Fig 1.
A frame with an open vocal tract (in this case, the schwa vowel in “the hill” during the metronome condition).
Dark blue is air and bone, light blue is tissue. A semi-polar grid is constructed over the airway (see analysis procedures for details). Gridlines were coloured by region. Red = Alveolar, yellow = Palatal, green = Velar, pink = Hyperpharyngeal, blue = Hypopharyngeal.
Fig 2.
Example alveolar movement traces from one participant who stutters.
Ten repetitions of the target utterance “the impossibility of” are overlaid (see “Analysis Procedure”, above). Each line represents one repetition. Data were reconstructed at 33.3 frames per second. Low constriction values mean the airway was more open.
Fig 3.
Coefficient of variation by group (PWS, CON), condition (“Met” and “NoMet”) and region (alveolar, palatal, velar).
Each point is the variability of one participant averaged across the three sentences. Violin plots are shown to visualize the distribution of data and its probability density for each group and condition separately for each region of the vocal tract. Boxes display median and interquartile range; whiskers represent an additional 1.5*IQR.
Fig 4.
Interaction between condition (“Met” and “No Met”) and group.
A line connects each participants’ data between the conditions. Data are averaged for each participant over region (alveolar, palatal, velar) and sentence. Boxplots display median and IQR, whiskers represent an additional 1.5*IQR.
Fig 5.
Average duration for each participant for the metronome (“Met”) and no metronome (“NoMet”) condition measured in frames (images reconstructed at 33 frames per second).
The mean of duration was centred for each of the three target utterances prior to averaging to account for differences in utterance length between the three utterances. Violin plots are shown to visualize the distribution of data and its probability density for each group and condition separately. Boxplots display median and interquartile range; whiskers represent an additional 1.5*IQR.