Bayesian just-so stories in psychology and neuroscience. Bowers, J. S. & Davis, C. J. Psychol Bull, 138(3):389–414, 2012. doi abstract bibtex According to Bayesian theories in psychology and neuroscience, minds and brains are (near) optimal in solving a wide range of tasks. We challenge this view and argue that more traditional, non-Bayesian approaches are more promising. We make 3 main arguments. First, we show that the empirical evidence for Bayesian theories in psychology is weak. This weakness relates to the many arbitrary ways that priors, likelihoods, and utility functions can be altered in order to account for the data that are obtained, making the models unfalsifiable. It further relates to the fact that Bayesian theories are rarely better at predicting data compared with alternative (and simpler) non-Bayesian theories. Second, we show that the empirical evidence for Bayesian theories in neuroscience is weaker still. There are impressive mathematical analyses showing how populations of neurons could compute in a Bayesian manner but little or no evidence that they do. Third, we challenge the general scientific approach that characterizes Bayesian theorizing in cognitive science. A common premise is that theories in psychology should largely be constrained by a rational analysis of what the mind ought to do. We question this claim and argue that many of the important constraints come from biological, evolutionary, and processing (algorithmic) considerations that have no adaptive relevance to the problem per se. In our view, these factors have contributed to the development of many Bayesian "just so" stories in psychology and neuroscience; that is, mathematical analyses of cognition that can be used to explain almost any behavior as optimal.
@Article{Bowers2012,
author = {Bowers, Jeffrey S. and Davis, Colin J.},
journal = {Psychol Bull},
title = {Bayesian just-so stories in psychology and neuroscience.},
year = {2012},
number = {3},
pages = {389--414},
volume = {138},
abstract = {According to Bayesian theories in psychology and neuroscience, minds
and brains are (near) optimal in solving a wide range of tasks. We
challenge this view and argue that more traditional, non-Bayesian
approaches are more promising. We make 3 main arguments. First, we
show that the empirical evidence for Bayesian theories in psychology
is weak. This weakness relates to the many arbitrary ways that priors,
likelihoods, and utility functions can be altered in order to account
for the data that are obtained, making the models unfalsifiable.
It further relates to the fact that Bayesian theories are rarely
better at predicting data compared with alternative (and simpler)
non-Bayesian theories. Second, we show that the empirical evidence
for Bayesian theories in neuroscience is weaker still. There are
impressive mathematical analyses showing how populations of neurons
could compute in a Bayesian manner but little or no evidence that
they do. Third, we challenge the general scientific approach that
characterizes Bayesian theorizing in cognitive science. A common
premise is that theories in psychology should largely be constrained
by a rational analysis of what the mind ought to do. We question
this claim and argue that many of the important constraints come
from biological, evolutionary, and processing (algorithmic) considerations
that have no adaptive relevance to the problem per se. In our view,
these factors have contributed to the development of many Bayesian
"just so" stories in psychology and neuroscience; that is, mathematical
analyses of cognition that can be used to explain almost any behavior
as optimal.},
doi = {10.1037/a0026450},
keywords = {Bayes Theorem; Biological Evolution; Brain, physiology; Humans; Mental Processes, physiology; Models, Psychological; Motor Skills, physiology; Neurosciences; Psychological Theory; Psychology},
language = {eng},
medline-pst = {ppublish},
pmid = {22545686},
school = {School of Experimental Psychology, University of Bristol, England. j.bowers@bristol.ac.uk},
timestamp = {2012.11.26},
}
Downloads: 0
{"_id":"twics2mLtT8rg6wJQ","bibbaseid":"bowers-davis-bayesianjustsostoriesinpsychologyandneuroscience-2012","author_short":["Bowers, J. S.","Davis, C. J."],"bibdata":{"bibtype":"article","type":"article","author":[{"propositions":[],"lastnames":["Bowers"],"firstnames":["Jeffrey","S."],"suffixes":[]},{"propositions":[],"lastnames":["Davis"],"firstnames":["Colin","J."],"suffixes":[]}],"journal":"Psychol Bull","title":"Bayesian just-so stories in psychology and neuroscience.","year":"2012","number":"3","pages":"389–414","volume":"138","abstract":"According to Bayesian theories in psychology and neuroscience, minds and brains are (near) optimal in solving a wide range of tasks. We challenge this view and argue that more traditional, non-Bayesian approaches are more promising. We make 3 main arguments. First, we show that the empirical evidence for Bayesian theories in psychology is weak. This weakness relates to the many arbitrary ways that priors, likelihoods, and utility functions can be altered in order to account for the data that are obtained, making the models unfalsifiable. It further relates to the fact that Bayesian theories are rarely better at predicting data compared with alternative (and simpler) non-Bayesian theories. Second, we show that the empirical evidence for Bayesian theories in neuroscience is weaker still. There are impressive mathematical analyses showing how populations of neurons could compute in a Bayesian manner but little or no evidence that they do. Third, we challenge the general scientific approach that characterizes Bayesian theorizing in cognitive science. A common premise is that theories in psychology should largely be constrained by a rational analysis of what the mind ought to do. We question this claim and argue that many of the important constraints come from biological, evolutionary, and processing (algorithmic) considerations that have no adaptive relevance to the problem per se. In our view, these factors have contributed to the development of many Bayesian \"just so\" stories in psychology and neuroscience; that is, mathematical analyses of cognition that can be used to explain almost any behavior as optimal.","doi":"10.1037/a0026450","keywords":"Bayes Theorem; Biological Evolution; Brain, physiology; Humans; Mental Processes, physiology; Models, Psychological; Motor Skills, physiology; Neurosciences; Psychological Theory; Psychology","language":"eng","medline-pst":"ppublish","pmid":"22545686","school":"School of Experimental Psychology, University of Bristol, England. j.bowers@bristol.ac.uk","timestamp":"2012.11.26","bibtex":"@Article{Bowers2012,\n author = {Bowers, Jeffrey S. and Davis, Colin J.},\n journal = {Psychol Bull},\n title = {Bayesian just-so stories in psychology and neuroscience.},\n year = {2012},\n number = {3},\n pages = {389--414},\n volume = {138},\n abstract = {According to Bayesian theories in psychology and neuroscience, minds\n\tand brains are (near) optimal in solving a wide range of tasks. We\n\tchallenge this view and argue that more traditional, non-Bayesian\n\tapproaches are more promising. We make 3 main arguments. First, we\n\tshow that the empirical evidence for Bayesian theories in psychology\n\tis weak. This weakness relates to the many arbitrary ways that priors,\n\tlikelihoods, and utility functions can be altered in order to account\n\tfor the data that are obtained, making the models unfalsifiable.\n\tIt further relates to the fact that Bayesian theories are rarely\n\tbetter at predicting data compared with alternative (and simpler)\n\tnon-Bayesian theories. Second, we show that the empirical evidence\n\tfor Bayesian theories in neuroscience is weaker still. There are\n\timpressive mathematical analyses showing how populations of neurons\n\tcould compute in a Bayesian manner but little or no evidence that\n\tthey do. Third, we challenge the general scientific approach that\n\tcharacterizes Bayesian theorizing in cognitive science. A common\n\tpremise is that theories in psychology should largely be constrained\n\tby a rational analysis of what the mind ought to do. We question\n\tthis claim and argue that many of the important constraints come\n\tfrom biological, evolutionary, and processing (algorithmic) considerations\n\tthat have no adaptive relevance to the problem per se. In our view,\n\tthese factors have contributed to the development of many Bayesian\n\t\"just so\" stories in psychology and neuroscience; that is, mathematical\n\tanalyses of cognition that can be used to explain almost any behavior\n\tas optimal.},\n doi = {10.1037/a0026450},\n keywords = {Bayes Theorem; Biological Evolution; Brain, physiology; Humans; Mental Processes, physiology; Models, Psychological; Motor Skills, physiology; Neurosciences; Psychological Theory; Psychology},\n language = {eng},\n medline-pst = {ppublish},\n pmid = {22545686},\n school = {School of Experimental Psychology, University of Bristol, England. j.bowers@bristol.ac.uk},\n timestamp = {2012.11.26},\n}\n\n","author_short":["Bowers, J. S.","Davis, C. J."],"key":"Bowers2012","id":"Bowers2012","bibbaseid":"bowers-davis-bayesianjustsostoriesinpsychologyandneuroscience-2012","role":"author","urls":{},"keyword":["Bayes Theorem; Biological Evolution; Brain","physiology; Humans; Mental Processes","physiology; Models","Psychological; Motor Skills","physiology; Neurosciences; Psychological Theory; Psychology"],"metadata":{"authorlinks":{}}},"bibtype":"article","biburl":"https://endress.org/publications/ansgar.bib","dataSources":["xPGxHAeh3vZpx4yyE","TXa55dQbNoWnaGmMq"],"keywords":["bayes theorem; biological evolution; brain","physiology; humans; mental processes","physiology; models","psychological; motor skills","physiology; neurosciences; psychological theory; psychology"],"search_terms":["bayesian","stories","psychology","neuroscience","bowers","davis"],"title":"Bayesian just-so stories in psychology and neuroscience.","year":2012}