Recent research suggests that phonotactic probability (the likelihood of occurrence of a sound sequence) and neighborhood density (the number of words phonologically similar to a given word) influence spoken language processing and acquisition across the lifespan in both normal and clinical populations. The majority of research in this area has tended to focus on controlled laboratory studies rather than naturalistic data such as spontaneous speech samples or elicited probes. One difficulty in applying current measures of phonotactic probability and neighborhood density to more naturalistic samples is the significant correlation between these variables and word length. This study examines several alternative transformations of phonotactic probability and neighborhood density as a means of reducing or eliminating this correlation with word length. Computational analyses of the words in a large database and reanalysis of archival data supported the use of z scores for the analysis of phonotactic probability as a continuous variable and the use of median transformation scores for the analysis of phonotactic probability as a dichotomous variable. Neighborhood density results were less clear with the conclusion that analysis of neighborhood density as a continuous variable warrants further investigation to differentiate the utility of z scores in comparison to median transformation scores. Furthermore, balanced dichotomous coding of neighborhood density was difficult to achieve, suggesting that analysis of neighborhood density as a dichotomous variable should be approached with caution. Recommendations for future application and analyses are discussed.

The influence of language structure on the recognition, production, and acquisition of spoken language has received attention recently as a means of understanding how adults and children use regularities in the environment to process and acquire spoken language. One area of emerging interest involves regularities related to form, or sound, characteristics of language. Two form characteristics that have been shown to affect language processing and acquisition across the lifespan are phonotactic probability and neighborhood density. The majority of research on phonotactic probability and neighborhood density has tended to focus on controlled laboratory studies (eg, Aslin, Saffran, & Newport, 1998, 1999; Beckman & Edwards, 1999; Edwards, Beckman, & Munson, 2004; Gathercole, Frankish, Pickering, & Peaker, 1999; Hollich, Jusczyk, & Luce, 2002; Jusczyk, Luce, & Charles-Luce, 1994; Luce & Pisoni, 1998; Munson, Edwards, & Beckman, 2003; Saffran, Aslin, & Newport, 1996; Storkel, 2001, 2002, 2003; Storkel & Rogers, 2000; Vitevitch, 2002a; Vitevitch, Armbruster, & Chu, 2004; Vitevitch & Luce, 1998, 1999; Vitevitch, Luce, Charles-Luce, & Kemmerer, 1997), rather than more naturalistically collected samples such as probe samples or conversational speech (but see German & Newman, 2004; Gierut & Morrisette, 1998; Gierut & Storkel, 2002; Gordon, 2002; Morrisette, 1999; Newman & German, 2002; Storkel, 2004a; Storkel & Gierut, 2002; Vitevitch, 1997, 2002b). One stumbling block to analyzing the effect of phonotactic probability and neighborhood density on naturalistic samples is the correlation between these variables and other form characteristics, such as word length. The goals of this paper are to review previous studies of phonotactic probability and neighborhood density, to consider the problems associated with applying these constructs to the analysis of naturalistic data, and to examine potential solutions to this problem.