Sound Proof: What’s the Evidence on Target Selection for Speech Sound Disorders

Did you know that target selection is an important variable in treatment outcomes? In fact, the therapy targets, rather than the exact treatment approach, may be the instrument of change.^1,2

Over time, clinicians have used a number of factors in target selection for intervention³, including:

• Age of the child
• Developmental level of the target sound
• Effect on overall intelligibility
• Error consistency
• Relevance of the sound to the child
• Severity of the disorder
• Stimulability

Many of these factors are originally based on the ease of learning individual sounds. However, a shift in thinking about target selection occurred that considers the influence of linguistic factors on learning, and more specifically, on system-wide generalization. The new approach hypothesizes that selection of certain target sounds has a cascade effect, and therefore would be more efficient. With these selections, the child learns not only the targeted sound or sound class but also other untrained aspects of their sound system.

Sound Learning Versus System Shifting

Per the flowchart below, target selection options can be classified into either Phonetic (or Articulatory) or Phonemic (Phonological) to describe the traditional, complexity and distance metric approaches:

The difference between the phonetic and phonemic perspectives is sound learning versus system shifting. That is, the traditional approach results in the child learning the target sound more easily because it is an earlier developing, stimulable sound. In contrast, the phonological approach should result in a system-wide change.

Digging deeper, a direct comparison of the factors highlighted above is summarized in the table below:

The traditional and complexity approaches consider the same factors related to individual sounds, but are otherwise opposites.⁴ Each focuses on sound characteristics and thus target selection is based on those same characteristics, regardless of the severity or nature of a child’s speech sound disorder.

The distance metric assumes the importance of the target sound’s function in a child’s system is greater than the characteristics of an individual sound.⁷ Consequently, the target sound is selected based on the phonetic distance parameters of maximal classification and maximal distinction rather than developmental norms, stimulability, consistency or knowledge. As a result, the target sound will differ for individual children based on their unique sound system.

The parameters of maximal classification and distinction are based on selecting targets that have the greatest phonetic distance between the child’s error and the target, as well as the greatest phonetic distance across target sounds. Increasing the phonetic distance in these two ways increases both the saliency of the sound(s) to be learned, as well as the frame of learning that needs to be achieved by the child. That is, the child may need to learn more than a single aspect of sound production, such as manner, place, or voicing.

What’s the Evidence?

Unfortunately, few studies have examined treatment outcomes relative to target selection. One study by Gierut, Morrisette, Hughes and Rowland (1996) examined the complexity approach.⁵ Their results revealed that selection of later developing sounds, representing least phonological knowledge, led to greater sound changes. Treatment of superordinate properties (phonemic factors) facilitated mastery of subordinate properties (phonetic factors), resulting in greater system-wide change.

Rvachew and Nowak (2001) conducted a similar investigation and reported opposite findings.⁶ They found that selection of early developing targets associated with greater productive phonological knowledge resulted in more progress toward the target sounds.

How do we interpret these conflicting results?

The answer may be in the type of learning that was measured, i.e., sound learning versus system-wide change. As discussed earlier, the traditional approach focuses on the ease of learning the sound whereas the complexity approach addresses system-wide change. Traditional targets will be easier to learn, whereas the complexity approach will lead to greater change across the system.

Williams (2006) compared the treatment of systemic (i.e. selection based on the function of sound in a system) to non-systemic (i.e. selection based on the characteristic of the sound) treatment targets.⁸ The results revealed the following:

• Systemic targets resulted in the greatest amount of system-wide change
• Non-systemic targets led to greater sound stabilization

These results generated the following proposed model that integrates both target selection approaches:

1. Initially, select targets directed at creating system-wide change (i.e., system shifting)
2. Followed by selecting targets that will stabilize those newly learned sounds (sound learning)

Making Sound Sense

The different approaches to target selection address different aspects of learning; i.e., sound learning versus system shifting. Although the results of the studies appear contradictory, it’s important to realize different types of outcomes were measured. When looking at the results from sound learning versus system shifting, the outcomes are consistent with the assumptions of each approach. Taken together, these results can inform clinicians about designing intervention that addresses a child’s learning needs and mirrors changes as a result of an intervention. Initial treatment targets should be selected that will result in the greatest amount of system-wide change (i.e., either complexity or distance metric approaches), and followed by target selection that will result in greater stabilization of newly learned contrasts (traditional approach).