Choosing the right words in speech sound treatment: (aka: No, seriously—is there a deck for this?)

When treating a young child with a speech sound disorder, how do you choose which words to work on? Many of us choose words we predict to be most meaningful to the child and most likely used on a regular basis (e.g. targeting /s/ in “sick” instead of “soot”). The author agrees that functional communication often motivates SLPs’ choices. Also convenience—for example, using a ready-made deck of cards, regardless of what words are in it, as long as it has the right target sounds in the right position(s).
In this paper, however, the author summarizes research to show that choosing stimuli this way is inadequate. Instead, lexical properties of words must be considered because of their significant impact on phonological learning. She suggests two things that we should be considering:

  • (1) WORD FREQUENCY: For the greatest learning benefit to both treated and untreated sounds and words, use words that are more frequently-occurring in the language
  • (2) PHONOLOGICAL NEIGHBORHOOD DENSITY (PND) and PHONOTACTIC PROBABILITY (PP): Recall that a word’s “phonological neighbors” are those that differ from the word by only one phoneme (e.g. neighbors of ‘dog’ would include ‘fog, log, dot, dig…’). A “dense neighborhood” is exactly what it sounds like—lots of phonological neighbors. Phonotactic probability refers to the frequency of sounds or sound sequences in a language. The author points out that “PND and PP tend to be highly correlated”, so groups them together for consideration. Words that have a high phonological neighborhood density and high phonotactic probability should be used as stimuli.

The author also points out that using words that satisfy both criteria (1) and (2) are best.
Resources are provided within the article for websites with phonological neighborhood density calculators.
Also—even though it’s not a major focus of the article, the author also points to research on the use of phototactically legal nonwords in therapy. Even though these are a word frequency of zero, evidence indicates that they may have a unique benefit for phonological instruction (see Cummings & Barlow, 2011Gierut & Morrisette, 2010Gierut et al., 2010).

See: Sosa, A. (2016). Lexical Decisions in the Treatment of Speech Sound Disorders in Children. Perspectives of the ASHA Special Interest Groups, SIG 1, 1(2), 57–65.