Modeling correlated noise is necessary to decode uncertainty

ABSTRACT Brain decoding algorithms form an important part of the arsenal of analysis tools available to neuroscientists, allowing for a more detailed study of the kind of information represented in patterns of cortical activity. While most current decoding algorithms focus on estimating a single, most likely stimulus from the pattern of noisy fMRI responses, the presence of noise causes this estimate to be uncertain. This uncertainty in stimulus estimates is a potentially highly relevant aspect of cortical stimulus processing, and features prominently in Bayesian or probabilistic models of neural coding. Here, we focus on sensory uncertainty and how best to extract this information with fMRI. We first demonstrate in simulations that decoding algorithms that take into account correlated noise between fMRI voxels better recover the amount of uncertainty (quantified as the width of a probability distribution over possible stimuli) associated with the decoded estimate. Furthermore, we show that not all correlated variability should be treated equally, as modeling tuning‐dependent correlations has the greatest impact on decoding performance. Next, we examine actual noise correlations in human visual cortex, and find that shared variability in areas V1‐V3 depends on the tuning properties of fMRI voxels. In line with our simulations, accounting for this shared noise between similarly tuned voxels produces important benefits in decoding. Our findings underscore the importance of accurate noise models in fMRI decoding approaches, and suggest a statistically feasible method to incorporate the most relevant forms of shared noise. HIGHLIGHTSInformation in a cortical response is best characterized by a probability distribution.Accurate decoding of this distribution requires accounting for correlated noise.Correlated noise is most relevant when it mimics a stimulus‐driven response.Including this noise in fMRI forward models enables the estimation of cortical uncertainty.

[1]  Sabine Kastner,et al.  Widespread correlation patterns of fMRI signal across visual cortex reflect eccentricity organization , 2015, eLife.

[2]  Janneke F. M. Jehee,et al.  Sensory uncertainty decoded from visual cortex predicts behavior , 2015, Nature Neuroscience.

[3]  Peter Dayan,et al.  The Effect of Correlated Variability on the Accuracy of a Population Code , 1999, Neural Computation.

[4]  M. A. Smith,et al.  Spatial and Temporal Scales of Neuronal Correlation in Primary Visual Cortex , 2008, The Journal of Neuroscience.

[5]  Christian K. Machens,et al.  Variability in neural activity and behavior , 2014, Current Opinion in Neurobiology.

[6]  Matthew F Nolan,et al.  Grid cells' need for speed , 2016, Nature Neuroscience.

[7]  Ehud Zohary,et al.  Correlated neuronal discharge rate and its implications for psychophysical performance , 1994, Nature.

[8]  Wei Ji Ma,et al.  Probabilistic brains: knowns and , 2013 .

[9]  E. DeYoe,et al.  Mapping striate and extrastriate visual areas in human cerebral cortex. , 1996, Proceedings of the National Academy of Sciences of the United States of America.

[10]  Peter van Gelderen,et al.  Reducing correlated noise in fMRI data , 2008, Magnetic resonance in medicine.

[11]  Ludmila I. Kuncheva,et al.  On the optimality of Naïve Bayes with dependent binary features , 2006, Pattern Recognit. Lett..

[12]  J. Gallant,et al.  Identifying natural images from human brain activity , 2008, Nature.

[13]  Kevin Murphy,et al.  Resting-state fMRI confounds and cleanup , 2013, NeuroImage.

[14]  D H Brainard,et al.  The Psychophysics Toolbox. , 1997, Spatial vision.

[15]  Daeyeol Lee,et al.  Effects of noise correlations on information encoding and decoding. , 2006, Journal of neurophysiology.

[16]  Nikolaus Kriegeskorte,et al.  Visual representations are dominated by intrinsic fluctuations correlated between areas , 2015, NeuroImage.

[17]  A. Pouget,et al.  Probabilistic brains: knowns and unknowns , 2013, Nature Neuroscience.

[18]  Harry Zhang,et al.  The Optimality of Naive Bayes , 2004, FLAIRS.

[19]  Wei Ji Ma,et al.  Neural coding of uncertainty and probability. , 2014, Annual review of neuroscience.

[20]  G. Glover,et al.  Retinotopic organization in human visual cortex and the spatial precision of functional MRI. , 1997, Cerebral cortex.

[21]  P. Berkes,et al.  Statistically Optimal Perception and Learning: from Behavior to Neural Representations , 2022 .

[22]  Jeremy Freeman,et al.  Orientation Decoding Depends on Maps, Not Columns , 2011, The Journal of Neuroscience.

[23]  Si Wu,et al.  Population Coding with Correlation and an Unfaithful Model , 2001, Neural Computation.

[24]  A. Dean.  The variability of discharge of simple cells in the cat striate cortex , 2004, Experimental Brain Research.

[25]  Michael Brady,et al.  Improved Optimization for the Robust and Accurate Linear Registration and Motion Correction of Brain Images , 2002, NeuroImage.

[26]  Yevgeniy B. Sirotin,et al.  Stimulus-Related Neuroimaging in Task-Engaged Subjects Is Best Predicted by Concurrent Spiking , 2014, The Journal of Neuroscience.

[27]  Konrad Kording,et al.  Annals of the New York Academy of Sciences Bayesian Models: the Structure of the World, Uncertainty, Behavior, and the Brain , 2022 .

[28]  Miranda Scolari,et al.  Estimating the influence of attention on population codes in human visual cortex using voxel-based tuning functions , 2009, NeuroImage.

[29]  D. Heeger,et al.  Cross-orientation suppression in human visual cortex. , 2011, Journal of neurophysiology.

[30]  C. D. De Dreu,et al.  Oxytocin-Motivated Ally Selection is Moderated by Fetal Testosterone Exposure and Empathic Concern , 2012, Front. Neurosci..

[31]  D. Knill,et al.  The Bayesian brain: the role of uncertainty in neural coding and computation , 2004, Trends in Neurosciences.

[32]  Rich Caruana,et al.  Predicting good probabilities with supervised learning , 2005, ICML.

[33]  D. Heeger,et al.  Decoding and Reconstructing Color from Responses in Human Visual Cortex , 2009, The Journal of Neuroscience.

[34]  Frank Tong,et al.  Perceptual Learning Selectively Refines Orientation Representations in Early Visual Cortex , 2012, The Journal of Neuroscience.

[35]  M. A. Smith,et al.  The spatial structure of correlated neuronal variability , 2016, Nature Neuroscience.

[36]  J W Belliveau,et al.  Borders of multiple visual areas in humans revealed by functional magnetic resonance imaging. , 1995, Science.

[37]  B. Finlay,et al.  Short-term response variability of monkey striate neurons , 1976, Brain Research.

[38]  Edward Awh,et al.  A Neural Measure of Precision in Visual Working Memory , 2013, Journal of Cognitive Neuroscience.

[39]  L. Pinneo.  On noise in the nervous system. , 1966, Psychological review.

[40]  Pedro M. Domingos,et al.  On the Optimality of the Simple Bayesian Classifier under Zero-One Loss , 1997, Machine Learning.

[41]  N. Logothetis,et al.  Neurophysiological investigation of the basis of the fMRI signal , 2001, Nature.

[42]  Jascha D. Swisher,et al.  Multiscale Pattern Analysis of Orientation-Selective Activity in the Primary Visual Cortex , 2010, The Journal of Neuroscience.

[43]  Damien J. Mannion,et al.  Orientation anisotropies in human visual cortex. , 2010, Journal of neurophysiology.

[44]  Wim Vanduffel,et al.  The Radial Bias: A Different Slant on Visual Orientation Sensitivity in Human and Nonhuman Primates , 2006, Neuron.

[45]  Roel H. R. Deckers,et al.  Quantifying the spatial resolution of the gradient echo and spin echo BOLD response at 3 Tesla , 2005, Magnetic resonance in medicine.

[46]  Bryon A. Mueller,et al.  A Survey of the Sources of Noise in fMRI , 2012, Psychometrika.

[47]  A. Pouget,et al.  Neural correlations, population coding and computation , 2006, Nature Reviews Neuroscience.

[48]  A. Dale,et al.  Cortical Surface-Based Analysis II: Inflation, Flattening, and a Surface-Based Coordinate System , 1999, NeuroImage.

[49]  Abraham Z. Snyder,et al.  Spurious but systematic correlations in functional connectivity MRI networks arise from subject motion , 2012, NeuroImage.

[50]  A. Pouget,et al.  Information-limiting correlations , 2014, Nature Neuroscience.

[51]  D G Pelli,et al.  The VideoToolbox software for visual psychophysics: transforming numbers into movies. , 1997, Spatial vision.

[52]  W. Bair,et al.  Correlated Firing in Macaque Visual Area MT: Time Scales and Relationship to Behavior , 2001, The Journal of Neuroscience.

[53]  Jonathan Winawer,et al.  GLMdenoise: a fast, automated technique for denoising task-based fMRI data , 2013, Front. Neurosci..

[54]  A. F. Adams,et al.  The Survey , 2021, Dyslexia in Higher Education.