著者
武藤 拓之
出版者
日本基礎心理学会
雑誌
基礎心理学研究 (ISSN:02877651)
巻号頁・発行日
pp.39.27, (Released:2021-04-01)
参考文献数
37

Hierarchical Bayesian modeling is a powerful and promising tool that aids experimental psychologists to flexibly build and evaluate interpretable statistical models that consider inter-individual and inter-trial variability. This article offers several examples of hierarchical Bayesian modeling to introduce the idea and to show its implementation with R and Stan. As a tutorial, it uses data from well-known experimental paradigms in perceptual and cognitive psychology. Specifically, I present linear models for correct response time data from a mental rotation task, probit models for binary choice data from two psychophysical tasks, and drift diffusion models for both response time and binary choice data from an Eriksen flanker task. The R and Stan scripts and data are available on the Open Science Framework repository at https://doi.org/10.17605/osf.io/2zxs6. The importance of model selection and the potential functions of open data practices in statistical modeling are also briefly discussed.

言及状況

外部データベース (DOI)

Twitter (11 users, 12 posts, 47 favorites)

<論文の紹介>[bot] 実験心理学でおなじみの課題を題材とした,ベイズ統計モデリングの解説論文です(分析スクリプト付き)。>武藤 (2021). 実験心理学者のための階層ベイズモデリング入門──RとStanによるチュートリアル── https://t.co/Un2BlULT7N

収集済み URL リスト