Speech development patterns and phonological awareness in preschool children
Mann, Virginia A
Received: 14 August 2006 /Accepted: 13 March 2007 /Published online: 24 May 2007
© The International Dyslexia Association 2007
To examine the association between speech production and early literacy skills, this study of 102 preschool children looked at phonological awareness in relation to whether children were delayed, typical, or advanced in their articulation of consonants. Using a developmental typology inspired by some of the literature on speech development (Kahn and Lewis, The Kahn-Lewis phonological analysis, 1986; Shriberg, Journal of Speech and Hearing Research 36(1):105-140, 1993a), we found that failure to master the early-8 consonants and a greater prevalence of certain types of production errors were associated with deficient phonological awareness. We also found that children who made no consonant errors had advanced phonological awareness relative to other children in the sample. In all cases, both productive speech patterns and speech errors were more closely linked with rhyme awareness than with phoneme awareness. The association between speech production and rhyme awareness may provide some new directions for the early preschool assessment of risk for reading problems.
Keywords Phonological awareness * Early literacy * Phoneme awareness * Rhyme awareness * Speech production
It is well established that the development of phonological awareness is closely tied to the development of alphabetic reading (Adams, 1990; Catts, 1989; Gottardo et al., 1996; Lyon, 1995; Muter & Snowling, 1998; Wagner & Torgesen, 1987). To the extent that phonological awareness depends upon mental representations of the phonological structure of language (Scarborough & Brady, 2002), one might expect the instantiation of phonological awareness to associate with the integrity of speech production and perception. Just as manipulating a word’s onset, rime or phonemes requires some means of internally representing its phonological structure, so do the language skills of producing and perceiving phonological structure.
To date, there has been growing support for a link between phonological awareness and speech perception. Echoing and extending an earlier statement by Brady and her colleagues (Brady et al., 1989) who suggested that differences between good and poor readers may lie in “the accuracy of formulating phonological representations” (p. 120), McBride-Chang (1996) and McBride-Chang et al. (1997) have argued for an influence of speech perception on reading that is moderated by phonological awareness. Directly bridging between deficient phoneme perception and phoneme awareness, Chiappe et al. (2001) have shown that variance in phoneme identification accounts for significant variance in phoneme deletion. They suggest that deficits in speech perception play a causal role in the deficient phonological processing of poor readers and that insufficiently differentiated phonological representations are a mediating link between speech perception and deficient phonological awareness. To date, the literature has not taken a concise position on the level at which phonological representation becomes problematic for the poor reader. Some research considers poor phonological representation in short-term memory processes (Brady et al., 1989; Mann et al, 1980) where other discussion centers on the mental lexicon (e.g., Elbro, 1996; Fowler & Swainson, 2004). It is presently unclear whether there are two different types of representational impairment or a single common denominator. In this study, we are neutral to this issue. We examine word production, vocabulary, naming speed, and judgments about rhyming words, all of which involve the mental lexicon. We also examine digit span, nonsense word repetition and the awareness of phonemes in nonsense words, which would place demands on short term memory. We shall return to this issue in the discussion.
In the present study, we focus on speech production in relation to the development of phonological awareness. Previous researchers have asked whether speech production problems are evident among children who are poor readers or those at significant risk by virtue of having an affected family member. They have also asked whether children with speech and language impairment are prone to reading problems, and they have examined the associations between individual differences in speech, reading, and phonological awareness. We use a combination of these approaches; we study the phoneme awareness of a population of children whom we classify as delayed, normal, or advanced according to the status of their articulatory development and we also study the ways in which these children’s phonological awareness skills more generally relate to the patterns of their misarticulations.
Previous studies of poor readers have implied problems with production, often in the face of other language problems (e.g., Dodd, 1995). Groups of poor readers have presented difficulties with the repetition of multisyllabic words (Snowling, 1981), nonwords (Snowling et al., 1986), and phonologically complex phrases (Catts, 1986). Problems with naming and other expressive language tasks are also reported in the literature on children at risk for reading problems (e.g., DeFries et al, 1999; Fowler & Swainson, 2004; Scarborough 1989). Children with speech-language problems have also been reported to have reading problems more often than the general population (Aram & Hall, 1989; Bishop & Adams, 1990; Catts, 1991, 1993; Lewis et al., 1989; Menyuk et al., 1991). In many of these cases, however, the basis of poor speech production is unclear. Children’s misarticulations could reflect a problem with articulatory control. They could reflect inadequate speech perception (e.g. “garbage-in-garbage-out”) or insufficient working memory resources (Gathercole & Baddeley, 1990). Problems with articulation, perception, and memory could also be part and parcel of a more general problem, internally representing the phonological structure of language (e.g., a pervasive problem with phonological representation, see Foy & Mann, 2001, 2003 for discussion).
With regard to the expressive language skills of children at familial risk for reading problems (DeFries et al, 1986, 1999; Gallagher et al., 2000; Scarborough, 1989, 1990, 1991), the Gallagher sample of children who developed reading difficulty (e.g., 57% of the familial risk group and 12% of a control group) showed impairments in expressive language and phonological awareness but also showed poorer vocabulary, short term memory, and letter knowledge. These impairments could not be traced to a weaker language and literacy environment (e.g., a parent who avoided reading) but were somehow inherent to the child. Compared to at-risk children who developed normal reading skills and to other children who did not have dyslexic family members, Scarborough’s familial risk subjects (DeFries et al., 1986, 1999; Gallagher et al, 2000; Scarborough, 1989, 1990, 1991) who later developed reading disabilities (e.g., 65% of the group) made disproportionately many productive phonological deficits during the preschool years and showed phonological (and syntactic) deficits at second grade. They also had significant receptive vocabulary and object naming difficulties. Similar results have been found for familial risk children in other studies (e.g., Byrne, 1996; Elbro et al., 1998; Lefly & Pennington, 1996). Working with a Finnish population of children at familial risk for reading problems, Lyytinen et al. (2001) have perhaps been the first to show that delayed language-development (e.g., being a late-talker) in a familial risk child is a particularly significant predictor of later reading problems.
With regard to reading problems among children with delayed expressive language, delayed talkers are disproportionately often poor readers (Catts et al, 2001; Lefly & Pennington, 1996; Menyuk et al., 1991) and have weaker phonological awareness (Hesketh et al, 2000; Webster & Plante, 1992a) and weaker speech perception (Evans et al., 2002; Joanisse et al., 2000; Sussman, 1993), although not all children with speech-language problems later develop reading difficulties (Bird et al., 1995; Bishop & Adams, 1990; Catts, 1993; Major & Bernhardt, 1998). One subgroup of children appears to be at unusually strong risk for reading problems, namely, children with expressive language problems that are unresolved before school entry (Bishop & Adams, 1990), especially those with concomitant receptive language problems (Catts, 1993) and concomitant expressive and receptive language impairments more generally (Leitao et al, 1997).
Children with unresolved expressive language impairment tend to employ fewer consonants than normally developing children (Paul & Jennnings, 1992; Ratner, 1994; Rescorla & Ratner, 1996; Scarborough & Dobrich, 1990; Whitehurst et al., 1991). Speech production accuracy (the number of phonemes pronounced correctly in an experimental task) in preschool children is linked to later phoneme awareness (Carroll et al, 2003). To us, this suggested the possibility of using the status of consonant production as an approach to clarifying the relation between speech production and phoneme awareness. We reasoned that, within this approach, two parameters could be considered. We could determine the set of consonants that children articulate, and we could describe their misarticulations. For example, many of the speech production errors of normal preschoolers (e.g., articulatory distortions) resolve spontaneously by school age or earlier in a predictable manner (Roberts et al., 1998; Shriberg et al., 1986). Most preschool children correctly produce ImI but many do not produce adult-like /s/ sounds. When they do misarticulate /s/, they tend to produce variants of the /s/ sound in terms of dimensions which are not phonemically salient in the target language, such as dentalized or lateralized productions of /s/. It is relatively uncommon for their misarticulations to involve a change in consonant manner, such as producing /t/ for /s/, a phonological error process that is referred to as “stopping”. A delay in mastering the age-appropriate set of consonants or a delay in outgrowing “normal” productive errors might reflect phonological problems that would lead to poor phonological awareness and reading difficulties; so might a more atypical tendency to persist in misarticulating some of the sounds that much younger children have mastered and a tendency to make atypical errors such as stopping.
The criteria as to what constitutes “normal” error patterns for children in the preschool years largely derive from two research approaches: typicality and disorder perspectives. The typicality perspective is taken by studies involving fairly large numbers of children recruited without regard to developmental problems (e.g., Porter & Hudson, 2001; Smit, 1993a, b). In these studies, error patterns that are infrequent are defined as “abnormal” (variously operationally defined, ranging from 15% to less than 5%) for a given age group. In contrast, the disorder perspective examines children recruited by virtue of having been diagnosed with expressive language/speech problems. It is taken by studies that make qualitative and quantitative contrasts between the speech errors of disordered group and children matched on various criteria such as age and gender (e.g., Roberts et al., 1998; Shriberg, 1993a, b). For the sake of clarity, we will use the phonological process names adopted by Shriberg and his colleagues (see Shriberg, 1993a; Shriberg et al, 1986, 1994, 1997, 2000; Shriberg & Kent 1995; Shriberg & Kwiatkowski, 1982, 1994) in discussing the results of these two approaches: deletion affinal consonants, syllable reduction, palatal fronting, velar fronting, consonant harmony, stopping of fricatives, and affricates (termed stopping), cluster simplification, and liquid simplification.
Typical acquisition of consonantal phonemes Surprisingly, few studies have examined speech production in large samples of normally developing children. These suggest that children tend to acquire productive control over most consonants by preschool age and that atypicality reflects delayed rather than deviant patterns of development. For example, Smit (1993a, 1993b) has shown that by age 4, at least 85% of the children in the IowaNebraska Articulation Norms Project were able to correctly produce the following sounds in initial and final position: /p, b, k, g, t, d, w, m, n, h, j/. They were less able to produce: /r, l, f, v, θ, ð, s, z, ∫, z, t∫, dz. No other consistent patterns were apparent. Porter and Hodson (2001) report that normative speech production in their sample (e.g., adequately produced in at least 85% of the opportunities) involved acquisition of all major phoneme classes by age 3 years, except for liquids (/1/ and /r/), which were most often acquired by age 5 years, and sibilants, which were acquired without lisps by age 7 years.1 Porter and Hudson’s error pattern analysis further revealed three types of consonant errors that were relatively atypical among preschoolers: omissions of singleton consonants (e.g., /da:/ for “dog”), omission of one consonant within a cluster (termed “cluster reductions”; e.g., /fæg/ for “flag”), and substitutions that involved either changes in manner or place of articulation (e.g., /t/ for /s/ or /d/ for /l/ for manner, /d/ for /g/ for place).2 Substitutions involving voicing were typical in their sample (2001), a finding which is mirrored in children’s errors on perceptual tasks (Treiman et al, 1998a).
In summary, studies of normally developing children suggest that by age 4, most children have acquired productive control over consonant manner and place of articulation, arguably the major bases for categorization of English phonemes. The exceptions to this are the liquids (/l/ and /r/), the strident phonemes (i.e., s, z, θ, ð, etc.), and clusters (e.g., /st/). Studies also suggest that omissions and inter-phoneme class substitutions that change manner or place are atypical phonological errors for preschoolers.
Disordered acquisition of consonant phonemes Studies of children identified as having abnormal speech-language development have pointed to a set of consonants that when misarticulated past a given age tend to give rise to the most profound abnormalities. Roberts et al. (1998) studied children whose speech development had been identified as delayed at 2 years (e.g., less than a 50-word vocabulary and no word combinations). By three years, close to half of the delayed children had caught up with their peers in verbalization rate (defined as the ratio of verbalizations to vocalizations). Those who did not catch up showed persistent problems that were unique in misarticulating some of the early-acquired phonemes /p, b, t, d, k, g, w, j/. Statistically significant differences were not found for error patterns; all children were most prone to consonant deletion, gliding, and stopping. Children also did not differ in the production of later developing phonemes.
Shriberg (1993a) has offered a similar comparison between preschool children with persistent articulation problems and those who obtained normal speech. However, his study, which involved children older than 3 years, employs a slightly different categorization of errors and obtains a slightly different result. He identified a three-step developmental sequence for acquisition of consonantal phonemes based on clustering in a rank-ordered sequence of percent correct consonants in speech-delayed children: the early-8, (p, b, j, n, w, d, m, h), the middle-8 (t, η, k, g, f, V, t∫, dz), and the late-8 (∫, θ, s, z, ð, l, r, z). Analogous to Roberts et al. (1998) and Shriberg et al. (1994) found that the only distinguishing characteristic between children whose speech did not normalize 1 year later (n=10) and those whose speech did normalize (n=44) was a nonsignificant trend for the non-normalized group to show lower performance on the early-8 phonemes. Shriberg et al. (1986) identified eight sound-change categories (natural phonologic processes) that describe over 90% of deletion and substitution errors of normally developing and speech-delayed children above 3 years (p. 145): errors of deletion (final consonant deletion, cluster simplification, unstressed syllable deletion) and substitution (stopping, liquid simplification, palatal fronting, velar fronting, and assimilation/consonant harmony). Shriberg and Kwiatkowski (1994) and Shriberg et al. (1994) also found no statistically significant differences in the use of these processes by speech delayed children whose speech normalized within 1 year compared to children whose speech did not normalize within 1 year.
In summary, studies of children who had early and persistent deficits in speech production indicate a trend towards select impairment on early developing phonemes during the preschool years. Patterns of phonological errors suggest no trend towards delayed children using more of some processes and less of others but do indicate what errors are typical and what ones are atypical.
Knowing what consonants are typically mastered by children in the preschool ages, and what ones are not, and what types of distortions can be expected, our question is whether children’s errors relate to their level of phonological awareness and bear any relation to other phonological skills such as speech perception or working/short term memory. Given that three year old children who have difficulty articulating the late developing phoneme /r/ show persistent problems with phonological sensitivity to this phoneme (Thomas & Senechal, 2004), one prediction in the present study involves the co-occurrence of articulatory problems and deficient awareness. ‘Articulatorily delayed’ children who persist in showing errors in the production of early developing sounds may do so because they are less able to represent the phonological structure of language. If so, the same weakness in representation may lead them to be less able to judge or manipulate phonological units as is required in rhyme or phoneme awareness tasks. Another prediction, unique to our study, involves the co-occurrence of advanced articulatory development and superior awareness. “Articulatorily advanced” children who exhibit more mature control over consonant articulation may do so because they possess relatively stronger abilities to represent phonological structure. If so, in association with this strength, they may also possess superior phonological awareness.
Studies of phonological awareness in children with phonological disorders consistently report that children with expressive phonological problems show deficits in at least some phonological awareness tasks, as compared to normally developing children (e.g., Bird & Bishop, 1992; Bird et al, 1995; Hesketh et al, 2000; Leitao et al., 1997; Webster & Plante, 1992a, b). For example, Hesketh and her colleagues studied children aged 3.5-5 years who had not previously received speech therapy but who had specific deficits in articulation (on the Edinburgh Articulation Test) in the face of normal scores on measures of language, vocabulary, nonverbal intelligence, and hearing. Compared to normally developing children, the speech disordered group had lower phonological awareness scores on a composite score for five tasks that did not require overt verbal responses (rhyme matching, word-initial matching, blending phonemes, word-initial segmentation and matching, and consonant deletion). The subtests that showed statistically significant differences were onset matching and word-initial segmentation and matching. Webster and Plante (1992a, b; 1995) and Webster et al. (1997) have consistently found that older children (aged 6 years 5 months to 8 years 6 months) with production errors that were not developmentally appropriate were also deficient in letter name knowledge (upper case) and phoneme awareness (but not rhyme awareness) compared to children who were developing normally. Preschool children with a history of speech sound disorders have been shown to have deficient phonological awareness and letter knowledge skills (Raitano et al, 2004). Recently, Rvachew and Grawburg (2006) showed that children with speech sound disorders (scored below 16th percentile of standardized measure of articulation skills) had lower phonological awareness skills than normally developing children and that vocabulary and speech perception skills successfully predicted phonological awareness development.
In this study, we examine phonological awareness and other pre-reading skills as a function of speech production errors. We use the Goldman-Fristoe Test of Articulation as scored from three vantage points: Shriberg’s typology, Porter and Hudson’s typology, and the Kahn-Lewis phonological processes used by Webster et al. (1997). We also examine children with advanced articulatory skills and those with persistent delays relative to normally developing peers, as we believe that it can be an important research strategy to sample the entire continuum of development and not just the normal-to-impaired.
In our prior research, we have been concerned with the separability of rhyme and phoneme awareness (for details, see Foy & Mann, 2001, 2003, 2006; Mann & Foy, 2003), and this has shaped our decision in the present paper to examine phonological awareness in terms of both rhyme and phoneme. As in prior work, we also continue to employ more than one task as a measure of awareness, including those that require judgment responses and those that require more complex articulations. A growing body of evidence (e.g., Bertelson et al., 1989; Hulme, 2002; Hulme et al, 2002; Morais et al., 1986) suggests that phoneme awareness and rhyme awareness are separate processes that make differential contributions to reading achievement. Our prior findings (Foy & Mann, 2001, 2003) accord with such evidence in suggesting that the awareness of rhyme is more closely aligned with phonological perception and production abilities where the awareness of phonemes relates to literacy and educational exposure (Foy & Mann, 2001, 2003).
In addition to measuring production and awareness, we measure several other language and literacy skills. Given that our preschool subjects are too young to be reading, we examine the pre-reading skills of letter name and sound knowledge. As we are concerned with possible explanations of a link between production and awareness, we measure some additional aspects of spoken language skill that might either be causes of a production impairment or symptoms of a pervasive problem with representing of the phonological structure of language. We measure the receptive skill of speech perception, the productive skill of naming, and three skills that involve both reception and expression and might possibly confound our interpretation of the results, namely, vocabulary, nonword repetition (Edwards & Lahey, 1998; Manis et al, 1997), and short term memory. All of these additional skills have a history of being related to reading ability (for review see Foy & Mann, 2001, 2003), although our prior work (Foy & Mann, 2001, 2003, 2006; Mann & Foy, 2003) has also offered little support for the conception that a unitary problem with the representation of phonological structure is the source of reading, speech, and phonological awareness problems.
Specifically, we test the hypotheses that:
1) Speech production will be linked with measures of early literacy skills.
2) Patterns of consonant errors will predict speech perception, vocabulary, naming, and digit span, and their relation to reading and phonological awareness insofar as they relate to the representation of phonological structure. In particular, atypical errors may associate with weaker phonological awareness, reading, and language skills.
Participants consisted of 102 children (52 girls and 50 boys) attending seven preschool programs or daycare programs for preschool-aged children in Southern California. The sample consisted of Caucasian, African American, Hispanic, Asian, and mixed-race children. Approximately half of the sample was non-Caucasian. The children (aged 4-6 years) were from low to upper middle class families who spoke English as their primary language. We excluded children with structural/organic deficits such as tongue thrust that may have affected speech production from the study. To be assured of including children who were at risk for reading problems, we made efforts to include a cohort of children with a family history of reading problems (i.e., parent or sibling with known reading problems). This effort led to the inclusion of 13 children at familial risk for reading problems. The at-risk children, who were all native speakers of English, were compared to 13 children matched on age, sex, bilingual status, and vocabulary, and showed no significant differences on any of these measures. A subsample of 70 children were reexamined 3 months later. We report here only the phoneme awareness and rhyme awareness measures at time 2. All of the children were reported by the parents/caregivers and teachers to be developing normally.
All of the preschools had same-age classroom groupings. None of the preschools explicitly taught “reading”, and none taught phonological awareness. Letter name and letter sound training varied from classroom to classroom, including classrooms where the only exposure to letter names involved computer games the children could choose to play if they so wished, and classrooms where letter names and sounds were introduced during the year before kindergarten entry.
Six children, who all attended one of the preschools, were bilingual3 (Spanish). All of the interactions with the staff at this preschool were in English. The final sample included 52 4-year olds, 45 5-year olds, and 5 6-year olds. Only children whose parents consented to participation and who themselves assented were included in the study. The study was conducted in February/March of the school year.
Assessment of reading using the Word Identification (real words) and Word Attack (nonwords) subtests of the Woodcock Reading Mastery Test-Revised (Woodcock, 1987) revealed that none of the children were able to read more than two words. We thus assumed that our sample consisted of nonreaders.
The Sounds-In-Words subtest of the Goldman-Fristoe Test of Articulation (GFTA) (Goldman & Fristoe, 1986), a standardized test of articulatory skill was administered to each child, with the responses transcribed phonetically on-line, tape-recorded, and later analyzed. The test contains 35 simple color stimulus pictures intended to elicit 44 responses. Internal reliability reported by the authors is .96 for females, and .94 for males. Test-retest reliability is .98 for initial, medial, and final sounds. Median percentages of inter-rater agreement for initial, medial, and final sounds are reported at 93, 90, and 90, respectively (http://ags.pearsonassessments.com/assessments/technical/gfta.asp).
The responses elicited from the children were audiotaped and phonetically transcribed by a licensed and certified speech-language pathologist. The samples were then rechecked by the researcher or a research assistant using the audiotapes. Any differences between the transcript analyses were resolved through repeated listening to the taped response. Twenty percent of the transcripts were randomly selected for analysis by an independent research assistant blind to the research hypotheses. The percentage of agreement of the consonants transcribed for each word was calculated. The average percent of agreement for the transcripts was 97.6% (range 95.2-100%). “Articulation” reflected the number of errors made on phonemes identified for testing in the Goldman-Fristoe Test of Articulation.
Speech delay analysis
Shriberg (1993a) identified a normal developmental sequence for acquisition of consonantal phonemes based on clustering in a rank-ordered sequence of percent correct consonants in speech-delayed children. These are the early-8 (p, b, j, n, w, d, m, h), middle-8 (t, η, k, g, f, v, t∫, d3), and the late-8 (∫, θ, s, z, ð, 1, r, 3). Errors made on the early-8 sounds (early-8 errors), middle-8 sounds (middle-8 errors), and late-8 sounds (late-8 errors) were subjected to separate analyses and are illustrated in Fig. 1 (after Shriberg, 1993b). We also included a category for errors on clusters (cluster errors).
Speech error analysis
Speech production errors were scored in two ways, reflecting perspectives in the literature.
Porter and Hudson’s typology Porter and Hodson (2001) found that singleton (single consonant) omissions, cluster (consonants occurring together, e.g., /s/ and /t/ in “stop”) reductions, and inter-phoneme class substitutions were atypical phonological errors in their preschool sample. We scored these as omissions, reductions, and substitutions, respectively. Substitutions of /ð/ and /θ/ (lisps) for /s/ and /z/ were not scored as inter-class substitutions by Porter and Hodson, as they were prevalent, and thus, were also excluded from our scoring of substitutions.
Kahn-Lewis phonological analysis The Kahn-Lewis Phonological Analysis (KLPA) (Kahn & Lewis, 1986) is designed as a speech production error analysis tool to be used with the GFTA (p.33). It assesses the use of phonological processes in responses containing speech production errors. Of these processes, 12 are considered developmentally appropriate for children aged 2-6 years according to the test’s standardization sample. The processes (as shown in Table 1) consist of all of the processes identified by Shriberg and Kwiatkowski (1982, 1994): deletion of final consonants, syllable reduction, palatal fronting, velar fronting, consonant harmony, stopping of fricatives, and affricates (termed stopping), cluster simplification, and liquid simplification. The Kahn-Lewis analysis also includes the additional processes of initial voicing, deaffrication, stridency deletion, and final devoicing. Due to the floor effects for some of the categories, we combined the normal processes into one category representing the sum of the 12 corresponding developmentally normal phonological processes. According to the Kahn-Lewis analysis, the following phonological processes are examples of processes that are not characteristic of normal phonological development and are termed non-developmental processes: deletion of initial consonants, glottal replacement, and backing to velars. Any other processes not among the 12 developmental processes are also considered nondevelopmental processes for the purposes of our analysis. The Nondevelopmental Processes scores represent the sum of the phonological processes that are not considered normal (i.e., the sum of the deletion of final consonants, glottal replacement, backing of velars, and all other phonological processes not among the 12 normal processes, as shown in Table 1).
Expressive language measures
Verbal short-term memory The Digit Span subtest (forward) of the WISC-R (Wechsler, 1992) provided a measure of verbal short-term memory, which has been linked to reading achievement and early reading skills (e.g., Marshall et al., 2001; Snowling et al., 1994), including phonological representations (Fowler, 1991).
Vocabulary The WPPSI Vocabulary subtest (Wechsler, 1992) was used as a measure of expressive vocabulary previously shown to correlate significantly with reading in school-aged children (Mahony et al., 2000; Singson et al., 2000). In this test, children are asked to give definitions for aurally presented words of increasing difficulty.
Early literacy skills
Letter knowledge The letter identification and letter sound subtests of the Concepts about Print Test (Clay, 1979) were administered. This test involves identification and naming of all upper and lower case letters in random order. Letter sound knowledge was assessed by re-administering the letter stimuli and asking children to provide the sound associated with each letter. The tasks were discontinued after eight consecutive failures, with the exception of letters in the child’s first name, all of which were then tested. In addition, letter name and letter sound knowledge was assessed for four clusters, as in Mann and Foy (2003). The letter-naming score reflects the summed scores on the letter identification tests for upper and lower case letters and cluster subtests. The letter sound score reflects the summed scores on the letter sound tests for upper and lower case letters, and the cluster sounds subtests. Letter name knowledge was assessed before letter sound knowledge, separated by several other tasks.
Phoneme awareness The materials consisted of practice trials and test items for each of six subtests assessing phoneme judgment, phoneme deletion, and phoneme substitution in both initial and final positions. These tests were previously used in Mann and Foy (2003); the two rhyme awareness tasks and the phoneme judgment task present real words as stimuli where the phoneme deletion task presents nonsense words. The subtests each consisted of two practice items and five test items. The tests were administered in standard order: phoneme judgment, phoneme deletion, and phoneme substitution. Raw scores on the phoneme judgment, phoneme deletion, and phoneme substitution segments of each test were summed to provide a single score for phoneme awareness.
In the phoneme judgment tests, the examiner presented a stimulus word, followed by two test words, and the children responded with the word that started (initial) or ended (final) with the same sound as the target word. This test requires the child to articulate but places minimal stress on the quality of the articulation as the child can misarticulate the matching word and still receive a correct score.
In the phoneme deletion tasks, the children were told to say what happens to words when the first (initial) or last (final) sound was taken out. After demonstration and practice, the children responded by indicating how the word would sound when the target sound was removed from each word. In the phoneme substitution tests, the children were told that the examiner liked the sound /k/, and were invited to change nonsense words by changing the first (initial) or last (final) sound to /k/. Following demonstration and practice, the children responded by changing the nonsense words into nonsense words that began (initial) or ended (final) with /k/. Phoneme deletion and phoneme substitution presumably place greater demands on articulatory control and require more robust phonological representations than phoneme judgment tasks.
Rhyme Awareness As in Mann and Foy (2003) the composite rhyme awareness score (RA) was derived by summing the raw scores on two rhyming tasks: rhyme recognition and rhyme production. In the rhyme recognition task, adapted from (Chaney, 1992), children saw three pictured objects at a time, two of which had names that rhymed. The examiner named the three objects, and pointed to them. The children were asked to point to the pictures that “rhymed” or “sounded almost the same.” After demonstration and three practice trials, the children indicated their responses to eight trials by pointing. This task required no articulation. In the rhyme production task, the children were asked to say, “what word rhymes with _____” for five trials consisting of common words (e.g., hop). This task did require articulation but either words or nonwords were scored as correct as long as they rhymed with the target word.
Additional phonological representation measures
Naming speed An English language adaptation (Foy & Mann, 2001) of Elbro’s naming task (see Elbro et al., 1998) was used to measure picture-naming speed. It is a simple naming task with color pictures taken from magazines. These names are within the vocabulary of 5-year-old children. Pictures from the same semantic category (e.g., chair, sofa, table) are presented three at a time on a single card, and the child is asked to name the objects depicted in the pictures as quickly as possible. The test has two trial items and 15 test items. If the child failed to name a picture, misnamed a picture, or took longer than 15 s to name the items on the card, the data from that item were disregarded in the naming data. The individual scores are average naming time in seconds.
Nonword repetition The modified Children’s Test of Nonword Repetition (Gathercole et al., 1994) was used to assess nonword repetition ability. To shorten the task, only the first five nonwords from two-syllable, three-syllable, and four-syllable nonwords were administered to the children. According to Gathercole et al. (1994), the phoneme sequences in each stimulus nonword conformed to the phonotactic rules of English, and within each number of syllables, the items were constructed to correspond to the dominant syllable stress patterns in English for words of that length (strong-weak for two-syllable nonwords, and strong-weak-weak for the three-syllable nonwords, and variable stress patterns for four-syllable words). The phoneme sequences for the nonwords thus are phonotactically and prosodically legal. Test-retest reliability was reported at .77. Pronunciation was modified for the American sample according to pronunciation by ten normally reading adults (see Foy & Mann, 2001). Online scoring has been previously reported at agreement on 97% of the items. Deletions, substitutions, and additions were all scored as errors. Percentage of correct words was calculated.
Speech perception was assessed in a computerized task using synthesized stimuli generated using the CSLU Speech Toolskit with a sampling rate of 16,000 samples/s. The target stimuli were derived from the Goldman-Fristoe-Woodcock Test of Auditory Discrimination (Goldman et al., 1970) and consisted of 12 minimal pairs of words that included the following contrasts in the initial position in words common to the vocabularies of young children (Mann & Foy, 2003). The word pairs contrasted voicing (for example, /b/ vs /p/, /t/ vs /d/), place of articulation (for example, /b/ vs /d/, /t/ vs /p/), and manner of articulation (for example, /S/ VS. /∫/). Each task (quiet and noise) consisted of 21 randomly ordered trials, with order within each pair also randomly determined for each participant. In the “quiet” condition, participants listened to the stimuli via noise-canceling earphones. In the “noise” condition, participants heard the same stimuli in a different order, also wearing the earphones, with the stimuli masked by white noise (0 SNR) 440 ms preceding and during the duration of the speech sound presentation. The quiet condition always preceded the noise condition, and in both conditions, pointing responses were recorded by an experimenter who was blind to the speech pair condition. Ambiguous responses (e.g., pointing in the center of the screen or at both sides of the screen simultaneously), failure to respond, and incorrect responses were scored as errors.
Participants were tested individually in quiet testing rooms on the school premises. They were rewarded with stickers as needed to ensure maximal motivation and attention. Testing was conducted in two sessions each lasting approximately 30 min and usually conducted on two separate days. The tests were presented in a fixed order: (session one) Vocabulary, phoneme and rhyme awareness, letter names, digit span, rapid naming, letter sounds, reading, and (session two): articulation, speech perception, nonword repetition.
Before analysis, the data were examined for missing values, fit between their distributions and the assumptions of multivariate analysis (Tabachnick & Fidell, 2001, pp. 98-99). To correct for the positive skewness and kurtosis in most of our measures, we used nonparametric statistics (Spearman correlations and Kruskal-Wallis analyses) to examine relationships between variables. These tests are appropriate statistics to use when variables have equivalent but non-normal distributions (Norusis, 2000). Regression analyses were used to examine predictive relations between speech production measures and our reading related-measures.
Hypothesis #1: Speech production will be linked with measures of early literacy skills
Spearman correlations revealed that the number of early-8 errors was significantly correlated with vocabulary, digit span, letter knowledge rhyme awareness, naming speed, and speech discrimination, as illustrated in Table 2. Middle-8 errors were also associated with these variables, but in addition, were associated with phoneme awareness and nonword repetition. Late-8 errors were significantly correlated with vocabulary, digit span, letter knowledge, rhyme awareness, naming speech, nonword repetition, and speech discrimination. Cluster errors were only associated with nonword repetition accuracy.
For the purpose of analysis, we then divided the participants into three groups:
(1) The Delayed Group (n=25): children who had at least one error deficits on the Shriberg early-8 sounds compared with children who made no errors on these sounds. Of the children at familial risk for reading problems, 38% (5/13) made errors on the early-8 sounds, compared to 17% (10/58) of the children not at familial risk.
(2) The Typical Group (n= 65): children who had no errors on the early-8 sounds, but who made errors on the late-8 sounds. Some of these children had deficits on the middle-8 sounds but some did not. Forty-six percent (6/13) of the children at familial risk for reading problems were members of this group compared to 72% (42/58) of the children not at familial risk.
(3) The Advanced Group (n-12): children who had no deficits on the early-8, middle-8, or late-8 sounds. Ten of these children also had no errors on the Goldman-Fristoe Test, including consonant clusters, which Shriberg did not include in his speech norms. Fifteen percent (2/13) of the children at familial risk for reading problems were members of this group compared to 10% (6/58) of the children not at familial risk.
To further examine the relationship between production of consonants (e.g., Shriberg’s developmental norms) and our major variables, we conducted Kruskal-Wallis tests of the dependent variables according to our three comparison groups: Delayed vs typical vs advanced, and then followed up significant findings with two-group comparisons using two-group Kruskal-Wallis tests (see Table 3). We had hypothesized that patterns of speech production skills would be linked with reading-related measures. Results of Kruskal-Wallis tests revealed significant differences between the groups for expressive vocabulary (H=7.31, p=.026), verbal short term memory (H=12.13, p=.002), rhyme awareness (H=15.85, p=.0001), letter knowledge (H=12.94, p=.002), speed of naming responses (H=6.41, p=.041), and accuracy of nonword repetition (H=14.38, p=.001), and speech discrimination (H=8.34, p=.015). Follow-up two-group comparisons to further explore differences between the delayed/advanced groups and the typical group, as illustrated in Table 3, showed that children in the Delayed group achieved significantly lower scores on expressive vocabulary (H=4.712, p=.03), verbal short term memory (H=7.19, p=.007), letter knowledge, (H=11.09, p=.001) rhyme awareness (H=7.76, p=.005), letter knowledge (H=11.09, p=.001), speech discrimination (H=6.30, p=.012), and slower naming responses (H=4.02, p=.045) than children in the typical group. Children in our advanced speech development group had significantly higher scores than the typical children for rhyme awareness (H=4.80, p=.028), and nonword repetition accuracy (H=12.13, p=.0001) and higher scores for vocabulary (H=6.11, p=.013), verbal short-term memory (H=9.62, p=.002), rhyming (H=14.30, p=.0001), letter knowledge (H=8.07, p=.004), speech discrimination (H=6.03, p=.014), nonword repetition accuracy (H=12.14, p=.0001), and faster naming responses (H=5.09, p=.024) than our delayed group.
Hypothesis #2: Consonantal errors are related to reading related measures
The most common developmental phonological processes used by the children in the present sample were, in descending order of frequency: stridency deletion (22% of the phonological processes used by the children), stopping (15.29%), liquid simplification (14.41%), cluster simplification (9.79%), palatal fronting (4.70%), and consonant harmony (4.07%). The normal developmental processes that occurred very infrequently in our sample were initial voicing (2.09%), final devoicing (1.87%), velar fronting (1.54%), deletion of final consonant (.7%), and syllable reduction (less than .1%). The most common “nondevelopmental” error we observed was addition of stridency (11.88% of the phonological processes). Initial and medial devoicing errors occurred in only 1.1 and 1.2%, respectively, of the processes. The following occurred less than 1% of the total number of phonological processes used by the children: frication and deletion of medial consonants, sound preferences, lateralization, unknown errors, stopping of liquids, backing to velars, addition of consonants, final voicing and medial voicing, nasalization, liquidization, deletion of final consonants, glottal replacement, dentalized, and laterialized productions.
The Kruskal-Wallis test for three-group comparisons (delayed vs typical vs advanced), revealed significant differences between the groups for Total Errors (H=33.70, p=.0001), Developmental Processes (H=31.36, p=.0001), Nondevelopmental Processes (H=17.00, p=.0001), Deletions (H=8.85, p=.012), and Substitutions (H=28.37, p=.0001).
Follow-up two-group comparisons to further explore differences between the delayed/ advanced groups and the typical group, as illustrated in Table 4, showed that children in the advanced group had significantly lower scores on Total Errors (H=29.70, p=.0001), Developmental Processes (H=27.51, p=.0001), Nondevelopmental Processes (H=15.28, p=.0001), Deletions (H=5.29,p=.021), and Substitutions (H=21.16, ^=.0001) than typical children. Typical children has significantly lower scores on Substitutions (H=4, p=.046), than delayed children, who also had significantly higher scores on Total Errors (H=24.49, p=.0001), Developmental Processes (H=24.63, p=.0001), Nondevelopmental Processes (H= 14.28, p=.0001), Deletions (H=8.13, p=.004), and Substitutions (H= 24.59, p=.0001).
Predicting reading-related measures from speech production Next, we examined which of the speech production process measures best predicted the reading related measures by conducting three separate regressions: (1) regressing the consonant type errors (early, middle, late, and cluster errors) on rhyme and phoneme awareness (separately), (2) regressing process types (substitution vs deletion errors) on rhyme and phoneme awareness (separately), and (3) regressing error type (developmentally normal vs nondevelopmental) on rhyme and phoneme awareness (separately).
The results showed that of the consonant type errors (early, middle, late, and clusters), only early-8 errors (β=-.30, p=.004) significantly predicted rhyme, adjusted R^sup 2^=.101, p=.006. None of these errors significantly predicted phoneme awareness. In a further test of this hypothesis, we also regressed these measures on the rhyme and phoneme awareness measures from the second testing (3 months later), revealing similar results in that early-8 errors alone (β=-.34, p=.006) predicted rhyme awareness at time 2, adjusted R^sup 2^=.089, p=.039, and that none of the consonant type errors predicted phoneme awareness at time 2.
The regression analyses revealed that process error types (substitutions vs deletion) failed to significantly predict rhyme or phoneme awareness.
The final way we distinguished speech production errors was by type (developmentally normal vs nondevelopmental). The results showed that nondevelopmental processes significantly predicted rhyme awareness at time 1 (β=-.28, p=.007) and time 2 (β=-.25, p=.036), adjusted R^sup 2^=.084, p=.005 and adjusted R^sup 2^=.067, p=.037, respectively, but these processes did not significantly predict phoneme awareness either at time 1 or time 2.
To control the effects of age on our results, we conducted hierarchical regressions, first entering age, and then examining whether the speech errors significantly predicted rhyme awareness and phoneme awareness. The results were unchanged; rhyme awareness was significantly and negatively predicted by nondevelopmental processes and by early-8 errors, even when age was controlled.
Finally, to examine the contribution of age, verbal short-term memory, and expressive vocabulary to these relationship, we entered age, digit span, and vocabulary separately into the first block, and then examined whether controlling these variables accounted for any of the significant relations between nondevelopmental errors and early-8 errors (β=-.25, p=.016) on rhyme awareness. We found that digit span emerged as a significant independent predictor of rhyme awareness (β=.20, p=.054), and removing its effects effectively removed the variance reflecting the relationship between nondevelopmental errors and rhyme awareness and in the relationship between early-8 errors and rhyme awareness. Vocabulary also emerged as an independent predictor of rhyme awareness (β=.342, p=.041), but only when developmental/ nondevelopmental error effects on rhyme awareness were examined. When the early-8, middle-8, late-8 and cluster errors were entered into the analysis after the effects of vocabulary were removed, early-8 errors continued to independently predict rhyme awareness (β=-.18, p=.048).
The goal of this study was to test two hypotheses about how patterns of speech articulation among a group of preschool children would be linked with measures of phonological awareness. We predicted a link on theoretical grounds: Speech production and phonological awareness both demand internal representations of phonological structure. Converging support from research with normally developing children and children with reading and speech disorders, and children at familial risk for these problems, had each provided some empirical evidence for our predictions. We decided to test them more completely by categorizing and analyzing consonant errors and to consider both negative and positive extremes of mastery. Thus, we considered not only children who had not mastered speech production of sounds typically achieved at 4 years but also those who were advanced for their age. We examined not only where the errors occurred (e.g., the early-8) but also whether these errors were developmentally normal (e.g., stopping, cluster reduction) or atypical of normal children at this age, and thus, “nondevelopmental” in nature. We began by classifying children’s errors in terms of what is typical for children above age 3 years, using the early-, middle-, and late-8 categories developed by Shriberg. We classified children as “delayed” if they made errors on consonants within the early-8, “typical” if their errors exclusively involved the middle- and late-8 consonants, and “advanced” if they made no errors at all. This allowed us to test the hypothesis, which we confirm, while controlling for the confound of age as a possible source of a link between articulation and speech sound awareness.
If the pace of speech development reflects the strength of phonological representations, as the literature suggests, we reasoned that children who persist in showing errors in the early developing sounds should demonstrate weaknesses in other skills that draw upon phonological representations. Our results support this hypothesis, but differences were stronger and more consistent for rhyme awareness than for phonological awareness. Consistent with previous findings (Foy & Mann, 2001, 2003, 2006; Mann & Foy, 2003; Muter, 1994), we have once again found that rhyme and phoneme awareness can be dissociated. They appear to involve different concomitants and are differentially associated with very early reading abilities. Performance on the phoneme awareness tasks was lower than on the rhyme task in our sample of nonreaders. A different pattern of results might be obtained in a sample of readers or older children who have stronger phoneme awareness skills. Furthermore, the older children in our sample may have had more exposure to instruction in alphabet knowledge than the younger children, confounding our results. Lack of control over instructional practices leads us to couch our conclusions about phoneme vs rhyme awareness relations somewhat tentatively.
A major innovation of our study was the inclusion of children who were advanced articulators in making no errors on our consonant inventory. Examining phonological skills among such children and showing that they are significantly advanced in rhyme awareness compared to their peer group provides an important complement to the study of deficient children and further strengthens the point that it is rhyme awareness that is most directly tied to articulation. We are able to distinguish superior rhyme awareness as a characteristic of children who exhibit early control over phonemes that are typically late to develop (e.g., the late-8). Although there is emerging evidence that advanced early speech perception skills predict later language development (Kühl, 2004; Kühl et al, 2005), our approach in considering advanced production skills is unprecedented in the literature.
What remains to be seen is whether articulatorily advanced children will go on to become superior readers, owing to their phonological production advantages and increased rhyme awareness. Mann and Singson (2000) have shown that the most salient differences between a group of advanced readers and nonreaders involve phoneme awareness and onset-rime awareness (but see Fletcher-Flinn and Thompson, 2000, for a case study of precocious reading in the absence of phoneme awareness). Stainthorp and Hughes (1998) found that children who learned to read by school-age had shown early and sustained advantages in rhyme awareness and letter knowledge. Although the advanced readers had higher levels of phoneme awareness than age and vocabulary-matched control initially and after school immersion, Stainthorp and Hughes (1998) showed that the relationship between phoneme awareness and reading was best described as reciprocal where the effect of rhyme awareness was causal.
Our second hypothesis about production and phonological awareness had proposed that specific patterns of phonological errors-the developmental nature of the speech production errors-would relate to phonological awareness and possibly to reading and language more generally. To this end, we investigated the pattern of children’s production errors (their “phonological processes” in the terms of Shriberg et al. (1986). According to the literature, the use of developmentally normal “processes” such liquid simplification, stopping and cluster simplification processes should be common in a preschool sample, whereas aberrant “nondevelopmental” processes such as backing to velars and deletion of initial consonants should be relatively rare. As we had predicted, errors that were developmentally not common were more strongly associated with weaker awareness, but only for rhyme awareness. Nondevelopmental errors may reflect phonological representations that are less well specified (see Elbro et al., 1998 for further discussion). Again, the fact that these processes were related to both lexical level and short-term memory problems suggests that the representation problems occur at more than one level.
That nondevelopmental errors were significantly associated with speech skills and rhyme awareness could be due to our focus on an older population than several of the previous studies: We had studied 4- to 6-year olds where Roberts et al. studied 3-year olds and Shriberg (1993a, b) studied 4-year olds. It may reflect our inclusion of a larger number of children whose speech had not normalized by age 4 to 6 years (e.g., 25 as compared to the 10 studied by Shriberg et al., 1994). It may also reflect our inclusion of children at familial risk for reading problems, given that delayed speech is an added risk for reading problems among such children (Lyytinen et al., 2001).
We had chosen to study the relation between speech production skills and early literacy skills because logic would suggest that they have a common involvement with phonological representations. Our results suggest that while the phoneme awareness measures were not apparently related to the articulatory measures, there were consistent associations between rhyme awareness and articulation. We take the association between perception, production, and awareness as consistent with a problem with the internal representation of phonological structure.
In our sample, short-term memory may be a moderator of the link between articulatory skill and rhyme awareness, supporting Gathercole and Baddeley’s (1990) postulations about the interconnectedness of working memory and articulation. Expressive vocabulary also may be related to rhyme awareness in our sample, consistent with the views of Walley and her colleagues about the role of vocabulary expansion in phonological representation. Walley (1993) has suggested that vocabulary growth plays a role in phoneme awareness by forcing representations to become more phonemic and segmentai as lexical neighborhoods increase in density. In turn, this may make it easier for children to draw associations between orthographic and speech-sound connections and may support clearer articulations of manner and place. Our finding that both short-term memory and vocabulary mediate the relation between rhyme and articulatory skill suggests that pervasive problems with phonological representations are implicated in this relation, as they have been in reading problems. Our results are consistent with a single problem with phonological representation, although not inconclusive, and a rigorous test will have to await future research. Measures of articulation, perception, and rhyme awareness were clearly intertwined in our population of 4- to 6-year olds, and phoneme awareness was most clearly associated with other products of literacy experience (Foy & Mann, 2003; Mann & Foy, 2003; Mann & Wimmer, 2002). Our previous studies of phonological awareness among preschool children had also linked rhyme awareness to speech skills and phoneme awareness to literacy experience (Foy & Mann, 2001, 2003, 2006; Mann & Foy, 2003). The fact that children who had trouble with the early-8 also had inferior letter knowledge may help to explain why Webster and Plante (Webster & Plante, 1992a, b, 1995; Webster et al., 1997) have consistently found that 6- to 8-year-old children with production errors are deficient in phoneme awareness (but not rhyme awareness). Poor speech skills and poor rhyme awareness may set the stage for poor letter knowledge, and in turn, poor phonological awareness. This would accord with Treiman’s views that learning to read and write letters may have a reciprocal effect on the development of phonological awareness (e.g., Treiman et al., 1998b).
While our primary purpose in this study was to examine patterns of articulation in relation to young children’s phonological awareness and pre-reading skills, an incidental contribution of our study has been to provide additional support for developmental typologies of phonological errors, most notably using Shriberg’s approach to typical development (e.g., his normal vs nondevelopmental processes). We would argue, on the basis of our data, that further research, using large sample sizes and phonological representation measures as covariates, would continue to validate these approaches and their utility as screening devices for reading problems. Our findings suggest that, for those concerned with a child’s prospects for literacy development, the study of speech production errors is most useful when both typical (e.g., mastery of the early-8) and atypical (nondevelopmental) phonological aspects of preschool speech development are taken into account, as both the locus and nature of errors can be indicative of delayed and advanced development. We may not have identified a single factor that can encompass all of speech and literacy development, but we have elucidated some interconnections that can provide directions for the early preschool assessment of risk for reading problems.
Acknowledgments The authors wish to thank the children and their parents who participated in this study, and the preschool administrators who made the research possible. The authors also gratefully acknowledge the assistance with testing, scoring, and reliability assessments of the following Loyola Marymount University students: Aqila Blakey, Jessica Flores, Noemi Mai, and Elva Rios.
1 It is important to note that Porter and Hodson used the APP-R analysis procedure for scoring speech production errors. This procedure “allows credit for productions within phoneme boundaries” (p. 169), which would allow scoring the use of/s/ for /z/ or /J/ as correct, although the error pattern analysis would categorize this production as a “substitution within phoneme classes” (see below). The /θ/ /ð/, and /h/ phonemes, as well as the affricates (/t∫,dz/) were not examined in the Porter and Hodson study.
2 Phonemes in English are classified along three dimensions: manner, place, and voicing. Substitutions involving voicing are not considered atypical for two primary reasons: judgments about voicing vary with context, and consonant devoicing is considered a characteristic of developing speech (Shriberg & Kent, 1995, p.79).
3 In order for a child to be classified as bilingual, the teacher or the parent had to report that the child spoke this language primarily at home and was somewhat fluent in this language. All of these children were also fluent in Enelish.
Adams, M. R. (1990). Beginning to read: Thinking and learning about print. Cambridge MA: MIT Press.
Aram, D., & Hall, N. (1989). Longitudinal follow-up of children with preschool communication disorders: Treatment implications. School Psychology Review, 18, 487-501.
Bertelson, P., de Gelder, B., Tfouni, L. V., & Morais, J. (1989). Metaphonological abilities of adult illiterates: New evidence of heterogeneity. European Journal of Cognitive Psychology, 1(3), 239-250.
Bird, J., & Bishop, D. V. M. (1992). Perception and awareness of phonemes in phonologically impaired children. European Journal of Disorders of Communication, 27, 289-312.
Bird, J., Bishop, D. V. M., & Freeman, N. H. (1995). Phonological awareness and literacy development in children with expressive phonological impairments. Journal of Speech and Hearing Research, 38, 446-462.
Bishop, D. V. M., & Adams, C. (1990). A prospective study of the relationship between Specific Language Impairment, phonological disorders and reading retardation. Journal of Child Psychology and Psychiatry and Applied Disciplines, 31(7), 1027-1050.
Brady, S., Poggie, E., & Rapala, M. M. (1989). Speech repetition abilities in children who differ in reading skill. Language and Speech, 32(2), 109-122.
Byrne, B. (1996). The leamability of the alphabetic principle: Children’s initial hypotheses about how print represents spoken language. Applied Psycholinguistics, 17(4), 401-426.
Carroll, J. M., Snowling, M. J., Hulme, C., & Stevenson, J. (2003). The development of phonological awareness in preschool children. Developmental Psychology, 38(5), 913-923.
Catts, H. (1986). Speech production/phonological deficits in reading-disordered children. Journal of Learning Disabilities, 19, 504-508.
Catts, H. (1989). Phonological processing deficits and reading disabilities. In A. G. Kamhi & H. Catts (Eds.), Reading disabilities. A developmental language perspective (pp. 101-132). Boston: Allyn and Bacon.
Catts, H. (1991). Early identification of dyslexia: Evidence from a follow-up study of speech-language impaired children. Annals of Dyslexia, 41, 163-177.
Catts, H. (1993). The relationship between speech-language impairments and reading disabilities. Journal of Speech and Hearing Research, 36, 948-958.
Catts, H., Fey, M. E., Zhang, X., & Tomblin, J. B. (2001). Estimating the risk of future reading difficulties in kindergarten children: A research-based model and its clinical implementation. Language, Speech, and Hearing Services in Schools, 32(1), 38-50.
Chaney, C. (1992). Language development, metalinguistic skills, and print awareness in 3-year old children. Applied Psycholinguistics, 13, 485-514.
Chiappe, P., Chiappe, D. L., & Siegel, L. (2001). Speech perception, lexicality and reading skill. Journal of Experimental Child Psychology, 80, 58-74.
Clay, M. (1979). The early detection of reading difficulties: A diagnostic survey with recovery procedures (2nd ed.). Portsmouth, NH: Heinemann Educational Books.
DeFries, J. C., Knopik, V. S., & Wadsworth, S. J. (1999). Colorado twin study of reading disability. In D. D. Duane (Ed.), Reading and attention disorders’, neurobiological correlates. Baltimore, MD: York Press Inc.
DeFries, J. C., Vogler, G. P., & LaBuda, M. C. (1986). Colorado Family Reading Study: An overview. In J. L. Fuller & E.C. Simmel (Eds.), Perspectives in behavior genetics (pp. 29-56). Hillsdale, NJ: Erlbaum.
Dodd, B. (1995). Differential diagnosis and treatment of children with speech disorder. London: Whurr Publishers.
Edwards, J., & Lahey, M. (1998). Nonword repetitions of children with specific language impairment: Exploration of some explanations for their inaccuracies. Applied Psycholinguistics, 19, 278-309.
Elbro, C. (1996). Early linguistic abilities and reading development: A review and a hypothesis. Reading and Writing: An Interdisciplinary Journal, 8, 453-485.
Elbro, C., Borstrom, I., & Petersen, D. K. (1998). Predicting dyslexia from kindergarten: The importance of distinctness of phonological representations of lexical items. Reading Research Quarterly, 33(1), 36-60.
Evans, J. L., Viele, K., Kass, R. E., & Tang, F. (2002). Grammatical morphology and perception of synthetic and natural speech in children with specific language impairments. Journal of Speech, Language, and Hearing Research, 45(3), 494-505.
Fletcher-Flinn, C. M., & Thompson, G. B. (2000). Learning to read with underdeveloped phonemic awareness but lexicalized phonological recoding: A case study of a 3-year-old. Cognition, 74, 177-208.
Fowler, A. E. (1991). How early phonological development might set the stage for phonological awareness. In S. Brady & D. Shankweiler (Eds.), Phonological processes in literacy: A tribute to Isabelle Y. Liberman (pp. 97-117). Hillsdale, NJ: Lawrence Erlbaum Associates.
Fowler, A. E., & Swainson, B. (2004). Relationships of naming skills to reading, memory, and receptive vocabulary: Evidence for imprecise phonological representations of words by poor readers. Annals of Dyslexia, 54(2), 247-280.
Foy, J. G., & Mann, V. A. (2001). Does strength of phonological representations predict phonological awareness in preschool children? Applied Psycholinguistics, 22(3), 301-325.
Foy, J. G., & Mann, V. A. (2003). Home literacy environment and phonological awareness: Differential effects for phoneme awareness and rhyme awareness. Applied Psycholinguistics, 24(1), 59-88.
Foy, J. G., & Mann, V. A. (2006). Changes in letter sound knowledge are associated with development of phonological awareness in preschool children. Journal of Research in Reading, 29(2), 143-161.
Gallagher, A., Frith, U., & Snowling, M. J. (2000). Precursors of literacy delay among children at genetic risk of dyslexia. Journal of Child Psychology and Psychiatry and Allied Disciplines, 41(2), 202-213.
Gathercole, S. E., & Baddeley, A. D. (1990). Phonological memory deficits in language disordered children: Is there a causal connection? Journal of Memory and Language, 29, 336-360.
Gathercole, S. E., Willis, C. S., Baddeley, A. D., & Emslie, H. (1994). The children’s test of nonword repetition: A test of phonological working memory. Memory. Special Issue: Memory tests and techniques, 2(2), 103-127.
Goldman, R., & Fristoe, M. (1986). Goldman-Fristoe test of articulation. Circle Pines, MN: American Guidance Service.
Goldman, R., Fristoe, M., & Woodcock, R. W. (1970). Goldman-Fristoe-woodcock test of auditory discrimination. Circle Pines, MN: American Guidance Service.
Gottardo, A., Stanovich, K. E., & Siegel, L. S. (1996). The relationships between phonological sensitivity, syntactic processing, and verbal working memory in the reading performance of third-grade children. Journal of Experimental Child Psychology, 63(3), 563-582.
Hesketh, A., Adams, C., & Nightingale, C. (2000). Metaphonological abilities of phonologically disordered children. Educational Psychology, 20(4), 483-499.
Hulme, C. (2002). Phonemes, rimes, and the mechanisms of early reading development. Journal of Experimental Child Psychology, 82, 58-64.
Hulme, C., Hatcher, P. J., Nation, K., Brown, A., Adams, J., & Stuart, G. (2002). Phoneme awareness is a better predictor of early reading skills than onset-rime awareness. Journal of Experimental Child Psychology, 82(1), 2-28.
Joanisse, M. F., Manis, F. R., Keating, P., & Seidenberg, M. S. (2000). Language deficits in dyslexic children: Speech perception, phonology and morphology. Journal of Experimental Child Psychology, 77 (1), 30-60.
Kahn, L. M. L., & Lewis, N. P. (1986). The Kahn-Lewisphonological analysis. Circle Pines, MN: American Guidance Service.
Kuhl, P. (2004). Early language acquisition: cracking the speech code. Nature Reviews Neuroscience, 5, 831-843.
Kuhl, P. K., Conboy, B. T., Padden, D., Nelson, T., & Pruitt, J. (2005). Early speech perception and later language development: Implications for the “critical period”. Language Learning and Development, 1, 237-264.
Lefly, D. L., & Pennington, B. F. (1996). Longitudinal study of children at high family risk for dyslexia: The first two years. In M. L. Rice (Ed.), Towards a genetics of language. Hillsdale NJ: Lawrence Erlbaum.
Leitao, S., Hogben, J., & Fletcher, J. (1997). Phonological processing skills in speech and language impaired children. European Journal of Disorders of Communication, 32, 91-111.
Lewis, B. A., Ekelman, B. L., & Aram, D. (1989). Familial phonological disorders. Journal of Speech and Hearing Research, 32, 713-724.
Lyon, G. R. (1995). Toward a definition of dyslexia. Annals of Dyslexia, 45, 3-27.
Lyytinen, H., Ahonen, T., Eklund, K., Guttorm, T. K., Laasko, M-L., Leinonen, S., et al. (2001). Developmental pathways of children with and without familial risk for dyslexia during the first years of life. Developmental Neuropsychology, 20(2), 535-554.
Mahony, D., Singson, M., & Mann, V. (2000). Reading ability and sensitivity to morphological relations. Reading & Writing, 12, 191-218.
Major, E. M., & Bernhardt, B. H. (1998). Metaphonological skills of children with phonological disorders before and after phonological intervention. International Journal of Language and Communication Disorders, 33, 413-444.
Manis, F. R., McBride-Chang, C., Seidenberg, M. S., Keating, P., Doi, L. M., & Petersen, A. (1997). Are speech perception deficits associated with developmental dyslexia? Journal of Experimental Child Psychology, 66, 211-235.
Mann, V. A., & Foy, J. G. (2003). Speech development, phonological awareness, and letter knowledge in preschool children. Annals of Dyslexia, 53, 149-173.
Mann, V. A., Liberman, I. Y., & Shankweiler, D. (1980). Children’s memory for sentences and word strings in relation to reading ability. Memory and Cognition, 8, 329-335.
Mann, V. A., & Singson, M. (2000). Precocious readers and the phonological basis of early reading ability. Paper presented at the annual meeting of the International Association for Research on Learning disabilities, Vancouver, British Columbia.
Mann, V. A., & Wimmer, H. (2002). Phoneme awareness and pathways into literacy: A comparison of German and American children. Reading and Writing: An Interdisciplinary Journal, 15(7-8), 653-682.
Marshall, C. M., Snowling, M. J., & Bailey, P. J. (2001). Rapid auditory processing and phonological ability in normal readers and readers with dyslexia. Journal of Speech, Language, and Hearing Research, 44 (4), 925-941.
McBride-Chang, C. (1996). Models of speech perception and phonological processing in reading. Child Development, 67, 1836-1856.
McBride-Chang, C., Wagner, R. K., & Chang, L. (1997). Growth modeling of phonological awareness. Journal of Educational Psychology, 89(4), 621-630.
Menyuk, P., Chesnick, M., Libergott, J., Komgold, B., D’Agostino, R., & Belanger, A. (1991). Predicting reading problems in at-risk children. Journal of Speech and Hearing Research, 34, 893-903.
Morais, J., Bertelson, P., Gary, L., & Alegria, J. (1986). Literacy training and speech segmentation. Cognition, 24, 45-64.
Muter, V. (1994). The influence of phonological awareness and letter knowledge on beginning reading and spelling development. In C. H. M. Snowling (Ed.), Reading development and dyslexia (pp. 45-62). London: Whurr.
Muter, V., & Snowling, M. (1998). Concurrent and longitudinal predictors of reading: The role of metalinguistic and short-term memory skills. Reading Research Quarterly, 33(3), 320-337.
Norusis, M. J. (2000). Guide to data analysis. Upper Saddle River, NJ.
Paul, R., & Jennnings, P. (1992). Phonological behavior in toddlers with slow expressive language development. Journal of Speech, Language, and Hearing Research, 35(1), 99-108.
Porter, J. H., & Hodson, B. W. (2001). Collaborating to obtain phonological acquisition data for local schools. Language, Speech, and Hearing Services in Schools, 32(3), 165-171.
Raitano, N. A., Pennington, B. E, Tunick, R. A., Boada, R., & Shriberg, L. D. (2004). Pre-literacy skills of subgroups of children with speech sound disorders. Journal of Child Psychology and Psychiatry, 45(4), 821-835.
Ratner, N. B. (1994). Phonological analysis of child spech. In J. Sokolov & C. Snow (Eds.), Handbook of research in child language development using CHILDES. Hillsdale, NJ: Erlbaum.
Rescorla, L., & Ratner, B. (1996). Phonetic profiles of toddlers with specific expressive language impairment (SLI-E). Journal of Speech and Hearing Research, 39(1), 153-166.
Roberts, J., Rescorla, L., Giroux, J., & Stevens, L. (1998). Phonological skills of children with Specific Expressive Language Impairment (SLI-E): Outcome at age 3. Journal of Speech, Language, and Hearing Research, 41, 374-384.
Rvachew, S., & Grawburg, M. (2006). Correlates of phonological awareness in preschoolers with speech sound disorders. Journal of Speech, Language & Hearing Research, 49, 74-87.
Scarborough, H. S. (1989). Prediction of reading disability from familial and individual differences. Journal of Educational Psychology, 61, 101-108.
Scarborough, H. S. (1990). Very early language deficits in dyslexic children. Child Development, 61, 1728-1743.
Scarborough, H. S. (1991). Antecedents to reading disability: Preschool language development and literacy experiences of children from dyslexic families. Reading and Writing: An Interdisciplinary Journal, 3, 219-233.
Scarborough, H. S., & Brady, S. (2002). Toward a common terminology for talking about speech and reading: A glossary of the “phon” words and some related terms. Journal of Literacy Research, 34, 299-334.
Scarborough, H. S., & Dobrich, W. (1990). Development of children with early language delay. Journal of Speech and Hearing Research, 33, 70-83.
Shriberg, L. D. (1993a). Four new speech and prosody-voice measures for genetics research and other studies in developmental phonological disorders. Journal of Speech and Hearing Research, 36(1), 105-140.
Shriberg, L. D. (1993b). Four new speech and prosody-voice measures for genetics research and other studies in developmental phonological disorders. Journal of Speech and Hearing Research, 36, 105-140.
Shriberg, L. D., Austin, D., Lewis, B. A., McSweeny, J. L., & Wilson, D. L. (1997). The Percentage of Consonants Correct (PCC) metric: Extensions and reliability data. Journal of Speech, Language & Hearing Research, 40, 708-722.
Shriberg, L. D., Friel-Patti, S., Flipsen, P., & Brown, R. L. (2000). Otitis media, fluctuant hearing loss, and speech-language outcomes: A preliminary structural equation model. Journal of Speech, Language, & Hearing Research, 43(1), 100-121.
Shriberg, L. D., & Kent, R. D. (1995). Clinical phonetics. New York: Macmillan.
Shriberg, L. D., & Kwiatkowski, J. (1982). Phonological disorders III: A procedure for assessing severity of involvement. Journal of Speech & Hearing Disorders, 47, 256-270.
Shriberg, L. D., & Kwiatkowski, J. (1994). Developmental phonological disorders I: A procedure for assessing severity of involvement. Journal of Speech & Hearing Disorders, 37, 1100-1126.
Shriberg, L. D., Kwiatkowski, J., Best, S., Hengst, J., & Terselic-Weber (1986). Characteristics of children with phonologic disorders of unknown origin. Journal of Speech and Hearing Research, 51, 140-161.
Shriberg, L. D., Kwiatkowski, J., & Gruber, F. A. (1994). Developmental phonological disorders: II. Short-term speech-sound normalization. Journal of Speech and Hearing Research, 37, 1127-1150.
Singson, M., Mahony, D., & Mann, V. (2000). The relation between reading ability and morphological skills: Evidence from derivational suffixes. Reading & Writing, 12, 219-252.
Smit, A. B. (1993a). Phonologic error distributions in the Iowa-Nebraska articulation norms project: consonant singletons. Journal of Speech, Language, & Hearing Research, 36, 553-547.
Smit, A. B. (1993b). Phonological error distributions in the Iowa-Nebrasks articulation norms project: Word initial consonant clusters. Journal of Speech and Hearing Research, 36, 931-947.
Snowling, M. (1981). Phonemic deficits in developmental dyslexia. Psychological Research, 43, 219-234.
Snowling, M., Goulandris, N., Bowlby, M., & Howell, P. (1986). Segmentation and speech perception in relation to reading skill: A developmental analysis. Journal of Experimental Child Psychology, 41, 489-507.
Snowling, M., Hulme, C., Smith, A., & Thomas, J. (1994). The effects of phonemic similarity and list length on children’s sound categorisation performance. Journal of Experimental Child Psychology, 58, 160-180.
Stainthorp, R., & Hughes, D. (1998). Phonological sensitivity and reading: Evidence from precocious readers. Journal of Research in Reading, 21(1), 53-68.
Sussman, J. E. (1993). Perception of formant transition cues to place of articulation in children with language impairments. Journal of Speech and Hearing Research, 36, 1286-1299.
Tabachnick, B. G., & Fidell, L. S. (2001). Using multivariate statistics (4th ed.) New York: Harper Collins.
Thomas, E., & Senechal, M. (2004). Long-term association between articulation quality and phoneme sensitivity: A study from age 3 to age 8. Applied Psycholinguistics, 25, 513-541.
Treiman, R., Broderick, V., Tincoff, R., & Rodriguez, K. (1998a). Children’s phonological awareness: Confusions between phonemes that differ only in voicing. Journal of Experimental Child Psychology, 68 (1), 3-21.
Treiman, R., Tincoff, R., Rodriguez, K., Monzaki, A., & Francis, D. J. (1998b). The foundations of literacy: Learning the sounds of letters. Child Development, 69(6), 1524-1540.
Wagner, R. K., & Torgesen, J. K. (1987). The nature of phonological processing and its causal role in the acquisition of reading skills. Psychological Bulletin, 30, 73-87.
Walley, A. C. (1993). The role of vocabulary development in children’s spoken word recognition and segmentation ability. Developmental Review, 13, 286-350.
Webster, P. E., & Plante, A. S. (1992a). Effects of phonological impairment on word, syllable and phoneme segmentation and reading. Language, Speech, and Hearing Services in Schools, 23, 176-182.
Webster, P. E., & Plante, A. S. (1992b). Phonologically impaired preschoolers: Rhyme with an eye toward reading. Perceptual and Motor Skills, 75, 176-182.
Webster, P. E., & Plante, A. S. (1995). Productive phonology and phonological awareness in preschool children. Applied Psycholinguistics, 16, 43-57.
Webster, P. E., Plante, A. S., & Couvillion, L. M. (1997). Phonological impairment and prereading: Update on a longitudinal study. Journal of Learning Disabilities, 30(4), 365-375.
Wechsler, D. (1992). Wechsler intelligence scale for children. London: Psychological Corporation.
Whitehurst, G., Smith, M., Fischel, J., Arnold, D., & Lonigan, C. (1991). The continuity of babble and speech in children with specific expressive language delay. Journal of Speech and Hearing Research, 34, 1121-1129.
Woodcock, R. M. (1987). Woodcock reading mastery-revised, form H. Circle Pines, MN: American Guidance Service.
This manuscript was peer reviewed, processed and accepted under the editorship of the immediate past editor Dr. Che Kan Leong.
V. A. Mann (*)
Department of Cognitive Sciences, University of California-Irvine, Irvine, CA 92697-5100, USA
J. G. Foy
Loyola Marymount University, Los Angeles, CA, USA
Copyright International Dyslexia Association Jun 2007
Provided by ProQuest Information and Learning Company. All rights Reserved