Evaluating generalised additive mixed modelling strategies for dynamic speech analysis

• Published in: Journal of phonetics Vol. 84; p. 101017
• Main Author: Sóskuthy, Márton
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01-01-2021
• Subjects: Dynamic speech analysis; Generalised additive mixed models; Random effects; Significance testing; Type I/II error simulations; Dynamic speech analysis; Random effects; Generalised additive mixed models; Type I/II error simulations; Significance testing
• ISSN: 0095-4470; 1095-8576
• DOI: 10.1016/j.wocn.2020.101017

•Type I/II error simulations provide guidance about GAMMs in phonetic research.•Simulations are run using both real and simulated formant contours and real pitch contours.•The paper provides advice on using random smooths and AR1 error components.•The paper provides guidance on when different methods of significance testing are appropriate.
Generalised additive mixed models (GAMMs) are increasingly popular in dynamic speech analysis, where the focus is on measurements with temporal or spatial structure such as formant, pitch or tongue contours. GAMMs provide a range of tools for dealing with the non-linear contour shapes and complex hierarchical organisation characteristic of such data sets. This, however, means that analysts are faced with non-trivial choices, many of which have a serious impact on the statistical validity of their analyses. This paper presents type I and type II error simulations to help researchers make informed decisions about modelling strategies when using GAMMs to analyse phonetic data. The simulations are based on two real data sets containing F2 and pitch contours, and a simulated data set modelled after the F2 data. They reflect typical scenarios in dynamic speech analysis. The main emphasis is on (i) dealing with dependencies within contours and higher-level units using random structures and other tools, and (ii) strategies for significance testing using GAMMs. The paper concludes with a small set of recommendations for fitting GAMMs, and provides advice on diagnosing issues and tailoring GAMMs to specific data sets. It is also accompanied by a GitHub repository including a tutorial on running type I error simulations for existing data sets: https://github.com/soskuthy/gamm_strategies.