortical layers, and cell types among others—that can be repurposed in a compositional manner to endow the animal with a highly flexible behavioral repertoire. Accordingly, their behaviors show their own modularity, yet such behavioral modules seldom correspond directly to traditional notions of modularity in brains. It remains unclear how to link neural and behavioral modularity in a compositional manner. We propose a comprehensive framework—compositional modes—to identify overarching compositionality spanning specialized submodules, such as brain regions. Our framework directly links the behavioral repertoire with distributed patterns of population activity, brain-wide, at multiple concurrent spatial and temporal scales. Using whole-brain recordings of zebrafish brains, we introduce an unsupervised pipeline based on neural network models, constrained by biological data, to reveal highly conserved compositional modes across individuals despite the naturalistic (spontaneous or task-independent) nature of their behaviors. These modes provided a scaffolding for other modes that account for the idiosyncratic behavior of each fish. We then demonstrate experimentally that compositional modes can be manipulated in a consistent manner by behavioral and pharmacological perturbations. Our results demonstrate that even natural behavior in different individuals can be decomposed and understood using a relatively small number of neurobehavioral modules—the compositional modes—and elucidate a compositional neural basis of behavior. This approach aligns with recent progress in understanding how reasoning capabilities and internal representational structures develop over the course of learning or training, offering insights into the modularity and flexibility in artificial and biological agents.

Unmotivated bias

In this talk, I will explore how social affective biases arise even in the absence of motivational factors as an emergent outcome of the basic structure of social learning. In several studies, we found that initial negative interactions with some members of a group can cause subsequent avoidance of the entire group, and that this avoidance perpetuates stereotypes. Additional cognitive modeling discovered that approach and avoidance behavior based on biased beliefs not only influences the evaluative (positive or negative) impressions of group members, but also shapes the depth of the cognitive representations available to learn about individuals. In other words, people have richer cognitive representations of members of groups that are not avoided, akin to individualized vs group level categories. I will end presenting a series of multi-agent reinforcement learning simulations that demonstrate the emergence of these social-structural feedback loops in the development and maintenance of affective biases.

Emergence of behavioural individuality from global microstructure of the brain and learning

Contribution of computational models of reinforcement learning to neurosciences/ computational modeling, reward, learning, decision-making, conditioning, navigation, dopamine, basal ganglia, prefrontal cortex, hippocampus

Decomposing motivation into value and salience

Humans and other animals approach reward and avoid punishment and pay attention to cues predicting these events. Such motivated behavior thus appears to be guided by value, which directs behavior towards or away from positively or negatively valenced outcomes. Moreover, it is facilitated by (top-down) salience, which enhances attention to behaviorally relevant learned cues predicting the occurrence of valenced outcomes. Using human neuroimaging, we recently separated value (ventral striatum, posterior ventromedial prefrontal cortex) from salience (anterior ventromedial cortex, occipital cortex) in the domain of liquid reward and punishment. Moreover, we investigated potential drivers of learned salience: the probability and uncertainty with which valenced and non-valenced outcomes occur. We find that the brain dissociates valenced from non-valenced probability and uncertainty, which indicates that reinforcement matters for the brain, in addition to information provided by probability and uncertainty alone, regardless of valence. Finally, we assessed learning signals (unsigned prediction errors) that may underpin the acquisition of salience. Particularly the insula appears to be central for this function, encoding a subjective salience prediction error, similarly at the time of positively and negatively valenced outcomes. However, it appears to employ domain-specific time constants, leading to stronger salience signals in the aversive than the appetitive domain at the time of cues. These findings explain why previous research associated the insula with both valence-independent salience processing and with preferential encoding of the aversive domain. More generally, the distinction of value and salience appears to provide a useful framework for capturing the neural basis of motivated behavior.

Feedback-induced dispositional changes in risk preferences

Contrary to the original normative decision-making standpoint, empirical studies have repeatedly reported that risk preferences are affected by the disclosure of choice outcomes (feedback). Although no consensus has yet emerged regarding the properties and mechanisms of this effect, a widespread and intuitive hypothesis is that repeated feedback affects risk preferences by means of a learning effect, which alters the representation of subjective probabilities. Here, we ran a series of seven experiments (N= 538), tailored to decipher the effects of feedback on risk preferences. Our results indicate that the presence of feedback consistently increases risk-taking, even when the risky option is economically less advantageous. Crucially, risk-taking increases just after the instructions, before participants experience any feedback. These results challenge the learning account, and advocate for a dispositional effect, induced by the mere anticipation of feedback information. Epistemic curiosity and regret avoidance may drive this effect in partial and complete feedback conditions, respectively.

Use case determines the validity of neural systems comparisons

Deep learning provides new data-driven tools to relate neural activity to perception and cognition, aiding scientists in developing theories of neural computation that increasingly resemble biological systems both at the level of behavior and of neural activity. But what in a deep neural network should correspond to what in a biological system? This question is addressed implicitly in the use of comparison measures that relate specific neural or behavioral dimensions via a particular functional form. However, distinct comparison methodologies can give conflicting results in recovering even a known ground-truth model in an idealized setting, leaving open the question of what to conclude from the outcome of a systems comparison using any given methodology. Here, we develop a framework to make explicit and quantitative the effect of both hypothesis-driven aspects—such as details of the architecture of a deep neural network—as well as methodological choices in a systems comparison setting. We demonstrate via the learning dynamics of deep neural networks that, while the role of the comparison methodology is often de-emphasized relative to hypothesis-driven aspects, this choice can impact and even invert the conclusions to be drawn from a comparison between neural systems. We provide evidence that the right way to adjudicate a comparison depends on the use case—the scientific hypothesis under investigation—which could range from identifying single-neuron or circuit-level correspondences to capturing generalizability to new stimulus properties

Localisation of Seizure Onset Zone in Epilepsy Using Time Series Analysis of Intracranial Data

There are over 30 million people with drug-resistant epilepsy worldwide. When neuroimaging and non-invasive neural recordings fail to localise seizure onset zones (SOZ), intracranial recordings become the best chance for localisation and seizure-freedom in those patients. However, intracranial neural activities remain hard to visually discriminate across recording channels, which limits the success of intracranial visual investigations. In this presentation, I present methods which quantify intracranial neural time series and combine them with explainable machine learning algorithms to localise the SOZ in the epileptic brain. I present the potentials and limitations of our methods in the localisation of SOZ in epilepsy providing insights for future research in this area.

On finding what you’re (not) looking for: prospects and challenges for AI-driven discovery

Recent high-profile scientific achievements by machine learning (ML) and especially deep learning (DL) systems have reinvigorated interest in ML for automated scientific discovery (eg, Wang et al. 2023). Much of this work is motivated by the thought that DL methods might facilitate the efficient discovery of phenomena, hypotheses, or even models or theories more efficiently than traditional, theory-driven approaches to discovery. This talk considers some of the more specific obstacles to automated, DL-driven discovery in frontier science, focusing on gravitational-wave astrophysics (GWA) as a representative case study. In the first part of the talk, we argue that despite these efforts, prospects for DL-driven discovery in GWA remain uncertain. In the second part, we advocate a shift in focus towards the ways DL can be used to augment or enhance existing discovery methods, and the epistemic virtues and vices associated with these uses. We argue that the primary epistemic virtue of many such uses is to decrease opportunity costs associated with investigating puzzling or anomalous signals, and that the right framework for evaluating these uses comes from philosophical work on pursuitworthiness.

Beyond Homogeneity: Characterizing Brain Disorder Heterogeneity through EEG and Normative Modeling

Electroencephalography (EEG) has been thoroughly studied for decades in psychiatry research. Yet its integration into clinical practice as a diagnostic/prognostic tool remains unachieved. We hypothesize that a key reason is the underlying patient's heterogeneity, overlooked in psychiatric EEG research relying on a case-control approach. We combine HD-EEG with normative modeling to quantify this heterogeneity using two well-established and extensively investigated EEG characteristics -spectral power and functional connectivity- across a cohort of 1674 patients with attention-deficit/hyperactivity disorder, autism spectrum disorder, learning disorder, or anxiety, and 560 matched controls. Normative models showed that deviations from population norms among patients were highly heterogeneous and frequency-dependent. Deviation spatial overlap across patients did not exceed 40% and 24% for spectral and connectivity, respectively. Considering individual deviations in patients has significantly enhanced comparative analysis, and the identification of patient-specific markers has demonstrated a correlation with clinical assessments, representing a crucial step towards attaining precision psychiatry through EEG.

Top-down models of learning and decision-making in BG

Physical Activity, Sedentary Behaviour and Brain Health

Prosocial Learning and Motivation across the Lifespan

2024 BACN Early-Career Prize Lecture Many of our decisions affect other people. Our choices can decelerate climate change, stop the spread of infectious diseases, and directly help or harm others. Prosocial behaviours – decisions that help others – could contribute to reducing the impact of these challenges, yet their computational and neural mechanisms remain poorly understood. I will present recent work that examines prosocial motivation, how willing we are to incur costs to help others, prosocial learning, how we learn from the outcomes of our choices when they affect other people, and prosocial preferences, our self-reports of helping others. Throughout the talk, I will outline the possible computational and neural bases of these behaviours, and how they may differ from young adulthood to old age.

Neural mechanisms governing the learning and execution of avoidance behavior

The nervous system orchestrates adaptive behaviors by intricately coordinating responses to internal cues and environmental stimuli. This involves integrating sensory input, managing competing motivational states, and drawing on past experiences to anticipate future outcomes. While traditional models attribute this complexity to interactions between the mesocorticolimbic system and hypothalamic centers, the specific nodes of integration have remained elusive. Recent research, including our own, sheds light on the midline thalamus's overlooked role in this process. We propose that the midline thalamus integrates internal states with memory and emotional signals to guide adaptive behaviors. Our investigations into midline thalamic neuronal circuits have provided crucial insights into the neural mechanisms behind flexibility and adaptability. Understanding these processes is essential for deciphering human behavior and conditions marked by impaired motivation and emotional processing. Our research aims to contribute to this understanding, paving the way for targeted interventions and therapies to address such impairments.

Visuomotor learning of location, action, and prediction

Probing neural population dynamics with recurrent neural networks

Large-scale recordings of neural activity are providing new opportunities to study network-level dynamics with unprecedented detail. However, the sheer volume of data and its dynamical complexity are major barriers to uncovering and interpreting these dynamics. I will present latent factor analysis via dynamical systems, a sequential autoencoding approach that enables inference of dynamics from neuronal population spiking activity on single trials and millisecond timescales. I will also discuss recent adaptations of the method to uncover dynamics from neural activity recorded via 2P Calcium imaging. Finally, time permitting, I will mention recent efforts to improve the interpretability of deep-learning based dynamical systems models.

Trends in NeuroAI - Brain-like topography in transformers (Topoformer)

Dr. Nicholas Blauch will present on his work "Topoformer: Brain-like topographic organization in transformer language models through spatial querying and reweighting". Dr. Blauch is a postdoctoral fellow in the Harvard Vision Lab advised by Talia Konkle and George Alvarez. Paper link: https://openreview.net/pdf?id=3pLMzgoZSA Trends in NeuroAI is a reading group hosted by the MedARC Neuroimaging & AI lab (https://medarc.ai/fmri | https://groups.google.com/g/medarc-fmri).

Mapping the Brain‘s Visual Representations Using Deep Learning

Generative models for video games (rescheduled)

Developing agents capable of modeling complex environments and human behaviors within them is a key goal of artificial intelligence research. Progress towards this goal has exciting potential for applications in video games, from new tools that empower game developers to realize new creative visions, to enabling new kinds of immersive player experiences. This talk focuses on recent advances of my team at Microsoft Research towards scalable machine learning architectures that effectively capture human gameplay data. In the first part of my talk, I will focus on diffusion models as generative models of human behavior. Previously shown to have impressive image generation capabilities, I present insights that unlock applications to imitation learning for sequential decision making. In the second part of my talk, I discuss a recent project taking ideas from language modeling to build a generative sequence model of an Xbox game.

Applied cognitive neuroscience to improve learning and therapeutics

Advancements in cognitive neuroscience have provided profound insights into the workings of the human brain and the methods used offer opportunities to enhance performance, cognition, and mental health. Drawing upon interdisciplinary collaborations in the University of California San Diego, Human Performance Optimization Lab, this talk explores the application of cognitive neuroscience principles in three domains to improve human performance and alleviate mental health challenges. The first section will discuss studies addressing the role of vision and oculomotor function in athletic performance and the potential to train these foundational abilities to improve performance and sports outcomes. The second domain considers the use of electrophysiological measurements of the brain and heart to detect, and possibly predict, errors in manual performance, as shown in a series of studies with surgeons as they perform robot-assisted surgery. Lastly, findings from clinical trials testing personalized interventional treatments for mood disorders will be discussed in which the temporal and spatial parameters of transcranial magnetic stimulation (TMS) are individualized to test if personalization improves treatment response and can be used as predictive biomarkers to guide treatment selection. Together, these translational studies use the measurement tools and constructs of cognitive neuroscience to improve human performance and well-being.

The multi-phase plasticity supporting winner effect

Aggression is an innate behavior across animal species. It is essential for competing for food, defending territory, securing mates, and protecting families and oneself. Since initiating an attack requires no explicit learning, the neural circuit underlying aggression is believed to be genetically and developmentally hardwired. Despite being innate, aggression is highly plastic. It is influenced by a wide variety of experiences, particularly winning and losing previous encounters. Numerous studies have shown that winning leads to an increased tendency to fight while losing leads to flight in future encounters. In the talk, I will present our recent findings regarding the neural mechanisms underlying the behavioral changes caused by winning.

Generative models for video games

Developing agents capable of modeling complex environments and human behaviors within them is a key goal of artificial intelligence research. Progress towards this goal has exciting potential for applications in video games, from new tools that empower game developers to realize new creative visions, to enabling new kinds of immersive player experiences. This talk focuses on recent advances of my team at Microsoft Research towards scalable machine learning architectures that effectively capture human gameplay data. In the first part of my talk, I will focus on diffusion models as generative models of human behavior. Previously shown to have impressive image generation capabilities, I present insights that unlock applications to imitation learning for sequential decision making. In the second part of my talk, I discuss a recent project taking ideas from language modeling to build a generative sequence model of an Xbox game.

Cell-type-specific plasticity shapes neocortical dynamics for motor learning

How do cortical circuits acquire new dynamics that drive learned movements? This webinar will focus on mouse premotor cortex in relation to learned lick-timing and explore high-density electrophysiology using our silicon neural probes alongside region and cell-type-specific acute genetic manipulations of proteins required for synaptic plasticity.

Learning representations of specifics and generalities over time

There is a fundamental tension between storing discrete traces of individual experiences, which allows recall of particular moments in our past without interference, and extracting regularities across these experiences, which supports generalization and prediction in similar situations in the future. One influential proposal for how the brain resolves this tension is that it separates the processes anatomically into Complementary Learning Systems, with the hippocampus rapidly encoding individual episodes and the neocortex slowly extracting regularities over days, months, and years. But this does not explain our ability to learn and generalize from new regularities in our environment quickly, often within minutes. We have put forward a neural network model of the hippocampus that suggests that the hippocampus itself may contain complementary learning systems, with one pathway specializing in the rapid learning of regularities and a separate pathway handling the region’s classic episodic memory functions. This proposal has broad implications for how we learn and represent novel information of specific and generalized types, which we test across statistical learning, inference, and category learning paradigms. We also explore how this system interacts with slower-learning neocortical memory systems, with empirical and modeling investigations into how the hippocampus shapes neocortical representations during sleep. Together, the work helps us understand how structured information in our environment is initially encoded and how it then transforms over time.

Maintaining Plasticity in Neural Networks

Nonstationarity presents a variety of challenges for machine learning systems. One surprising pathology which can arise in nonstationary learning problems is plasticity loss, whereby making progress on new learning objectives becomes more difficult as training progresses. Networks which are unable to adapt in response to changes in their environment experience plateaus or even declines in performance in highly non-stationary domains such as reinforcement learning, where the learner must quickly adapt to new information even after hundreds of millions of optimization steps. The loss of plasticity manifests in a cluster of related empirical phenomena which have been identified by a number of recent works, including the primacy bias, implicit under-parameterization, rank collapse, and capacity loss. While this phenomenon is widely observed, it is still not fully understood. This talk will present exciting recent results which shed light on the mechanisms driving the loss of plasticity in a variety of learning problems and survey methods to maintain network plasticity in non-stationary tasks, with a particular focus on deep reinforcement learning.

Neuromatch 5

Neuromatch 5 (Neuromatch Conference 2022) was a fully virtual conference focused on computational neuroscience broadly construed, including machine learning work with explicit biological links. After four successful Neuromatch conferences, the fifth edition consolidated proven innovations from past events, featuring a series of talks hosted on Crowdcast and flash talk sessions (pre-recorded videos) with dedicated discussion times on Reddit.

OPTIMIZING VISUOSPATIAL WORKING MEMORY THROUGH ADAPTIVE DIGITAL LEARNING: BEHAVIORAL AND EEG EVIDENCE

FENS Forum 2026

Analysis of burst sequences in mouse prefrontal cortex during learning

Bernstein Conference 2024

Learning predictable factors from sequences: it’s not only about slow features

Bernstein Conference 2024

Biological-plausible learning with a two compartment neuron model in recurrent neural networks

Bernstein Conference 2024

Building mechanistic models of neural computations with simulation-based machine learning

Bernstein Conference 2024

Competition and integration of sensory signals in a deep reinforcement learning agent

Bernstein Conference 2024

Effect of experience on context-dependent learning in recurrent networks

Bernstein Conference 2024

Continual learning using dendritic modulations on view-invariant feedforward weights

Bernstein Conference 2024

Controversial Opinions on Model Based and Model Free Reinforcement Learning in the Brain

Bernstein Conference 2024

Correcting cortical output: a distributed learning framework for motor adaptation

Bernstein Conference 2024

The cost of behavioral flexibility: a modeling study of reversal learning using a spiking neural network

Bernstein Conference 2024

Dynamics of Supervised and Reinforcement Learning in the Non-Linear Perceptron

Bernstein Conference 2024

DelGrad: Exact gradients in spiking networks for learning transmission delays and weights

Bernstein Conference 2024

Dendrites endow artificial neural networks with accurate, robust and parameter-efficient learning

Bernstein Conference 2024

Computational implications of motor primitives for cortical motor learning

Bernstein Conference 2024

Dual role, single pathway: A pyramidal cell model of feedback integration in function and learning

Bernstein Conference 2024

Efficient learning of deep non-negative matrix factorisation networks

Bernstein Conference 2024

Enhancing learning through neuromodulation-aware spiking neural networks

Bernstein Conference 2024

Evaluating Memory Behavior in Continual Learning using the Posterior in a Binary Bayesian Network

Bernstein Conference 2024

Evolutionary algorithms support recurrent plasticity in spiking neural network models of neocortical task learning

Bernstein Conference 2024

Excitatory and inhibitory neurons exhibit distinct roles for task learning, temporal scaling, and working memory in recurrent spiking neural network models of neocortex.

Bernstein Conference 2024

Experiment-based Models to Study Local Learning Rules for Spiking Neural Networks

Bernstein Conference 2024

Dissecting the Factors of Metaplasticity with Meta-Continual Learning

COSYNE 2022

A feedback control algorithm for online learning in Spiking Neural Networks and Neuromorphic devices

Bernstein Conference 2024

Identifying cortical learning algorithms using Brain-Machine Interfaces

Bernstein Conference 2024

Inhibition-controlled Hebbian learning unifies phenomenological and normative models of plasticity

Bernstein Conference 2024

Knocking out co-active plasticity rules in neural networks reveals synapse type-specific contributions for learning and memory

Bernstein Conference 2024

Learning an environment model in real-time with core knowledge and closed-loop behaviours

Bernstein Conference 2024

Learning Hebbian/Anti-Hebbian networks in continuous time

Bernstein Conference 2024

Learning neuronal manifolds for interacting neuronal populations

Bernstein Conference 2024

How Do Bees See the World? A (Normative) Deep Reinforcement Learning Model for Insect Navigation

Bernstein Conference 2024

Learning and using predictive maps for strategic planning

Bernstein Conference 2024

Linking Spontaneous Synaptic Activity to Learning

Bernstein Conference 2024

Neuromodulated online cognitive maps for reinforcement learning

Bernstein Conference 2024

Object detection with deep learning and attention feedback loops

Bernstein Conference 2024

Quantifying the learning dynamics of single subjects in a reversal learning task with change point analysis

Bernstein Conference 2024

Replay of Chaotic Dynamics through Differential Hebbian Learning with Transmission Delays

Bernstein Conference 2024

Response variability can accelerate learning in feedforward-recurrent networks

Bernstein Conference 2024

'Reusers' and 'Unlearners' display distinct effects of forgetting on reversal learning in neural networks

Bernstein Conference 2024

Co-Design of Analog Neuromorphic Systems and Cortical Motifs with Local Dendritic Learning Rules

Bernstein Conference 2024