Paper accepted in the Journal of the Acoustical Society of America
Be on the lookout for "Context effects in perception of vowels differentiated by F1 are not influenced by variability in talkers’ mean F1 or F3" in the Journal of the Acoustical Society of America later this year! This paper started as first author Hannah Mills' Culminating Undergraduate Experience in the fall of 2020, complete with online poster presentation. Then, with revision and another experiment added, became this paper. Trial-by-trial variability in talkers' mean f0 can diminish the size of spectral context effects in vowel identification (Assgari & Stilp, 2015 JASA; Assgari, Theodore, & Stilp, 2019 JASA), but that was far from the only acoustic property varying across talkers' voices. Here, surprisingly, we showed that variability in mean F1 (which was relevant to the target vowels, which were primarily differentiated in F1) had no bearing on the size of the context effects. Also surprisingly, variability in mean F3 (which has a history in talker differentiation and vocal tract scaling) also had no bearing on the context effects. These results challenged our own theoretical motivation behind these experiments, showing that just because an acoustic property is variable or consistent doesn't guarantee that it matters for the perceptual task at hand.