Fig 1.
Example Non-nutritive suck (NNS) bursts: Infant NNS is arranged in bursts of sucking with pause periods for respiration.
Each burst contains cycles within it, which are depicted as black dots in this image. The frequency (Hz) within a burst is measured by the number of cycles per second. The strength of the suck cycle, or amplitude (cmH20), is measured by examining the top of each cycle where the black dot is located. This image depicts 50s of NNS data with 3 NNS bursts. Burst 1 has 27 cycles/burst (burst duration of 16 sec), Burst 2 has 15 cycles/burst (burst duration 8 seconds) and Burst 3 with 19 cycles/burst (burst duration 11 seconds). Amplitude of each cycle (black dot) can be determined looking at the y-axis.
Fig 2.
LENA vest (left) and Voice onset time example (Right). Left: An infant wearing the LENAvest during recording (The parent of the infant pictured in this manuscript has given written informed consent (as outlined in PLOS consent form) to publish these case details). Right: Praat window with waveform and spectrogram from a stop consonant-vowel production. Voice onset time is measured from the burst to the start of the subsequent vowel, indicated by the highlighted portion.
Table 1.
Perception of stop consonants during initial calculation of voice onset time.
Fig 3.
Voice onset time medians (blue dots) and ranges for each infant.
Fig 4.
Relationships between voice onset time variability and non-nutritive suck.
Relationship between the actual ranked coefficient of variation (CoV) of voice onset time and the CoV of voice onset time predicted by the regression model that included NNS measures of burst duration, frequency, amplitude, and burst amount. NNS measures of burst duration and intraburst frequency were significant predictors. Shaded area represents 95% confidence interval around the regression line.
Table 2.
Correlation matrix of non-nutritive suck (NNS) measures.