For speech-language pathologists

Which sound should you treat first? Developmental vs complexity approaches

Published · Updated

The short answer

There’s no single right answer. Traditional practice treats early-developing, stimulable sounds first, which in one trial gave more progress on the treated sounds. The complexity approach treats later-developing, nonstimulable or more marked targets, such as clusters, expecting wider generalization to untreated sounds. ASHA also lists impact on intelligibility and relevance to the child and family. Head-to-head evidence is small and mixed, so the choice rests on the child’s profile, the family’s priorities and the time available.

The criteria SLPs use

ASHA’s Practice Portal lists several ways of selecting initial targets for children with speech sound disorders (ASHA):

Target selection approaches listed by ASHA
ApproachHow targets are chosen
Client-specificRelevance to the child and family (for example, a sound in the child’s name), stimulability, or how visible the sound is (for example /f/ rather than /k/)
IntelligibilityThe errors that most affect intelligibility, such as omissions or initial consonant deletion
DevelopmentalThe order in which typically developing children acquire sounds
ComplexityMore complex, linguistically marked elements not in the child’s system, such as clusters, fricatives and affricates, or the least stimulable sounds, to encourage cascading generalization
Dynamic systemsSimple targets that don’t introduce new feature contrasts, to stabilize the foundations for later, more complex sounds
SystemicTargets chosen for their function in the child’s system, maximally distinct from each other and from the child’s error in place, voice and manner

These aren’t mutually exclusive, and the criteria can point in opposite directions: a developmental view favours early sounds the child almost has, while a complexity view favours late, unknown ones. ASHA also notes that treatment selection depends on the child’s age, the type of errors, severity and the effect on intelligibility (ASHA).

Developmental and stimulable targets

The traditional logic is to teach what comes next in typical development. In a review of 15 US sources (6 articles and 9 assessments) covering 18,907 children, most consonants were acquired by 5;0 at a 90% criterion, with /ɹ/, /ð/ and /ʒ/ by 5;0–5;11 and /θ/ by 6;0–6;11. Ages varied across studies, and the authors recommend drawing on several sources (Crowe & McLeod, 2020). See speech sounds by age.

The complexity approach

Gierut’s complexity approach turns the developmental logic around: treating more complex, marked properties is expected to bring greater benefit to the child’s whole sound system, with a cascading effect on untreated sounds. Gierut acknowledges that this can seem counterintuitive, that the supporting studies used single-subject designs, and that the children studied had phonological errors with no other delays; some children may need simpler input because of attention or processing limits (Gierut, 2007).

Storkel’s tutorial on putting it into practice weighs characteristics of the targets (developmental norms and implicational universals, where knowing a complex sound implies knowing simpler ones) and the child’s knowledge of them (accuracy and stimulability), with free worksheets and probes for preschool children with phonological disorders (Storkel, 2018).

Uptake is low. Storkel’s tutorial was written partly because US clinicians are less familiar with the complexity approach than with other evidence-based options (Storkel, 2018). In a UK survey of 166 speech and language therapists, complexity approaches such as the empty set and two- to three-element clusters were among the least used, and some respondents were unsure what they involved (Hegarty et al., 2018).

Intelligibility, function and the family

What the evidence comparing approaches shows

Studies comparing simpler and more complex targets
StudyChildrenFinding
Gierut et al., 19969, in two single-subject studiesSimilar amounts of change; later-acquired targets brought faster, system-wide gains
Rvachew & Nowak, 200148, assigned to two target-selection strategiesEarly, partly known targets: more progress on treated sounds; no difference in generalization
Rvachew & Bernhardt, 20106, reanalyzed from the 2001 trialSimple targets: more progress and emergence of complex untreated sounds; complex targets: little gain
Maggu et al., 2021 (meta-analysis)50, across 12 single-case studiesComplex input improved treated complex sounds (d = 1.08) and untreated simpler sounds (d = 2.69)
Krueger & Storkel, 20238, aged 4–5 or 7–8Late-acquired sounds learned as effectively by younger children

The evidence is genuinely mixed. Support for complexity comes mostly from single-case studies: in the meta-analysis, all 12 included studies used single-case designs, only two used blinding, and the authors acknowledge possible publication bias; it didn’t include the 2001 trial (Maggu et al., 2021). The one larger comparison favoured simpler targets for progress on the treated sounds but found no difference in generalization (Rvachew & Nowak, 2001). A systematic review of preschool speech sound interventions found most of the evidence was lower-graded (Wren et al., 2018).

Putting it together

Questions SLPs ask

Should I treat stimulable or nonstimulable sounds first?

Both are defensible. A trial of 48 children found faster progress on treated sounds with early, partly known targets, while complexity studies report wider generalization from nonstimulable, later-developing targets. Neither has decisive evidence.

What is the complexity approach in speech therapy?

A way of choosing targets that treats more complex sounds first, such as clusters, later-developing or nonstimulable sounds, expecting simpler untreated sounds to improve too. Most of its evidence comes from single-case studies.

Is it too early to treat a late-developing sound in a young child?

Not necessarily. In a small study, children aged 4 to 5 learned late-acquired sounds as effectively as children aged 7 to 8, and the authors concluded that delaying treatment of individual sounds is unnecessary.

How does intelligibility affect target selection?

ASHA lists choosing the errors that most affect intelligibility as a criterion in its own right. Low intelligibility and poor stimulability also predicted errors that were less likely to resolve without therapy.

Sources

Join the Early Adopter Program