Currently, I am conducting a clinical efficacy research program, examining the effects of two different types of treatment for phonological (i.e., speech articulation and sound usage) disorders. The research program is examining the role of word lexicality during the treatment of phonological disorders. The specific research question is quite direct: Does speech treatment involving real words induce greater change than does speech treatment involving nonwords? While the research question may be simple, the outcome of the research program may impact how speech treatment protocols are applied to future individuals with phonological disorders. Thus, the current children in my research program directly benefit from my research due to improvement in their speech production. Additionally, future children with phonological disorders and their clinicians will also benefit once the results are published in a clear and concise manner so that the research methods, designs, and results are understandable to busy speech pathologists, who often have minimal amounts of time to research new treatment methodologies.

Speech production data (from treatment studies) alone are insufficient to detail the complexities of a child’s phonological system; children’s perceptual abilities must also be measured. While addressing speech treatment approaches is interesting, valuable, and necessary work, I am also seeking to expand the research and knowledge base surrounding possible underlying causes of phonological disorders. It is possible that the underlying problem of a productive phonological disorder may be perceptual in nature. While traditional treatment studies do not utilize methodologies that can measure perceptual knowledge, my research program currently uses behavioral and electrophysiological measures to provide data that are both relevant to and independent of treatment outcomes.

My research program has taken the approach to address how the perception of speech sounds is related to the production of speech sounds at multiple levels in an attempt to identify where the potential breakdown in the system occurs. Again, the basic research principle is simple: if children do no perceive, process, and store a speech sound in the correct manner, it is quite unlikely that the sound will be produced accurately. The first step in accurate phonological production is the ability to perceive differences between two sounds. Thus, one behavioral experiment presents children with two different pictures with rhyming labels (e.g., “ring” and “wing”) and then asks the child to point to one or the other. The experiment is catered to the child’s individual speech abilities and aims to pair a child’s incorrect production (e.g., saying “wing” for “ring”) with the correct adult production (e.g., “ring”).

Even if children can perceive the difference between two sounds, the auditory input must be stored, represented, and accessed correctly within the brain. Thus, the next step in accurate phonological production is the ability to compare a heard sound to the sounds already stored within the phonemic (i.e., sound) inventory in the brain. The second behavioral experiment presents children with a single picture (e.g., “ring”) and then asks the child on separate occasions, “is this a ring” or “is this a wing”. Here, the child must make the decision as to whether or not the spoken word label is appropriate for the presented picture, comparing the spoken word with his own internalized word label. Again, this experiment is catered to a child’s individual speech and is meant to potentially tax his ability to perceive subtle differences between word productions.

The final component to my current research program is examining how perceptual abilities measured at the brain level may or may not be different from those measured at the behavioral level. The research question for the electrophysiological experiments is also straightforward: Will children with phonological disorders show attenuated neural responses to sounds that they cannot produce, as compared to the sounds they do produce accurately? In the event-related potential (ERP) task, children are presented a majority of the time with a single syllable sound that they can produce (e.g., “ba”) and then a small portion of the time they are presented with three other sounds: a sound they can produce (e.g., “ga”), their treatment sound (e.g., “ra”), and a third sound they cannot produce (e.g., “la”). If indeed phonological disorders are due to problems detecting the perceptual differences between sounds, then results from this experiment may support this claim. More specifically, it is expected that the children with phonological disorders will demonstrate smaller or insignificant neural responses (i.e., Mismatch Negativity responses; MMN) to the syllables they cannot produce accurately. This result is expected because if a child is capable of producing a sound correctly in different contexts, then his mental representation of that sound must be correct and successful discrimination from another syllable is performed. However, when the deviant syllable is poorly perceived the discrimination is less precise.

While each of the pieces of my research program asks strong research questions and can stand alone, the important aspect of my work is how they also tie together to allow for more in-depth interpretation of the data. Specifically, both the behavioral and electrophysiological experiments are being conducted on the children prior to entering into treatment, in addition to when they have completed the treatment protocol. Thus, any potential change in perceptual ability due to the treatment experience and the auditory bombardment of a single treatment sound can be documented. It is also possible that some children will not show productive speech change by the end of the treatment. In other words, some children will still not be able to correctly produce their chosen treatment sound correctly. This is where most treatment studies end; the child would be classified as one that showed no change due to treatment. However, it is quite possible that there are other reasons as to why the child was unable to accurately produce the sound, while still being able to accurately perceive the sound. Since accurate perception is the first step in accurate production, by measuring the perceptual abilities at the end of treatment, a more accurate treatment prognosis can be made. If perceptual change is seen, then it should theoretically be only a matter of time before behavioral speech production change is observed.