“…We also tested whether our measures were significant predictors of iconicity ratings against a strong baseline of psycholinguistic variables that have been shown to be associated with such construct. Indeed, it has been shown that iconicity ratings are negatively associated with word length (Perry et al, 2015;Winter et al, 2017Winter et al, , 2022, log frequency (Perry et al, 2015(Perry et al, , 2018Winter et al, 2017Winter et al, , 2022, age of acquisition (Perry et al, 2018;Winter et al, 2022), concreteness (Hinojosa et al, 2021;Winter et al, 2022), and contextual diversity ; furthermore, they are positively correlated with perceptual experience (Sidhu & Pexman, 2018;Winter et al, 2017Winter et al, , 2022. In our analyses, we included these measures as covariates, to test the predictive power of our metrics while controlling for other possible confounds.…”