Premium
WIENER FILTER ESTIMATION OF TRANSFER FUNCTIONS
Author(s) -
Kessel Robert
Publication year - 2004
Publication title -
journal of the experimental analysis of behavior
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.75
H-Index - 61
eISSN - 1938-3711
pISSN - 0022-5002
DOI - 10.1901/jeab.2004.81-289
Subject(s) - wiener filter , transfer function , fidelity , filter (signal processing) , computer science , noise (video) , monte carlo method , algorithm , linear filter , function (biology) , mathematics , artificial intelligence , statistics , engineering , computer vision , telecommunications , evolutionary biology , electrical engineering , image (mathematics) , biology
The use of a Wiener filter estimate for the linear transfer function can significantly improve the description of behavioral dynamics. This report presents a two‐pass, Monte‐Carlo‐based algorithm that is well suited to repeated‐trials local average measurements. The Wiener filter transfer functions strongly suppress noise artifacts as well as allow reliable transfer function determination under a much wider class of reinforcement schedules. Implications of expanding the possible form of experimental design are considered along with improvements in the fidelity of resulting predictions.