New Partner Projects

This project investigates how animals make decisions in uncertain environments by studying the neural mechanisms underlying flexible foraging behavior. Head-fixed mice perform a continuous virtual foraging task in which they decide when to exploit a reward source and when to move to a new one, requiring them to infer hidden changes in reward availability from past outcomes. Using Neuropixels electrophysiology, behavioral measurements, and video recordings of facial movements, the project will examine how distributed brain circuits encode decision variables and behavioral strategies. By comparing neural activity across animals and across tasks, the project aims to determine whether behavioral variability can be explained by continuous differences in underlying computational processes, rather than by discrete strategies.

The IBL Core will support the project by helping standardize and organize large-scale datasets generated during these experiments, ensuring compatibility with the International Brain Laboratory data ecosystem. This includes guidance on data formatting, quality control procedures, automated processing pipelines for electrophysiology and video data, and integration of behavioral datasets with neural recordings. The collaboration will also support data sharing and open science practices by preparing curated datasets and documentation for public release, enabling the broader neuroscience community to reuse these data and analytical tools.

This project investigates how distributed brain circuits support rapid learning during reward-based sensorimotor decision-making. In a head-fixed behavioral paradigm, mice learn within a single session to respond appropriately to sensory stimuli associated with reward. The task relies on active whisker-based sensing, where mice detect sensory cues and receive feedback through whisker contact with the environment during learning. Using simultaneous Neuropixels recordings from multiple brain regions, the project will measure neuronal activity across the brain while animals perform the task, allowing researchers to examine how neural dynamics evolve during learning. By analyzing activity across cortical and subcortical structures, the study aims to reveal how sensory signals, whisker-based feedback, motor planning, and reward-related information interact across large-scale brain networks to guide behavior during learning.

The IBL Core will support the project by collaborating on the analysis of these large-scale electrophysiology datasets and by providing guidance on data management, electrophysiology data processing, and quality control procedures. The collaboration will also contribute to scalable analysis pipelines for studying brain-wide neural activity and functional interactions between neurons across regions. In addition, the IBL Core will provide consultation on behavioral video analysis and support efforts to document and standardize the experimental setup and task, enabling the paradigm to be adopted by other laboratories and facilitating broader collaboration within the neuroscience community.

This project aims to investigate the role of breathing rhythms in behavior and brain activity by integrating respiration measurements into the International Brain Laboratory (IBL) head-fixed decision-making paradigm. Breathing is increasingly recognized as an important physiological signal that can coordinate neural activity across brain regions and influence sensory processing, cognition, and decision-making. The project will develop computational methods to decode respiration directly from orofacial video recordings collected during behavior. By combining ground-truth breathing measurements with high-resolution behavioral video, the team will train and validate models capable of reconstructing breathing signals from video features. If successful, this approach would enable respiration to be inferred from existing video recordings, potentially adding a new physiological dimension to the IBL brain-wide dataset.

The IBL Core will support the project by providing consultation on the design of an IBL-compatible behavioral rig incorporating respiration measurements, guidance on video analysis and pose-tracking pipelines, and advice on data organization compatible with IBL data standards. The collaboration will also support the development of scalable pipelines for video processing and data synchronization. In later stages, the breathing-decoding model may be deployed on IBL datasets to explore how respiration relates to behavior and brain-wide neural activity, while the Cavelli laboratory will develop and validate the decoding algorithms and generate the experimental datasets.

This project aims to link detailed facial biomechanics with brain-wide neural activity by integrating the Cheese3D multiview facial tracking system with the International Brain Laboratory (IBL) ecosystem. Cheese3D is a high-speed, multi-camera system that reconstructs three-dimensional facial kinematics of head-fixed mice with micron-scale precision across multiple anatomical landmarks, including the eyes, ears, whisker pad, and jaw. By capturing coordinated motion across the face, the system provides a rich behavioral readout that can reflect internal states and neural activity. The project will combine these multiview facial recordings with electrophysiology to investigate how facial muscle dynamics relate to neural population activity across distributed motor and limbic circuits.

The collaboration with the IBL Core will focus on integrating Cheese3D with modern machine-learning approaches and IBL data infrastructure. This includes training multiview pose estimation models using LightningPose, applying self-supervised video representation learning through the BEAST framework, and linking facial movement embeddings with Neuropixels recordings. Together, these efforts aim to create a scalable pipeline for multiview facial tracking and neural decoding, enabling researchers to infer internal state and neural activity from facial movement patterns. The resulting datasets, models, and analysis tools will be released as open-source resources to support broader use within the neuroscience community.

This project, led jointly by the Neuroinformatics Unit (NIU) at UCL, the Allen Institute, and the International Brain Laboratory (IBL), aims to build sustainable, general-purpose software infrastructure for systems neuroscience. The collaboration focuses on developing tools that allow researchers to more easily compare analysis pipelines and build flexible workflows for large-scale behavioural and imaging datasets.

Two priority developments will be pursued. The first is PoseInterface, a general-purpose framework for running and comparing video analysis pipelines, including pose estimation tools such as LightningPose. The second is the development of a modular multiphoton imaging framework, conceptually analogous to SpikeInterface, which will provide standardized infrastructure for constructing and benchmarking imaging processing pipelines.

These tools will be developed collaboratively and tested on real experimental datasets, with the goal of integrating them into existing neuroscience workflows where appropriate. By building reusable, open-source infrastructure and promoting community contributions, the project aims to reduce duplication across neuroscience software ecosystems and support the development of flexible, interoperable, and maintainable data-processing tools for the broader neuroscience community.

This project investigates how distributed brain networks integrate valence signals during cost–benefit decision-making. In a head-fixed behavioral paradigm, mice make choices that balance potential rewards against costs such as delay, effort, or risk. The study focuses on the basolateral amygdala (BLA) as a central hub for valence processing. By optogenetically activating projection-defined BLA subregions while performing brain-wide Neuropixels recordings, the project will examine how valence-related signals propagate through downstream brain regions and influence decision-making circuits across the brain. This approach will allow researchers to determine how changes in positive or negative valence bias choices and shape neural activity across distributed networks, providing mechanistic insight into how emotional signals influence decision processes. Because dysregulation of BLA-prefrontal interactions is implicated in psychiatric disorders such as depression, anxiety, and addiction, the project also aims to establish a general experimental framework for studying how emotional processing shapes decision-making at the circuit level.

The IBL Core will support the project by collaborating on the development and standardization of the experimental hardware and software infrastructure used in the task. This includes technical support for a synchronization device, integration with the emerging bpod-core behavioral control framework, and guidance on open-source hardware documentation and community distribution. The collaboration will also provide consultation on electrophysiology rig design, best practices for repeated Neuropixels recordings across multiple brain regions, and standardized procedures for electrophysiology data quality control to ensure compatibility with large-scale neuroscience data analysis workflows.

This project aims to develop a foundation model of primate behavior that captures the biomechanics and neural control of complex movements. Using multiview video recordings of macaques performing dexterous tasks, the project will build models that reconstruct detailed limb kinematics and infer underlying biomechanical states such as joint angles and muscle dynamics. By combining multiview pose estimation with biomechanical modeling and simulation, the project seeks to generate anatomically meaningful representations of movement that go beyond conventional keypoint tracking. These models will support the study of naturalistic motor behavior and provide a framework for linking detailed movement dynamics with neural activity.

The collaboration with the IBL Core will focus on developing robust multiview pose estimation pipelines, integrating biomechanical constraints into model training, and establishing evaluation benchmarks for behavioral foundation models. The project will also explore how different behavioral representations, such as kinematic trajectories or muscle-driven simulations, relate to neural activity. The resulting tools, models, and analysis pipelines will be released as open-source resources to support the broader neuroscience community and advance the study of complex motor behavior in primates.