Facial motion capture is a technology used in computer animation to simulate the facial expressions of real people when creating animated characters for video games or movies. By converting facial movements into a digital database, based on specific reference points on an actor’s face, a more realistic animated character can be created. Now, a new study by researchers at New York University’s Steinhardt School of Culture, Education, and Human Development shows that facial motion capture can be used to distinguish between children with childhood apraxia of speech (CAS) and those with various other speech impairments. CAS, a type of motor speech disorder, is characterized by deficits in planning and executing the precise movements required for creating speech sounds. CAS causes delayed speech development as well as atypical patterns of speech. Children with CAS progress much slower during speech therapy than children with other speech impairments.
The current study used facial motion capture technology to track the movements of the jaw, upper lip and lower lip of children with CAS versus those with other speech impairments, and those without speech impairments. The researchers related these movement patterns to the perceived accuracy of speech by the listener, since current diagnostic and treatment protocols rely predominantly on the listener’s judgment of accurate speech. The authors found that facial motion capture was able to detect motor impairments in CAS subjects whose words were perceived by listeners to be pronounced correctly. Moreover, the subtle differences detected by the technology indicated that the lip and jaw movements of children with CAS were more variable than those in children with other speech impairments or unaffected children. Such movement variability may underlie the differential response to therapy. Uncovering the basis of poor response to treatment may facilitate the investigation of novel, more effective treatment strategies, and underscores the over-reliance on measuring speech sounds by auditory interpretation alone.