Presentation 2006-06-15
A hidden Markov model for human information processing during a POMDP task
Masanori OKURA, Wako YOSHIDA, Shin ISHII,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Observable information from the real world is often incomplete. In the machine learning field, decision making problems in uncertain environments are formulated as partially observable Markov decision processes (POMDPs). The action selection based on the estimation of unobservable (hidden) true states using observation sequence is effective in POMDPs and reasonable as human decision-making process. In this report, we conducted an fMRI experiment using a maze navigation task, and formulated subject's information processing as a hidden Markov model. Our model, where the action selection policies were switched depending on the level of confidence for estimation, well reproduced subjects' behavior. We used the cognitive loads evaluated by our model as regression functions of imaging analysis, and found that the estimation of hidden states is realized in the prefrontal cortex.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) partially observable environment / hidden Markov model / prefrontal cortex / functional magnetic resonance imaging (fMRI)
Paper # NC2006-9
Date of Issue

Conference Information
Committee NC
Conference Date 2006/6/8(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A hidden Markov model for human information processing during a POMDP task
Sub Title (in English)
Keyword(1) partially observable environment
Keyword(2) hidden Markov model
Keyword(3) prefrontal cortex
Keyword(4) functional magnetic resonance imaging (fMRI)
1st Author's Name Masanori OKURA
1st Author's Affiliation Nara Institute of Science and Technology()
2nd Author's Name Wako YOSHIDA
2nd Author's Affiliation Nara Institute of Science and Technology
3rd Author's Name Shin ISHII
3rd Author's Affiliation Nara Institute of Science and Technology
Date 2006-06-15
Paper # NC2006-9
Volume (vol) vol.106
Number (no) 101
Page pp.pp.-
#Pages 6
Date of Issue