Awarded to Professor Frederick Eberhardt and Professor Ralph Adolphs
Causal Discovery Algorithms for Neuroimaging Data
Adolphs and Eberhardt will apply new computational methods to determine whether it is possible to extract causal inferences from large, resting-state fMRI datasets in humans. These "causal discovery algorithms" leverage temporal sequencing and other factors to infer whether activity in one brain region is driven by activity in another brain region, or vice-versa. If successful, this method could have a transformative effect on the interpretation of correlative fMRI data.
Awarded to Professor Marianne Bronner and Professor Lior Pachter
Tensor-Based Modelling of Nervous System Development Using Multi-Plexed Single-Cell
Bronner and Pachter, in collaboration with Professors Joel Tropp and Venkat Chandrasekaran in CMS, will develop algebraic methods to try to infer developmental sequences ("time-lapse" models) in the differentiation of cells in the peripheral nervous system, based on single-cell RNA sequencing carried out on cell mixtures from zebrafish embryos at different developmental stages, where all cells from each stage are labeled by a DNA "bar-code." This will convert conventional 2D matrices of cells x genes into 3D tensors of cells x genes x developmental stages. Developing these tensor computational methods should allow this approach to be extended to many different settings.
Awarded to Professor Henry Lester and Professor Axel Scherer
Measuring Brain and Interstitial Nicotine with Wireless Sensors
Lester and Scherer will collaborate on the development and application of novel wireless sensors for measuring brain and interstitial levels of nicotine. This technology, if successful, would greatly facilitate within-animal studies of the pharmacokinetics and pharmacodynamics of nicotine, with high temporal resolution (100 ms). It has potentially important implications for the treatment of nicotine addiction in humans, and is potentially extensible to measuring other types of brain chemicals as well.
T&C CHEN CENTER FOR SYSTEMS NEUROSCIENCE AWARDS
Interfaces Between Systems
Awarded to Assistant Professor Elizabeth Hong and Assistant Professor Joe Parker
Dissecting the Neural Interface by a Symbiotic Animal and its Host
Hong and Parker will dissect the neural interface between ants and beetles in social symbiosis by identifying the chemicals and chemoreceptors beetles use to locate and successfully interact with ants. Techniques they will use include: neurotranscriptome profiling, phylogenetic analysis, gas chromotography-mass spectroscopy, and electrophysiology.
Awarded to Professor Henry Lester and Assistant Professor Matt Thomson
Molecular Basis for the Antidepressant Effects of Ketamine
Lester and Thomson will dissect the interface between depression, neural network excitability, and gene expression, by studying the effects of ketamine and other anti-depressants on neural activity and gene transcription in neuronal cell culture. Techniques they will use include: single-cell RNA-seq, electrophysiology, calcium imaging, and computational analysis.
Awarded to Professor John Doyle
Mathematical Foundations for Understanding Human Sensorimotor and Homiostic Control
Doyle will explore how a new mathematical theory of layered architectures can be applied to the brain, with a focus on the role of feedback in the visual system. He will clarify the similarities and differences between the theory and the actual anatomy and physiology of vision.
Awarded to Professor Michelle Effros
An Information Theoretic View of the Neruonal Mechanisms of Memory
Effros will explore the phenomena of memory in a network of simple, biologically realistic model neurons. She will apply information theory to understand what the optimal code is for information storage, how storing new memories affects old memories, how memories can be sequentially updated, and what the capacity limits are.
T&C CHEN CENTER FOR SOCIAL AND DECISION NEUROSCIENCE AWARDS
Awarded to Assistant Professor Dean Mobbs
Attack or Escape: The Role of the Human Hypothalamus in Reflexive Switching in Survival States
Mobbs will use MVPA methods to analyze fMRI activity in human hypothalamus at high-resolution, 1.5-1.7mm, using the new 3T Prisma upgraded awarded by NSF. Using a lifelike simplified attack-escape, they hope to identify changes in the hypothalamus and other regions as subjects switch from artificial predator to prey.
Awarded to Professor Ralph Adolphs and Professor John O'Doherty
Using Reinforcement-Learning as a Window into Between-Subject Variability in the Neural Computations Underlying Psychiatric Dysfunction
Adolphs and O'Doherty will conduct fMRI during four well-characterized human choice tasks, involving reward, model-based and model-free learning, exploration, and mentalizing. Behavior of brain activity of populations of patients with diagnosed OCD and autism spectrum disorder (ASD) will be compared to control subjects to search for novel functional biomarkers of these disorders.
Awarded to Professor Christopher Hitchcock, Professor Steven Quartz and Professor Shin Shimojo
A Novel Paradigm to Investigate Human Intention and Agency: A Neuroscience and
Hitchcock, Quartz and Shimojo will investigate a motor illusion. Subjects move their thumb at any time then choose, which then triggers an external stimulation-induced (TMS) thumb twitch. Subjects report feeling that the machine can ‘read their mind', because their thumb movement feels involuntary. This result challenges the traditional philosophical view that introspection is direct, suggesting instead that the feeling of personal agency is an inference.
T&C CHEN BRAIN-MACHINE INTERFACE CENTER AWARDS
Invasive or Non-Invasive Recording Technology
Awarded to Professor Markus Meister
Electrode Pooling: A Method to Boost the Yield of Multi Neuron Recordings
Being able to record from large numbers of neurons is essential for brain-machine interface (BMI) studies as well as systems neuroscience research. Meister, in collaboration with Dr. Tim Harris from Janelia, proposes a novel design for silicon probes that can increase channel counts by drastically reducing the number of wires needed on these multi-channel probes.
Computational Neural Prosthetics
Awarded to Professor Azita Emami
Efficient Machine Learning Algorithms for Closed-Loop Brain-Machine Interfaces
Real-time and fast learning algorithms are required to decode neural signals for BMI applications. Emami proposes to develop efficient machine learning algorithms for closed-loop BMIs. She plans to implement these learning algorithms on a single integrated circuit, and will be collaborating with Professor Yisong Yue.
Neural Plasticity for BMI learning
Awarded to Professor Joel Burdick
A Pilot Study of Joint Cortical and Spinal Plasticity During Spinally Stimulated Recovery of Voluntary Control in a rat model of severe SCI
Burdick and colleagues have shown, in rats and humans, that spinal cord stimulation below an incomplete lesion can enhance volitional control. He proposes, in rat studies, combined spinal cord and motor cortex stimulation to see if plasticity is enhanced by affecting both ends of the cortico-spinal pathway.