FMAT: The Fill-Mask Association Test

The Fill-Mask Association Test ('FMAT') is an integrative, versatile, and probability-based method that uses Masked Language Models to measure conceptual associations or relations (e.g., attitudes, biases, stereotypes, social norms, cultural values) as propositional representations in natural language. The supported language models include 'BERT' (Devlin et al., 2018) <arXiv:1810.04805> and its model variants available at 'Hugging Face' <>. 'Python' ('conda') environment and the 'transformers' module can be installed automatically using the FMAT_load() function. Methodological references and technical details are provided at <>.

Version: 2023.8
Depends: R (≥ 4.0.0)
Imports: PsychWordVec, psych, reticulate, text, data.table, stringr, forcats, glue, cli, purrr, plyr, dplyr, tidyr
Suggests: bruceR, nlme, parallel
Published: 2023-08-11
Author: Han-Wu-Shuang Bao ORCID iD [aut, cre]
Maintainer: Han-Wu-Shuang Bao <baohws at>
License: GPL-3
NeedsCompilation: no
SystemRequirements: Python (>= 3.6.0)
Materials: README NEWS
CRAN checks: FMAT results


Reference manual: FMAT.pdf


Package source: FMAT_2023.8.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
macOS binaries: r-release (arm64): FMAT_2023.8.tgz, r-oldrel (arm64): FMAT_2023.8.tgz, r-release (x86_64): FMAT_2023.8.tgz, r-oldrel (x86_64): FMAT_2023.8.tgz


Please use the canonical form to link to this page.