Bayesian population receptive field modelling

ABSTRACT We introduce a probabilistic (Bayesian) framework and associated software toolbox for mapping population receptive fields (pRFs) based on fMRI data. This generic approach is intended to work with stimuli of any dimension and is demonstrated and validated in the context of 2D retinotopic mapping. The framework enables the experimenter to specify generative (encoding) models of fMRI timeseries, in which experimental stimuli enter a pRF model of neural activity, which in turns drives a nonlinear model of neurovascular coupling and Blood Oxygenation Level Dependent (BOLD) response. The neuronal and haemodynamic parameters are estimated together on a voxel‐by‐voxel or region‐of‐interest basis using a Bayesian estimation algorithm (variational Laplace). This offers several novel contributions to receptive field modelling. The variance/covariance of parameters are estimated, enabling receptive fields to be plotted while properly representing uncertainty about pRF size and location. Variability in the haemodynamic response across the brain is accounted for. Furthermore, the framework introduces formal hypothesis testing to pRF analysis, enabling competing models to be evaluated based on their log model evidence (approximated by the variational free energy), which represents the optimal tradeoff between accuracy and complexity. Using simulations and empirical data, we found that parameters typically used to represent pRF size and neuronal scaling are strongly correlated, which is taken into account by the Bayesian methods we describe when making inferences. We used the framework to compare the evidence for six variants of pRF model using 7 T functional MRI data and we found a circular Difference of Gaussians (DoG) model to be the best explanation for our data overall. We hope this framework will prove useful for mapping stimulus spaces with any number of dimensions onto the anatomy of the brain. HIGHLIGHTSWe introduce a Bayesian toolbox for population receptive field (pRF) mapping.Neuronal and haemodynamic parameters are estimated per voxel or per region.Hypotheses can be tested by comparing pRF models based on their evidence.The uncertainty over parameters (such as pRF size) is estimated and visualised.We establish face validity using simulations and test‐rest reliability with 7 T fMRI.

[1]  Karl J. Friston,et al.  Physiologically informed dynamic causal modeling of fMRI data , 2015, NeuroImage.

[2]  William D. Penny,et al.  Comparing Dynamic Causal Models using AIC, BIC and Free Energy , 2012, NeuroImage.

[3]  A. Schleicher,et al.  Cytoarchitectonic mapping of the human dorsal extrastriate cortex , 2012, Brain Structure and Function.

[4]  Brian A. Wandell,et al.  Population receptive field estimates in human visual cortex , 2008, NeuroImage.

[5]  Timothy E. J. Behrens,et al.  Organizing conceptual knowledge in humans with a gridlike code , 2016, Science.

[6]  Simon B. Eickhoff,et al.  A new SPM toolbox for combining probabilistic cytoarchitectonic maps and functional imaging data , 2005, NeuroImage.

[7]  Klaas E. Stephan,et al.  A hemodynamic model for layered BOLD signals , 2016, NeuroImage.

[8]  Karl J. Friston,et al.  Neural masses and fields in dynamic causal modeling , 2013, Front. Comput. Neurosci..

[9]  Karl J. Friston,et al.  Post hoc Bayesian model selection , 2011, NeuroImage.

[10]  Karl J. Friston,et al.  Variational free energy and the Laplace approximation , 2007, NeuroImage.

[11]  Karl J. Friston,et al.  Bayesian model selection for group studies , 2009, NeuroImage.

[12]  Fang Fang,et al.  Position shifts of fMRI-based population receptive fields in human visual cortex induced by Ponzo illusion , 2015, Experimental Brain Research.

[13]  B. Wandell,et al.  Cortical Maps and White Matter Tracts following Long Period of Visual Deprivation and Retinal Image Restoration , 2010, Neuron.

[14]  J. Hennig,et al.  The Processing of First- and Second-Order Motion in Human Visual Cortex Assessed by Functional Magnetic Resonance Imaging (fMRI) , 1998, The Journal of Neuroscience.

[15]  D. M. Broussard,et al.  The bidirectionality of motor learning in the vestibulo-ocular reflex is a function of cerebellar mGluR1 receptors. , 2010, Journal of neurophysiology.

[16]  Karl J. Friston,et al.  Bayesian model selection for group studies — Revisited , 2014, NeuroImage.

[17]  A. Cowey,et al.  Preferential representation of the fovea in the primary visual cortex , 1993, Nature.

[18]  E. S. Pearson,et al.  On the Problem of the Most Efficient Tests of Statistical Hypotheses , 1933 .

[19]  Bruce L. McNaughton,et al.  Representation of Three-Dimensional Space in the Hippocampus of Flying Bats , 2013 .

[20]  M. Merzenich,et al.  Representation of the cochlear partition of the superior temporal plane of the macaque monkey. , 1973, Brain research.

[21]  Karl J. Friston,et al.  Comparing hemodynamic models with DCM , 2007, NeuroImage.

[22]  R. W. Rodieck.  Quantitative analysis of cat retinal ganglion cell response to visual stimuli. , 1965, Vision research.

[23]  Karl J. Friston,et al.  Gradient-based MCMC samplers for dynamic causal modelling , 2016, NeuroImage.

[24]  Eduardo A. Aponte,et al.  mpdcm: A toolbox for massively parallel dynamic causal modeling , 2016, Journal of Neuroscience Methods.

[25]  G. Holmes.  Ferrier Lecture - The organization of the visual cortex in man , 1945, Proceedings of the Royal Society of London. Series B - Biological Sciences.

[26]  Karl J. Friston,et al.  Nonlinear event‐related responses in fMRI , 1998, Magnetic resonance in medicine.

[27]  E. S. Pearson,et al.  On the Problem of the Most Efficient Tests of Statistical Hypotheses , 1933 .

[28]  R. Buxton,et al.  Modeling the hemodynamic response to brain activation , 2004, NeuroImage.

[29]  Elia Formisano,et al.  Processing of Natural Sounds in Human Auditory Cortex: Tonotopy, Spectral Tuning, and Relation to Voice Sensitivity , 2012, The Journal of Neuroscience.

[30]  Karl J. Friston,et al.  Efficient gradient computation for dynamical models , 2014, NeuroImage.

[31]  S. Dumoulin,et al.  Modeling center-surround configurations in population receptive fields using fMRI. , 2012, Journal of vision.

[32]  Karl J. Friston,et al.  A Metropolis–Hastings algorithm for dynamic causal models , 2007, NeuroImage.

[33]  J. Changeux.  The Ferrier Lecture 1998 The molecular biology of consciousness investigated with genetically modified mice , 2006, Philosophical Transactions of the Royal Society B: Biological Sciences.

[34]  D. Hubel,et al.  Receptive fields of single neurones in the cat's striate cortex , 1959, The Journal of physiology.

[35]  Arto Luoma,et al.  Bayesian Model Selection , 2016 .

[36]  Jonathan Winawer,et al.  Computational neuroimaging and population receptive fields , 2015, Trends in Cognitive Sciences.

[37]  K. Amunts,et al.  Brodmann's Areas 17 and 18 Brought into Stereotaxic Space—Where and How Variable? , 2000, NeuroImage.

[38]  B. Robertson.  A celebration of the 50th anniversary of David Hubel and Torsten Wiesel's Receptive fields of single neurones in the cat's striate cortex , 2009, The Journal of physiology.

[39]  Dwight J. Kravitz,et al.  A Retinotopic Basis for the Division of High-Level Scene Processing between Lateral and Ventral Human Occipitotemporal Cortex , 2015, The Journal of Neuroscience.

[40]  Karl J. Friston,et al.  Dynamic causal modelling , 2003, NeuroImage.

[41]  Karl J. Friston,et al.  Dynamic causal modeling with neural fields , 2012, NeuroImage.

[42]  Karl J. Friston,et al.  Neural fields, spectral responses and lateral connections , 2011, NeuroImage.

[43]  Masahiro Takei,et al.  Human resource development and visualization , 2009, J. Vis..

[44]  Serge O. Dumoulin,et al.  Slice-sampled Bayesian PRF mapping , 2016, bioRxiv.

[45]  D. Heeger,et al.  Two Retinotopic Visual Areas in Human Lateral Occipital Cortex , 2006, The Journal of Neuroscience.

[46]  A. Schleicher,et al.  Ventral visual cortex in humans: Cytoarchitectonic mapping of two extrastriate areas , 2007, Human brain mapping.

[47]  Matthew J. Beal,et al.  The variational Bayesian EM algorithm for incomplete data: with application to scoring graphical model structures , 2003 .

[48]  William D. Penny,et al.  Estimating neural response functions from fMRI , 2014, Front. Neuroinform..

[49]  A. Schleicher,et al.  Cytoarchitecture of the human lateral occipital cortex: mapping of two extrastriate areas hOc4la and hOc4lp , 2016, Brain Structure and Function.

[50]  Karl J. Friston,et al.  Contrast gain control and horizontal interactions in V1: A DCM study , 2014, NeuroImage.