Toward Goal-Driven Neural Network Models for the Rodent Whisker-Trigeminal System

In large part, rodents see the world through their whiskers, a powerful tactile sense enabled by a series of brain areas that form the whisker-trigeminal system. Raw sensory data arrives in the form of mechanical input to the exquisitely sensitive, actively-controllable whisker array, and is processed through a sequence of neural circuits, eventually arriving in cortical regions that communicate with decision-making and memory areas. Although a long history of experimental studies has characterized many aspects of these processing stages, the computational operations of the whisker-trigeminal system remain largely unknown. In the present work, we take a goal-driven deep neural network (DNN) approach to modeling these computations. First, we construct a biophysically-realistic model of the rat whisker array. We then generate a large dataset of whisker sweeps across a wide variety of 3D objects in highly-varying poses, angles, and speeds. Next, we train DNNs from several distinct architectural families to solve a shape recognition task in this dataset. Each architectural family represents a structurally-distinct hypothesis for processing in the whisker-trigeminal system, corresponding to different ways in which spatial and temporal information can be integrated. We find that most networks perform poorly on the challenging shape recognition task, but that specific architectures from several families can achieve reasonable performance levels. Finally, we show that Representational Dissimilarity Matrices (RDMs), a tool for comparing population codes between neural systems, can separate these higher-performing networks with data of a type that could plausibly be collected in a neurophysiological or imaging experiment. Our results are a proof-of-concept that goal-driven DNN networks of the whisker-trigeminal system are potentially within reach.

[1]  David P. Friedman,et al.  Physiological evidence for serial processing in somatosensory cortex. , 1987, Science.

[2]  D. J. Felleman,et al.  Distributed hierarchical processing in the primate cerebral cortex. , 1991, Cerebral cortex.

[3]  M. Goodale,et al.  Separate visual pathways for perception and action , 1992, Trends in Neurosciences.

[4]  M. Armstrong‐James,et al.  Flow of excitation within rat barrel cortex on striking a single vibrissa. , 1992, Journal of neurophysiology.

[5]  George A. Miller,et al.  WordNet: A Lexical Database for English , 1995, HLT.

[6]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[7]  Patrick J. F. Groenen,et al.  Modern Multidimensional Scaling: Theory and Applications , 2003 .

[8]  Y. Iwamura Hierarchical somatosensory processing , 1998, Current Opinion in Neurobiology.

[9]  F. Rice,et al.  Similarities and differences in the innervation of mystacial vibrissal follicle–sinus complexes in the rat and cat: A confocal microscopic study , 2002, The Journal of comparative neurology.

[10]  A. Grinvald,et al.  Spatiotemporal Dynamics of Sensory Responses in Layer 2/3 of Rat Barrel Cortex Measured In Vivo by Voltage-Sensitive Dye Imaging Combined with Whole-Cell Voltage Recordings and Neuron Reconstructions , 2003, The Journal of Neuroscience.

[11]  R. Kakigi,et al.  Serial processing in the human somatosensory system. , 2004, Cerebral cortex.

[12]  M. Diamond,et al.  Neuronal Encoding of Texture in the Whisker Sensory Pathway , 2005, PLoS biology.

[13]  Per Magne Knutsen,et al.  Haptic Object Localization in the Vibrissal System: Behavior and Performance , 2006, The Journal of Neuroscience.

[14]  E. Ahissar,et al.  Parallel Thalamic Pathways for Whisking and Touch Signals in the Rat , 2006, PLoS biology.

[15]  David S. Greenberg,et al.  Spatial Organization of Neuronal Population Responses in Layer 2/3 of Rat Barrel Cortex , 2007, The Journal of Neuroscience.

[16]  M. Diamond,et al.  Neuronal Activity in Rat Barrel Cortex Underlying Texture Discrimination , 2007, PLoS biology.

[17]  Delbert Dueck,et al.  Clustering by Passing Messages Between Data Points , 2007, Science.

[18]  D. Kleinfeld,et al.  'Where' and 'what' in the whisker sensorimotor system , 2008, Nature Reviews Neuroscience.

[19]  Nikolaus Kriegeskorte,et al.  Frontiers in Systems Neuroscience Systems Neuroscience , 2022 .

[20]  Faouzi Ghorbel,et al.  A simple and efficient approach for 3D mesh approximate convex decomposition , 2009, 2009 16th IEEE International Conference on Image Processing (ICIP).

[21]  Martin Deschenes,et al.  Vibrissal afferents from trigeminus to cortices , 2009, Scholarpedia.

[22]  K. Svoboda,et al.  Neural Activity in Barrel Cortex Underlying Vibrissa-Based Object Localization in Mice , 2010, Neuron.

[23]  Yoram Singer,et al.  Adaptive Subgradient Methods for Online Learning and Stochastic Optimization , 2011, J. Mach. Learn. Res..

[24]  Yoshua Bengio,et al.  Understanding the difficulty of training deep feedforward neural networks , 2010, AISTATS.

[25]  Joseph H. Solomon,et al.  The Morphology of the Rat Vibrissal Array: A Model for Quantifying Spatiotemporal Patterns of Whisker-Object Contact , 2011, PLoS Comput. Biol..

[26]  Cullen B. Owens,et al.  Anatomical Pathways Involved in Generating and Sensing Rhythmic Whisker Movements , 2011, Front. Integr. Neurosci..

[27]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[28]  David D. Cox,et al.  Hyperopt: A Python Library for Optimizing the Hyperparameters of Machine Learning Algorithms , 2013, SciPy.

[29]  Lucie A. Huet,et al.  Modeling Forces and Moments at the Base of a Rat Vibrissa during Noncontact Whisking and Whisking against an Object , 2014, The Journal of Neuroscience.

[30]  Yoshua Bengio,et al.  On the Properties of Neural Machine Translation: Encoder–Decoder Approaches , 2014, SSST@EMNLP.

[31]  Ha Hong,et al.  Performance-optimized hierarchical models predict neural responses in higher visual cortex , 2014, Proceedings of the National Academy of Sciences.

[32]  Stefan Carlsson,et al.  CNN Features Off-the-Shelf: An Astounding Baseline for Recognition , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops.

[33]  Daniel L. K. Yamins,et al.  Deep Neural Networks Rival the Representation of Primate IT Cortex for Core Visual Object Recognition , 2014, PLoS Comput. Biol..

[34]  Nikolaus Kriegeskorte,et al.  Deep Supervised, but Not Unsupervised, Models May Explain IT Cortical Representation , 2014, PLoS Comput. Biol..

[35]  D. Kleinfeld,et al.  Vibrissa Self-Motion and Touch Are Reliably Encoded along the Same Somatosensory Pathway from Brainstem through Thalamus , 2015, PLoS biology.

[36]  Mitra Hartmann Vibrissa mechanical properties , 2015, Scholarpedia.

[37]  Leonidas J. Guibas,et al.  ShapeNet: An Information-Rich 3D Model Repository , 2015, ArXiv.

[38]  J. DiCarlo,et al.  Using goal-driven deep learning models to understand sensory cortex , 2016, Nature Neuroscience.

[39]  Lucie A. Huet,et al.  Simulations of a Vibrissa Slipping along a Straight Edge and an Analysis of Frictional Effects during Whisking , 2016, IEEE Transactions on Haptics.

[40]  M. Hartmann,et al.  Spatiotemporal Patterns of Contact Across the Rat Vibrissal Array During Exploratory Behavior , 2016, Front. Behav. Neurosci..