Word frequency and neighborhood density are properties of lexical organization that differentially influence spoken-word recognition. This study examined whether these same properties also affect spoken-word production, particularly as related to children with functional phonological delays. The hypothesis was that differential generalization would be associated with a word's frequency and its neighborhood density when manipulated as input in phonological treatment. Using a multiple baseline across subjects design, 8 children (aged 3;10 to 5;4) were randomly enrolled in 1 of 4 experimental conditions targeting errored sounds in high-frequency, low-frequency, high-density, or low-density words. Dependent measures were generalization of treated sounds and untreated sounds within and across manner classes as measured during and following treatment. Results supported a hierarchy of phonological generalization by experimental condition. The clinical implications lie in planning for generalization through the input presented in treatment. Theoretically, the results demonstrate that lexical organization of words in the mental lexicon interacts with phonological structure in learning.

Lexical diffusion, as characterized by interword variation in production, was examined in phonological acquisition. The lexical variables of word frequency and neighbourhood density were hypothesized to facilitate sound change to varying degrees. Twelve children with functional phonological delays, aged 3;0 to 7;4, participated in an alternating treatments experiment to promote sound change. Independent variables were crossed to yield all logically possible combinations of high/low frequency and high/low density in treatment; the dependent measure was generalization accuracy in production. Results indicated word frequency was most facilitative in sound change, whereas, dense neighbourhood structure was least facilitative. The salience of frequency and avoidance of high density are discussed relative to the type of phonological change being induced in children's grammars, either phonetic or phonemic, and to the nature of children's representations. Results are further interpreted with reference to interactive models of language processing and optimality theoretic accounts of linguistic structure.

The efficacy of teaching sounds in developmental sequence as defined by age norms was evaluated in two independent investigations. Study I was a within-subject evaluation using an alternating treatments design, with three children each receiving treatment on one early-acquired and one later-acquired phoneme relative to chronological age. Study II was an across-subject evaluation involving six children in a staggered multiple baseline paradigm, whereby three subjects were each taught one early-acquired sound and three other subjects were taught one later-acquired sound relative to chronological age. Phonological change was measured on probes of sounds excluded from each child's phonemic inventory. General results indicated that: (a) quantitatively, change in treated phonemes and manner classes was equivocal following treatment of early-acquired and later-acquired phonemes; (b) qualitatively, the onset of change was immediate following treatment of later-acquired phonemes, but delayed following treatment of early-acquired phonemes; and (c) treatment of later-acquired phonemes led to system-wide changes in untreated sound classes, whereas treatment of early-acquired phonemes did not. These findings were considered relative to clinical intervention and theories of phonological acquisition.

The effect of word-level variables on expressive phonology has not been widely studied, although the properties of words likely bear on the emergence of sound structure (Stoel-Gammon, 2011). Eight preschoolers, diagnosed with phonological delay, were assigned to treatment to experimentally induce gains in expressive phonology. Erred sounds were taught using stimulus words that varied orthogonally in neighborhood density and word frequency as the independent variables. Generalization was the dependent variable, defined as production accuracy of treated and untreated (erred) sounds. Blocked comparisons showed that dense neighborhoods triggered greater generalization, but frequency did not have a clear differential effect. Orthogonal comparisons revealed graded effects, with frequent words from dense neighborhoods being optimal for generalization. The results contrast with prior literature, which has reported a sparse neighborhood advantage for children with phonological delay. There is a suggestion that children with phonological delay require greater than usual cue redundancy and convergence to prompt expressive phonological learning.

The effects of the age of acquisition (AoA) of words were examined in the clinical treatment of 10 preschool children with phonological delays. Using a single-subject multiple-baseline experimental design, children were enrolled in one of four conditions that varied the AoA of the treated words (early vs. late acquired) relative to their corresponding word frequency (high vs. low frequency). Phonological generalization to treated and untreated sounds in error served as the dependent variable. Results showed that late acquired words induced greater generalization, with an effect size four times greater than early acquired words, whereas the effects of word frequency were minimized. Results are discussed relative to hypotheses about the role of AoA in language acquisition and the relevance of this variable for phonological learning.