Parallel probability density approximation

Probability density approximation (PDA) is a nonparametric method of calculating probability densities. When integrated into Bayesian estimation, it allows researchers to fit psychological processes for which analytic probability functions are unavailable, significantly expanding the scope of theories that can be quantitatively tested. PDA is, however, computationally intensive, requiring large numbers of Monte Carlo simulations in order to attain good precision. We introduce Parallel PDA (pPDA), a highly efficient implementation of this method utilizing the Armadillo C++ and CUDA C libraries to conduct millions of model simulations simultaneously in graphics processing units (GPUs). This approach provides a practical solution for rapidly approximating probability densities with high precision. In addition to demonstrating this method, we fit a piecewise linear ballistic accumulator model (Holmes, Trueblood, & Heathcote, 2016) to empirical data. Finally, we conducted simulation studies to investigate various issues associated with PDA and provide guidelines for pPDA applications to other complex cognitive models.

[1]  C. Robert.  Simulation of truncated normal variables , 2009, 0907.4010.

[2]  Cajo J. F. ter Braak,et al.  A Markov Chain Monte Carlo version of the genetic algorithm Differential Evolution: easy Bayesian computing for real parameter spaces , 2006, Stat. Comput..

[3]  Jennifer S Trueblood,et al.  Bayesian analysis of the piecewise diffusion decision model , 2018, Behavior research methods.

[4]  E. Parzen.  On Estimation of a Probability Density Function and Mode , 1962 .

[5]  Brandon M. Turner,et al.  Likelihood-Free Methods for Cognitive Science , 2018 .

[6]  John K Kruschke,et al.  Bayesian data analysis. , 2010, Wiley interdisciplinary reviews. Cognitive science.

[7]  W. Schwarz.  ON THE CONVOLUTION OF INVERSE GAUSSIAN AND EXPONENTIAL RANDOM VARIABLES , 2002 .

[8]  Bo Hu,et al.  Distributed evolutionary Monte Carlo for Bayesian computing , 2010, Comput. Stat. Data Anal..

[9]  James L. McClelland.  On the time relations of mental processes: An examination of systems of processes in cascade. , 1979 .

[10]  James L. McClelland,et al.  Loss aversion and inhibition in dynamical models of multialternative choice. , 2004, Psychological review.

[11]  Andrew Gelman,et al.  The No-U-turn sampler: adaptively setting path lengths in Hamiltonian Monte Carlo , 2011, J. Mach. Learn. Res..

[12]  William R. Holmes,et al.  The impact of speed and bias on the cognitive processes of experts and novices in medical image decision-making , 2017, Cognitive Research: Principles and Implications.

[13]  Andrew Heathcote,et al.  A new framework for modeling decisions about changing information: The Piecewise Linear Ballistic Accumulator model , 2016, Cognitive Psychology.

[14]  Andrew Heathcote,et al.  Dynamic models of choice , 2018, Behavior Research Methods.

[15]  Brandon M. Turner,et al.  A generalized, likelihood-free method for posterior estimation , 2014, Psychonomic bulletin & review.

[16]  P. J. Green,et al.  Density Estimation for Statistics and Data Analysis , 1987 .

[17]  Bradley C. Love,et al.  Learning in Noise: Dynamic Decision-Making in a Variable Environment. , 2009, Journal of mathematical psychology.

[18]  M. Dawson,et al.  Fitting the ex-Gaussian equation to reaction time distributions , 1988 .

[19]  Marius Usher,et al.  Testing Multi-Alternative Decision Models with Non-Stationary Evidence , 2011, Front. Neurosci..

[20]  Philip L. Smith,et al.  Diffusion theory of decision making in continuous report. , 2016, Psychological review.

[21]  Conrad Sanderson,et al.  Armadillo: a template-based C++ library for linear algebra , 2016, J. Open Source Softw..

[22]  Andrew Heathcote,et al.  Fitting Wald and ex-Wald distributions to response time data: An example using functions for the S-PLUS package , 2004, Behavior research methods, instruments, & computers : a journal of the Psychonomic Society, Inc.

[23]  Thomas V. Wiecki,et al.  The Quality of Response Time Data Inference: A Blinded, Collaborative Assessment of the Validity of Cognitive Models , 2018, Psychonomic Bulletin & Review.

[24]  Brandon M. Turner,et al.  A method for efficiently sampling from distributions with correlated dimensions. , 2013, Psychological methods.

[25]  R. Plevin,et al.  Approximate Bayesian Computation in Evolution and Ecology , 2011 .

[26]  James L. McClelland,et al.  The time course of perceptual choice: the leaky, competing accumulator model. , 2001, Psychological review.

[27]  J. Rosenthal,et al.  Optimal scaling for various Metropolis-Hastings algorithms , 2001 .

[28]  Brandon M. Turner,et al.  Approximate Bayesian computation with differential evolution , 2012 .

[29]  Max Grossman,et al.  Professional CUDA C Programming , 2014 .

[30]  R. Sekuler,et al.  A specific and enduring improvement in visual motion discrimination. , 1982, Science.

[31]  Radford M. Neal.  An improved acceptance procedure for the hybrid Monte Carlo algorithm , 1992, hep-lat/9208011.

[32]  A. Goldenshluger,et al.  Bandwidth selection in kernel density estimation: Oracle inequalities and adaptive minimax optimality , 2010, 1009.1016.

[33]  E. Wagenmakers,et al.  Psychological interpretation of the ex-Gaussian and shifted Wald parameters: A diffusion model analysis , 2009, Psychonomic bulletin & review.

[34]  William R. Holmes,et al.  A practical guide to the Probability Density Approximation (PDA) with improved implementation and error characterization , 2015 .

[35]  Andrew Heathcote,et al.  An introduction to good practices in cognitive modeling , 2015 .

[36]  T. Zandt,et al.  How to fit a response time distribution , 2000, Psychonomic bulletin & review.

[37]  Birte U. Forstmann,et al.  Parameter recovery for the Leaky Competing Accumulator model , 2017 .

[38]  R. Hohle.  INFERRED COMPONENTS OF REACTION TIMES AS FUNCTIONS OF FOREPERIOD DURATION. , 1965, Journal of experimental psychology.

[39]  Kam-Wah Tsui,et al.  Distributed Evolutionary Monte Carlo with Applications to Bayesian Analysis , 2005 .

[40]  Scott D. Brown,et al.  The simplest complete model of choice response time: Linear ballistic accumulation , 2008, Cognitive Psychology.

[41]  Roger Ratcliff,et al.  The Diffusion Decision Model: Theory and Data for Two-Choice Decision Tasks , 2008, Neural Computation.

[42]  B. Silverman,et al.  Kernel Density Estimation Using the Fast Fourier Transform , 1982 .

[43]  B. Silverman,et al.  Algorithm AS 176: Kernel Density Estimation Using the Fast Fourier Transform , 1982 .

[44]  Bradley P. Carlin,et al.  Bayesian measures of model complexity and fit , 2002 .

[45]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[46]  P. Cisek,et al.  Decisions in Changing Conditions: The Urgency-Gating Model , 2009, The Journal of Neuroscience.

[47]  Paul Cisek,et al.  Decision making by urgency gating: theory and experimental support. , 2012, Journal of neurophysiology.

[48]  S. Sisson,et al.  Likelihood-free Markov chain Monte Carlo , 2010, 1001.2058.

[49]  H. Kile,et al.  Bandwidth Selection in Kernel Density Estimation , 2010 .

[50]  Andrew Gelman,et al.  Handbook of Markov Chain Monte Carlo , 2011 .

[51]  Roger Ratcliff,et al.  A Theory of Memory Retrieval. , 1978 .

[52]  Francis Tuerlinckx,et al.  Efficient simulation of diffusion-based choice RT models on CPU and GPU , 2016, Behavior research methods.