Ehrlich, D. B. | bioRxiv Science

bioRxiv2021

Single-Cell Transcriptional Profiling of the Adult Corticospinal Tract Reveals Forelimb and Hindlimb Molecular Specialization

The corticospinal tract (CST) is refractory to repair after CNS trauma, resulting in chronic debilitating functional motor deficits after spinal cord injury. While novel pro-axon growth activators have stimulated plasticity and regeneration of corticospinal neurons (CSNs) after injury, robust functional recovery remains elusive. These repair strategies are sub-optimal in part due to underexplored molecular heterogeneity within the developing and adult CST. In this study, we combine retrograde CST tracing with single-cell RNA sequencing to build a comprehensive atlas of CSN subtypes. By comparing CSNs to non-spinally projecting neurons in layer Vb, we identify pan-CSN markers including Wnt7b. By leveraging retrograde tracing, we are able to compare forelimb and hindlimb projecting CSNs, identifying subtype-specific markers, including Cacng7 and Slc16a2 respectively. These markers are expressed in embryonic and neonatal CSNs and can be used to study early postnatal patterning of the CST. Our results provide molecular insight into the differences between anatomically distinct CSN subtypes and provide a resource for future screening and exploitation of these subtypes to repair the damaged CST after injury and disease.

neuroscience↗

bioRxiv2021

Geometry of neural computation unifies working memory and planning

Real-world tasks require coordination of working memory, decision making, and planning, yet these cognitive functions have disproportionately been studied as independent modular processes in the brain. Here we propose that contingency representations, defined as mappings for how future behaviors depend on upcoming events, can unify working memory and planning computations. We designed a task capable of disambiguating distinct types of representations. Our experiments revealed that human behavior is consistent with contingency representations, and not with traditional sensory models of working memory. In task-optimized recurrent neural networks we investigated possible circuit mechanisms for contingency representations and found that these representations can explain neurophysiological observations from prefrontal cortex during working memory tasks. Finally, we generated falsifiable predictions for neural data to identify contingency representations in neural data and to dissociate different models of working memory. Our findings characterize a neural representational strategy that can unify working memory, planning, and context-dependent decision making.

neuroscience↗

bioRxiv2020

PsychRNN: An Accessible and Flexible Python Package for Training Recurrent Neural Network Models on Cognitive Tasks

Task-trained artificial recurrent neural networks (RNNs) provide a computational modeling framework of increasing interest and application in computational, systems, and cognitive neuroscience. RNNs can be trained, using deep learning methods, to perform cognitive tasks used in animal and human experiments, and can be studied to investigate potential neural representations and circuit mechanisms underlying cognitive computations and behavior. Widespread application of these approaches within neuroscience has been limited by technical barriers in use of deep learning software packages to train network models. Here we introduce PsychRNN, an accessible, flexible, and extensible Python package for training RNNs on cognitive tasks. Our package is designed for accessibility, for researchers to define tasks and train RNN models using only Python and NumPy without requiring knowledge of deep learning software. The training backend is based on TensorFlow and is readily extensible for researchers with TensorFlow knowledge to develop projects with additional customization. PsychRNN implements a number of specialized features to support applications in systems and cognitive neuroscience. Users can impose neurobiologically relevant constraints on synaptic connectivity patterns. Furthermore, specification of cognitive tasks has a modular structure, which facilitates parametric variation of task demands to examine their impact on model solutions. PsychRNN also enables task shaping during training, or curriculum learning, in which tasks are adjusted in closed-loop based on performance. Shaping is ubiquitous in training of animals in cognitive tasks, and PsychRNN allows investigation of how shaping trajectories impact learning and model solutions. Overall, the PsychRNN framework facilitates application of trained RNNs in neuroscience research. Visual Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=100 SRC="FIGDIR/small/321752v1_ufig1.gif" ALT="Figure 1"> View larger version (25K): org.highwire.dtl.DTLVardef@1541513org.highwire.dtl.DTLVardef@133c86borg.highwire.dtl.DTLVardef@aaf099org.highwire.dtl.DTLVardef@99d55b_HPS_FORMAT_FIGEXP M_FIG Example workflow for using PsychRNN. First, the task of interest is defined, and a recurrent neural network model is trained to perform the task, optionally with neurobiologically informed constraints on the network. After the network is trained, the researchers can investigate network properties including the synaptic connectivity patterns and the dynamics of neural population activity during task execution, and other studies, e.g. those on perturbations, can be explored. The dotted line shows the possible repetition of this cycle with one network, which allows investigation of training effects of task shaping, or curriculum learning, for closed-loop training of the network on a progression of tasks. C_FIG Significance StatementArtificial recurrent neural network (RNN) modeling is of increasing interest within computational, systems, and cognitive neuroscience, yet its proliferation as a computational tool within the field has been limited due to technical barriers in use of specialized deep-learning software. PsychRNN provides an accessible, flexible, and powerful framework for training RNN models on cognitive tasks. Users can define tasks and train models using the Python-based interface which enables RNN modeling studies without requiring user knowledge of deep learning software. PsychRNNs modular structure facilitates task specification and incorporation of neurobiological constraints, and supports extensibility for users with deep learning expertise. PsychRNNs framework for RNN modeling will increase accessibility and reproducibility of this approach across neuroscience subfields.

neuroscience↗