How much to trust the senses: likelihood learning.

Our brain often needs to estimate unknown variables from imperfect information. Our knowledge about the statistical distributions of quantities in our environment (called priors) and currently available information from sensory inputs (called likelihood) are the basis of all Bayesian models of perception and action. While we know that priors are learned, most studies of prior-likelihood integration simply assume that subjects know about the likelihood. However, as the quality of sensory inputs change over time, we also need to learn about new likelihoods. Here, we show that human subjects readily learn the distribution of visual cues (likelihood function) in a way that can be predicted by models of statistically optimal learning. Using a likelihood that depended on color context, we found that a learned likelihood generalized to new priors. Thus, we conclude that subjects learn about likelihood.

[1]  Konrad Paul Kording,et al.  Learning Priors for Bayesian Computations in the Nervous System , 2010, PloS one.

[2]  A. Yuille,et al.  Object perception as Bayesian inference. , 2004, Annual review of psychology.

[3]  S. Kitazawa,et al.  Bayesian Calibration of Simultaneity in Audiovisual Temporal Order Judgments , 2012, PloS one.

[4]  Eli Brenner,et al.  Reweighting visual cues by touch. , 2011, Journal of vision.

[5]  M. Kawato,et al.  Acquisition and contextual switching of multiple internal models for different viscous force fields , 2003, Neuroscience Research.

[6]  M. Miyazaki,et al.  Testing Bayesian models of human coincidence timing. , 2005, Journal of neurophysiology.

[7]  Heinrich H. Bülthoff,et al.  Touch can change visual slant perception , 2000, Nature Neuroscience.

[8]  D. Burr,et al.  The Ventriloquist Effect Results from Near-Optimal Bimodal Integration , 2004, Current Biology.

[9]  Kazuyuki Aihara,et al.  A Bayesian Model of Sensory Adaptation , 2011, PloS one.

[10]  M. Ernst,et al.  Humans integrate visual and haptic information in a statistically optimal fashion , 2002, Nature.

[11]  Denis G. Pelli,et al.  ECVP '07 Abstracts , 2007, Perception.

[12]  Makoto Miyazaki,et al.  Acquisition of Multiple Prior Distributions in Tactile Temporal Order Judgment , 2012, Front. Psychology.

[13]  Michael S Landy,et al.  Combining Priors and Noisy Visual Cues in a Rapid Pointing Task , 2006, The Journal of Neuroscience.

[14]  M. Ernst,et al.  The statistical determinants of adaptation rate in human reaching. , 2008, Journal of vision.

[15]  Konrad Paul Kording,et al.  Sensory Cue Integration , 2011 .

[16]  David R. Wozny,et al.  Computational Characterization of Visually Induced Auditory Spatial Adaptation , 2011, Front. Integr. Neurosci..

[17]  Kazuyuki Aihara,et al.  Bayesian Inference Explains Perception of Unity and Ventriloquism Aftereffect: Identification of Common Sources of Audiovisual Stimuli , 2007, Neural Computation.

[18]  Peter W Battaglia,et al.  Humans Trade Off Viewing Time and Movement Duration to Improve Visuomotor Accuracy in a Fast Reaching Task , 2007, The Journal of Neuroscience.

[19]  Hugo L. Fernandes,et al.  Differential Representations of Prior and Likelihood Uncertainty in the Human Brain , 2012, Current Biology.

[20]  Anthony M. Zador,et al.  Asymmetric Dynamics in Optimal Variance Adaptation , 1998, Neural Computation.

[21]  Konrad Paul Kording,et al.  Bayesian integration in sensorimotor learning , 2004, Nature.

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

[23]  M. Banks,et al.  Visual–Haptic Adaptation Is Determined by Relative Reliability , 2010, The Journal of Neuroscience.

[24]  Michael N. Shadlen,et al.  Temporal context calibrates interval timing , 2010, Nature Neuroscience.

[25]  David C Knill,et al.  Adapting internal statistical models for interpreting visual cues to depth. , 2010, Journal of vision.

[26]  R C Miall,et al.  System Identification Applied to a Visuomotor Task: Near-Optimal Human Performance in a Noisy Changing Task , 2003, The Journal of Neuroscience.

[27]  J. Kalaska,et al.  Colored context cues can facilitate the ability to learn and to switch between multiple dynamical force fields. , 2011, Journal of neurophysiology.

[28]  Max Berniker,et al.  Bayesian approaches to sensory integration for motor control. , 2011, Wiley interdisciplinary reviews. Cognitive science.

[29]  Alan A. Stocker,et al.  Sensory Adaptation within a Bayesian Framework for Perception , 2005, NIPS.

[30]  R. Jacobs,et al.  Experience-dependent integration of texture and motion cues to depth , 1999, Vision Research.

[31]  M. Kawato,et al.  Random presentation enables subjects to adapt to two opposing forces on the hand , 2004, Nature Neuroscience.

[32]  Konrad P. Körding,et al.  Uncertainty of Feedback and State Estimation Determines the Speed of Motor Adaptation , 2009, Front. Comput. Neurosci..

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

[34]  Eli Brenner,et al.  Speed judgments of three-dimensional motion incorporate extraretinal information. , 2011, Journal of vision.