An Animal Model for Stuttering-Related Part-Word Repetitions
About the presenter: David B. Rosenfield, M.D. is Director of the Stuttering Center Speech Motor Control Laboratory, Department of Neurology, Baylor College of Medicine/ The Methodist Hospital. His a Professor within the Department of Neurology and holds a joint appointment in the Department of Otorhinolaryngology and Communicative Sciences. His Center engages in multiple research projects, including genetics, functional MRI, and analysis of speech signals. He established the first aviary in the Texas Medical Center and investigates physiological and pharmalogical features of modeling stuttering in birdsong. He and his colleagues have also established an intensive therapy program for people who stutter, offering therapy for adults as well as children.
An Animal Model for Stuttering-Related Part-Word Repetitions
by David B.Rosenfield, M.D. Nagalapura S. Viswanath, Ph.D., Santosh A. Helekar, M.B.B.S., Ph.D.
from PLACE
Abstract
We present an animal model of stuttering, using song output from zebra finch songbirds. We demonstrate that 7 percent of laboratory-bred zebra finches produce song syllable repetitions. Further, fifty per-cent of birds born to non-syllable-repeating adults, when raised by syllable-repeating adults, develop song with a predominance of syllable repetitions in which the repeated syllables are not identical to those that are repeated by their adult tutors. Our data supports the thesis that occurrence of multiple repetitions of song syllables, which share similarities with part-word repetitions in human stuttering, may be dependent on innate as well as acquired factors.
Proposal: An Animal Model for Stuttering-Related Part-Word Repetitions
Acquisition of zebra finch birdsong and the learning of human speech share certain characteristics: 1) Both are spectrally as well as temporally complex learned vocal behaviors; 2) Each occurs early in development (Konishi, 1965; Marler, 1970); 3) They critically depend on appropriate auditory feedback (Konishi, 1965; Marler, 1970); 4) Both are associated with specific areas of brain specialized for vocal learning and production (Nottebohm et. al, 1976). Zebra finches offers a unique opportunity to investigate the neurological basis of learning complex time-varying information, especially as it relates to speech-motor output.
There are two separate phases of song learning in zebra finches. During sensory learning ( 15- 35 days), young birds hear and memorize a parent tutor song, or "template" (Marler, 1970; Slater et. al, 1988). During the subsequent sensorimotor phase (35-90 days), the bird sings and refines its song output until the song approximates the memorized song template. Young birds do not need to hear the tutor during this latter period of vocal practice, but do require normal hearing (Konishi, 1965). There is a discreet cerebral anatomic circuit for the learning of song, separate from the well-identified cerebral pathways necessary for normal production (reviewed in Doupe, 1997). In other words, the parts of the brain necessary for learning the song are separate from the parts necessary to produce the song in an adult.
Human speech, opposed to language, is primarily motor output. Stuttering is a disturbance of the speech-motor control system (e.g., people who stutter know what they want to say, but can not say it). Our laboratory is investigating song irregularities in the songbird, zebra finch, to identify putative animal models of stuttering. In a detailed analysis of song pattern irregularities in our zebra finch colony (Helekar, Marsh, Viswanath and Rosenfield, 1999, under review), we have observed that 7 per-cent of laboratory-bred zebra finches repeat single song syllables, similar to part-word repetitions of human stuttering (Helekar, Viswanath and Rosenfield, 1999, under review). These repetitive patterns are not copied in their entirety by juvenile birds from adult male tutors, during song learning. Yet, the syntactic rule of repeating syllables can be learned by offspring of non-syllable repeating birds that are raised artificially in the company of adult tutors who are repeaters. This highlights the significance of a birdsong irregularity to stuttering speech, while demonstrating that learned acquisition of song syntax (i.e., temporal frame) can occur independently from song syllable content (i.e., spectral content) (Helekar, Viswanath, Rosenfield, 1999, under review). This modeling of human stuttering is bolstered by the fact that, as in human speech, delayed auditory feedback in zebra finches produces song irregularities resembling stuttering (Leonardo and Konishi, 1999).
Our animal model of stuttering provides an opportunity to investigate how the brain learns complex, time-varying information, using birdsong and a human clinical correlate -- speech. We are currently attempting to substantiate further this model, by applying it to the non-adult learning of speech, investigating whether there is an increased prevalence of "stuttering" among juvenile zebra finches, and hope to have data on this issue as well, by the time of the Conference. If this is similar to humans, it should be at least four times more prevalent among young birds than among adult birds. This further analysis might also provide an animal model for "normal stuttering of childhood."
We shall discuss our completed investigations in the adult birds as well as our current investigations in the juvenile birds. Our further goal is to investigate the neurophysiology and neuropharmacology of birdsong, with the goal of investigating possible analogues of stuttering from the perspective of non-linguistic neuromotor control.
September 3, 1999
