Relationships between duration and variability measures of children's speech

Abstract

In recent years, several studies have found that young children often have longer speech segment durations than older children and adults do and that young children are also typically more variable across repetitions of stimuli. It has generally been suggested that both of these findings are indications of increased speech motor control as children become older. If both measures are reflections of neuromotor maturation of the speech production system, it seems that they should be highly correlated with one another and that they should lead to similar conclusions about the speech production abilities of individual subjects. However, data from children ranging from about 3–10 years of age indicate that duration and variability measures often are not very strongly correlated. For example, children who have shorter segment durations do not always show less variability than children with longer durations, and children with greater variability do not necessarily have longer segment durations. In addition, it appears that children may achieve adultlike duration values before measures of their variability become adultlike. Possible implications of these findings regarding children's speech timing control will be discussed.