The classification of types or subgroups of speech sound disorder (SSD) has been of interest in the last decade, because the heterogeneity in this population has been acknowledged and examined. Currently, there are several widely recognized classification systems based on etiology (Shriberg, 2010), surface speech patterns (Broomfield & Dodd, 2004; Dodd, Zhu, Crosbie, Holm, & Ozanne, 2006), or underlying deficits (Stackhouse & Wells, 1997). The latter two systems focus more squarely on symptoms and proximal causes of possible subtypes of SSD. The etiological system is aimed at delineating diagnostic markers related to underlying genetic influences. Although proposed subtypes differ across these systems, similarities among them suggest that there are at least three broad groups: (1) a group with speech delay/disorder whose errors are primarily substitutions and omissions of differing degrees of involvement/severity; (2) a group with speech errors that are articulatory in nature, characterized by distortions; and (3) a group with speech, segmental, and suprasegmental features that are consistent with dysarthria, apraxia, or both. Such classification systems are relevant not only for furthering our characterization of childhood SSD, but more importantly for testing the efficiency of current treatments and development of new treatments uniquely suited for a specific subtype. In this issue, investigators explore the classification of subtypes with different methods, explore behavioral traits associated with unique subtypes, and examine the efficiency of several interventions.
The issue begins with an age-old question, revisited by Barbara Dodd, regarding whether or not it is important to distinguish between a delay and a disorder. Dodd asks this question because it pertains to the children's underlying cognitive-linguistic skills and, in particular, deficits that may separate groups and be important for developing effective interventions. Dodd examines cognitive flexibility and nonverbal rule abstraction in two groups; one is differentially diagnosed as delayed because there are speech error patterns seen in typical development, but not appropriate any longer for the child's age. The second group is differentially diagnosed as disordered because of the presence of speech error patterns that are atypical during development, as represented in a large normative sample. The disordered group performed more poorly on a task requiring shifting attention between conceptual domains and on one involving rule abstraction using computer animation. Such tasks also involve memory and inhibition and thus point to general cognitive processes in phonological acquisition. Dodd suggests that knowledge of such deficits can inform treatment so that activities focus on rules and patterns, as well as expressive output, and that this should lead to improvements in phonological awareness and literacy.
These most challenging difficulties in the area of literacy and academic achievement for children with persistent SSD are also at the heart of the work described by Lewis and colleagues in the next article. They too examine cognitive-linguistic skills thought to be closely associated with behavioral traits of disorders such as SSD for the purposes of genetic analysis. Measures of oral motor skills, phonological awareness, phonological memory, speeded naming, and vocabulary are compared for a large sample of 4- to 7-year-olds referred from clinical caseloads of practicing speech-language pathologists. Skills are compared as the sample is partitioned using three different classification systems to identify different subgroups—one based on severity, one on the cooccurrence of language impairment (LI) and reading disorder (RD) with SSD, and the last on subtypes encountered in clinical practice. Phonological memory, phonological awareness, and vocabulary differentiated subtypes among each of the classification systems. Children with cooccurring SSD + LI + RD performed more poorly in both phonological memory and phonological awareness than children with SSD only or SSD and just one other cooccurring disorder. Children with SSD + LI also performed more poorly on these measures than children with SSD only. These results reinforce the clinical impression that SSD often cooccurs with other disorders and manifests a variety of underlying deficits best explained by the Multiple Deficit Model (Pennington, 2006; Pennington & Bishop, 2009).
Appealing to this same multiple deficit model, the next article describes an intervention study involving preschool children with severe cooccurring SSD + LI and multifaceted needs. These needs in oral language, speech intelligibility, and phonological awareness/emerging literacy were addressed differently using two previously tested and proven, efficacious interventions. Tyler, Gillon, and colleagues sought to determine whether a phonological awareness intervention could be successfully provided in younger children with cooccurring impairments and to examine the extent of direct and indirect gains in skills addressed across both interventions. Intervention groups made significant gains in all measures – speech, language, and phoneme awareness—and the gains in percent consonants correct were large. There were, however, only trends favoring better performance for the skills targeted specifically by each intervention. Such results magnify the impact of multiple underlying deficits and long-term need for interventions that target affected cognitive-linguistic skills.
Also in search of skills that are linked to behavioral traits and shed light on genetic influences, Peter and Raskind explore motor abilities in multigenerational families. Their approach presumes that it may be more fruitful to study multiple generations of families for unique subtypes rather than a large heterogeneous population to identify candidate genes. They investigate repetitive and alternating speech and hand motor tasks, adjusting for change in these tasks across the life span. What their results provide is evidence that genetic mechanisms cluster differently in different families. Two families shared similar motor deficits underlying SSD consistent with childhood apraxia of speech (CAS). Clinically, this verifies the importance of identifying the crucial underlying skill(s) and designing and testing interventions aimed at diminishing the behavioral symptoms to eliminate long-term effects. It also means that if we have the luxury to be able to identify the probable underlying mechanism in any one family, we will know more about the type of intervention to provide.
In the final article, Preston and Koenig explore whether phonetic variability in repeated productions can be used to identify subgroups in children with residual SSD with an average age of 12;1 They approach this little studied group with no a priori assumptions about subtype classifications. Although acoustic and transcription-based measures were not correlated and did not differentiate groups with high or low variability, there were associations between transcription-based measures and language skills. Once again, these findings lead us to questions concerning the relation between less accurate speech production and poorer language skills.
As a group, the articles in this issue bring researchers and clinicians closer to understanding the critical underlying skills that may be disrupted in different types of SSD, whether those are identified by type of surface speech error pattern or cooccurring condition such as LI or RD. Those critical skills appear to be primarily cognitive-linguistic; phonological memory, phonological awareness, vocabulary, cognitive flexibility, and rule abstraction. For some cases, especially those where CAS is suspected, there may be slowed motor sequencing, detectable across generations of a family. Clinically, these findings implore us to investigate interventions targeted specifically to these skills in well-defined and documented clinical subgroups. Such is the kind of intervention study reported in this issue. It joins a growing literature that suggests for children with multiple deficits, interventions must integrate oral language and phoneme awareness/early literacy skills. One important element is emerging from this intervention research: Speech intelligibility need not be the sole focus of the intervention. As long as speech production activities, such as minimal pairs, are integrated with other goals, gains appear to be a downstream byproduct of various cognitive-linguistic interventions. Thus, when we are assured that a motor impairment is not operating, an integrated focus on linguistic rule and pattern learning, phonological memory, and phonological awareness may be the most efficacious for clinical cases from the typical subgroups we encounter as clinicians.
Ann A. Tyler, PhD, CCC-SLP
Broomfield J., Dodd B. (2004). Children with speech and language disability: caseload characteristics. International Journal of Language and Communication Disorders, 39, 303–334.
Dodd B., Zhu H., Crosbie S., Holm A., Ozanne A. (2006). Diagnostic evaluation of articulation and phonology. San Antonio, TX: Psychological Corporation.
Pennington B. F. (2006). From single to multiple deficit models of developmental disorders. Cognition 101, 385–413.
Pennington B. F., Bishop D. V. M. (2009). Relations among speech, language, and reading disorders. Annual Review of Psychology, 60, 283–306.
Shriberg L. D. (2010). Childhood speech sound disorders: From postbehaviorism to the postgenomic era. In R. Paul & P. Flipsen (Eds.), Speech sound disorders in children (PP. 1–33). San Diego, CA: Plural Publishing.
Stackhouse J., Wells B. (1997). Children's speech and literacy difficulties. London: Whurr.