The power of vowels: Contributions of vowel, consonant and digit RAN to clinical approaches in reading development
Introduction
Understanding the cognitive processes involved when children learn to read is an endeavor that many researchers have pursued over the last decades. Reading skills are key for children's academic success (Stanovich, 1993). Skilled readers experience substantial print exposure and will consequently develop knowledge in many different domains. By contrast, poor readers experience more difficulties in decoding, engage less in reading-related activities, and are likely to develop negative attitudes towards reading itself. Thus, poor readers miss opportunities to develop efficient reading comprehension strategies, which will hinder learning in other academic domains. This mechanism refers to the so-called Matthew effect in academic achievement (Stanovich, 1986).
Much research on early reading development has therefore focused on the study of reliable cognitive correlates for explaining reading progress on the one hand and poor reading skills on the other hand. If children do not develop the fundamental skills necessary for reading acquisition, reading difficulties may inevitably develop. Grasping the type of cognitive processes involved in early reading development may therefore help to develop more accurate diagnostic tools to identify young children at risk for reading difficulties and tailor targeted interventions to avoid severe reading difficulties later on.
Importantly, beginning readers need to develop sensitivity to the sound structure of spoken words and knowledge of letter names and sounds (Treiman, 2000). Reading development has been repeatedly related to alphabetic knowledge (Ehri, Nunes, Stahl, & Willows, 2001), phonological awareness (Muter et al., 2004, Torgesen et al., 1994), vocabulary (e.g., Muter et al., 2004, Verhoeven and Leeuwe, 2008), verbal working memory (Gathercole and Baddeley, 1993, Kibby et al., 2014) and rapid automatized naming speed (Araújo et al., 2015, Kibby et al., 2014, Papadopoulos et al., 2016, Schatschneider et al., 2004, Wolf and Bowers, 1999). Swanson, Trainin, Necoechea, and Hammill's (2003) meta-analysis highlights moderate and similar correlations between phonological awareness, rapid automatized naming (RAN) and word reading. The influence of phonological awareness on reading has been intensively studied and reviewed (Melby-Lervåg et al., 2012, Rayner et al., 2001).
In a review on several longitudinal studies, Scarborough (1998) emphasized that RAN performance in young children was at least as consistent a predictor for later reading skills as phoneme awareness and letter name knowledge. Other studies support the finding that the skills underlying RAN are crucial for reading development (e.g., Araújo et al., 2011, Kirby et al., 2003, Kirby et al., 2008). RAN refers to the performance of quickly naming a series of well-known visual stimuli such as familiar objects, colors, letters and digits (Denckla & Rudel, 1974). Norton and Wolf (2012) suggested that RAN and reading involve closely related cognitive processes, such as attention to the stimuli, visual processes for recognition and discrimination, integration of visual information with stored orthographic and phonological codes, access and retrieval of phonological codes and articulation (also see Manis et al., 1999, Wolf and Bowers, 1999). These shared processes turn RAN performance into a promising predictor of early reading skills (Kirby et al., 2003, Landerl and Wimmer, 2008, Lervåg and Hulme, 2009, Manis et al., 1999).
RAN measures have been widely conducted to 1) predict early and later reading skills in typically developing children at different ages (e.g., Araújo et al., 2011, de Jong and van der Leij, 1999, Kirby et al., 2008, Lervåg and Hulme, 2009, Scarborough, 1998, Protopapas et al., 2013, Schatschneider et al., 2004, Vaessen and Blomert, 2013), and 2) to identify children at risk for learning difficulties and disorders such as dyslexia (e.g., Araújo et al., 2010, Bowers and Wolf, 1993, Furnes and Samuelsson, 2010, McBride-Chang and Manis, 1996, Norton and Wolf, 2012, Pan et al., 2013).
It seems however that the strength and independence of RAN contributions depend on multiple factors, such as on the age or school level of the participants (e.g., kindergarten versus second graders), on the reading outcomes evaluated (e.g., isolated words, text comprehension, speed), on the orthography (e.g., opaque vs. transparent) and on the visually presented stimuli in the RAN task (e.g., objects vs. letters). In some studies RAN appeared to be a stronger predictor in young and poor readers than in older, more experienced readers (Araújo et al., 2015, McBride-Chang and Manis, 1996, but see Landerl & Wimmer, 2008). Vaessen and Blomert (2013) also found the opposite pattern. In a recent study with second and sixth graders, Protopapas et al. (2013) found differences in the RAN - reading relationship throughout reading development. According to them, one common set of cognitive processes may contribute to early reading accuracy and speed while other common processes may be involved in later reading outcomes.
Furthermore, RAN performance appears to be a stronger predictor for reading fluency than for reading accuracy in primary school children (Araújo et al., 2011, Schatschneider et al., 2004, Song et al., 2016). Others distinguish between alphanumeric RAN (letters and digits; de Jong, 2011, Lervåg and Hulme, 2009, Meyer et al., 1998, Vaessen and Blomert, 2013) and non-alphanumeric RAN (objects and colors; de Jong and van der Leij, 1999, Furnes and Samuelsson, 2010, Kirby et al., 2003, Landerl and Wimmer, 2008). Research has repeatedly shown that reading performance was more strongly related to alphanumeric RAN tasks than to non-alphanumeric RAN tasks (see for a review Araújo et al., 2015, Bowey, 2005, Song et al., 2016). Before formal alphabetization though, RAN performance is very often assessed by non-alphanumeric RAN tasks (e.g., color RAN, Georgiou, Tziraki, Manolitsis, & Fella, 2013). But some researchers found that kindergarteners' letter RAN performance was a stronger predictor for reading development than object RAN performance (Schatschneider et al., 2004). Kindergarten letter RAN correlated strongly with first grade performance in reading (i.e., accuracy and fluency; r = 0.65) assessed by the TOWRE (i.e., Test of Word Reading Efficiency). Interestingly, Vaessen and Blomert (2013) found a stronger correlation between digit RAN and reading than between letter RAN and reading in first to sixth graders. Conversely, Araújo et al. (2015) did not find a significant difference in the relationships between digit RAN, letter RAN and reading respectively.
Although the relationship between RAN and reading development has been broadly studied, there are still remaining questions regarding the clinical use of RAN among clinical practitioners. Norton and Wolf (2012, p.446) emphasize: “RAN is one of the most robust early indicators of potential reading difficulties, along with phonological skills and letter name and sound knowledge.” RAN tasks take only few minutes to administer and require brief training to administer and score. Moreover, according to these authors a second advantage of RAN is the time component. Reading progress is often assessed in terms of accuracy although speed and automaticity are also key components of what it means to be a good reader. Thus, reading assessments should include timed measures. Nevertheless, fewer studies compared the unique contribution of different visually presented stimuli (e.g., letters, colors, digits) used in RAN to reading outcomes (but see Araújo et al., 2015, for a recent review; Compton, 2003, Schatschneider et al., 2004). Thus, it appears that RAN measures account for differences in early reading skills but their predictive power clearly depends on the stimuli used in the rapid naming task. While letter and digit naming involve linguistic, meaning “asemantic” coding (without access to semantic representations, Rath et al., 2015), colors involve semantic lexical coding. These visually presented stimuli are processed in different areas of the visual system although in close proximity.
Previous studies focused less on the influence of the type of visually presented stimuli of the RAN task (but see Compton, 2003, for letter substitutions in the letter RAN). The four most common RAN measures (i.e., color, objet, digit and letter) were generally administered and a composite RAN score was computed to analyze the contribution of general RAN to reading development. However, we hypothesize that selecting the RAN measure that is most sensitive in predicting reading development at different ages would improve assessment, more precise identification and consequently intervention practices.
Besides, most letter RANs include a mix of vowels and consonants without explanations regarding the usage of these two categories of letters within a task. For example, Schatschneider et al. (2004) used 5 lowercase letters in their RAN letter task (e.g., a, d, o, s, p). Evans and colleagues reported that in their study, the letters a, d, o, s, and p were named correctly by 73%, 47%, 98%, 95%, and 89% of the children respectively (Evans, Bell, Shaw, Moretti, & Page, 2006). Accordingly, letter name knowledge should be taken into account when administering letter RAN tasks in children at different ages. Specifically, due to the early mastery of vowels, Evans et al. (2006) emphasized that vowels should be treated as a separate category of letters.
A major goal of the current study was therefore to examine the specific contributions of RAN measures with different types of alphanumeric stimuli (i.e., vowels, consonants and digits) to different reading outcomes at different grade levels in beginning readers.
Infants are more familiar with vowels than with consonants. Research on oral phonological development has shown that simple vowels (i.e., a, e, i, o, u) are acquired first, before consonants (Yang & Zhu, 2010). By the end of 6 months, babies engage in babbling (McCarthy, 1960) and start combining vowels and consonants such as for instance “a” and “t”, or “a” and “m” to produce the well-known “tatata”, “mamama”. In the first months of life, infants use about one consonant for every four vowels, whereas by one year, infants use 1 consonant for every 2 vowels. By two years of age infants use both, vowels and consonants (Irwin, 1946). Accordingly the most used speech sounds in the two first years of an infant are vowels (Irwin & Chen, 1946). Given the anatomy of the vocal tract, children start to produce the sounds that are the simplest for them to articulate, which means vowels (Cairns, 1996). Also, Donegan (2013) suggests that vowels occur earlier than consonants due to vowels' higher amplitude and larger duration. Recent studies on speech acquisition point to a functional distinction in vowels and consonants (Carreiras et al., 2009, Carreiras and Price, 2008) and to different contributions of vowels and consonants to language development (Nespor, Peña, & Mehler, 2003). Both may contribute differently to lexical access with consonants being more salient than vowels. Pons and Toro (2010) found that 11-months old infants were sensitive to differences between vowels and consonants before a lexicon was fully developed. In this study infants were able to process structural generalizations over vowels but not over consonants. While vowels are simple sounds, consonants are more difficult to produce. Vowels are produced by position of the tongue and lips and they can be spoken alone and subsequently appear earlier in language development.
Regarding reading acquisition, the main concern is about matching phonemes and graphemes (i.e., written letters) to form complete word representations. The child must learn “how a given writing system relates to spoken units in his or her language”, (Rayner et al., 2001). This matching requires that phonemes are isolated from speech stream. Insofar as vowels are simultaneously sounds, syllables and phonemes, they are easier to isolate and thus to match to the corresponding letters. Compared with vowels, consonants are more abstract. Vowels have relatively longer duration and their sounds can be clearly heard and isolated (e.g., “a” in “pâtes, bar” in French; English translation: noodles, bar); (cf. Rayner et al., 2001). The pronunciation of a consonant (e.g., “l” in “vélo”, “lit”; translated: bike, bed) is more dependent on the preceding/following vowel (coarticulation phenomenon) and the child may have more difficulties to a make mental representation of phonemes in the beginning of reading acquisition without help. Isolating vowels and matching vowels and letters would constitute the most precocious and easiest way to enter reading and probably spelling. As a consequence we expect vowel naming to be an early predictor of the beginning of reading. In contrast to vowels, consonants are more numerous, generally less frequent and harder to isolate in the speech stream. In addition, they can't be produced without associating them with a vowel. As a consequence, we expect consonants to be matched later than vowels to their corresponding letters.
In sum, the above-mentioned differences in vowel and consonant processing encouraged this study to have a close look at the specific contribution of vowel RAN and consonant RAN to reading development in beginning readers. We suggest that vowels and consonants may have distinct predictive values for reading development most probably due to their time of acquisition and to the grade level under investigation. Especially vowel RAN may be particularly important for early reading, while consonant RAN might become a stronger predictor later on in reading development when the naming of consonants is learned and practiced.
For the purpose of our study, we therefore designed two independent letter RAN measures, one requiring the naming of vowels and the other requiring the naming of consonants. We suggest that dividing the standard letter RAN into two measures, a vowel and a consonant RAN, might add new information to the existing body of research on the predictive value of alphanumeric RAN measures to reading outcomes. Moreover, this approach might provide new insights on the influence of the task stimuli used when investigating relationships between RAN and reading development. Hence, the findings may help inform practitioners which RAN task to perform at different stages of reading development.
To the best of our knowledge there is no study to date that examined the unique contribution of vowel, consonant, and digit RAN on early reading outcomes. In the present study we therefore opted for a mixed cross-sectional and repeated measurement design. Our focus was on the specific contribution of distinct RAN measures to reading skills developing from kindergarten to first grade (Group 1) and from first grade to second grade (Group 2). The present research investigates two main research hypotheses. Firstly, we hypothesize that vowel RAN performance (in kindergarteners and first graders) independently predicts reading skills later in first and second grade, after controlling for consonant and digit RAN. Due to the higher frequency of vowels, and the precocious knowledge and use of vowels compared to consonants, we hypothesize that the naming of vowels is more strongly related to early reading performance than consonant RAN. We also hypothesize that vowel RAN should be stronger related to reading outcomes than digit RAN because both, vowel RAN and reading rely on letter processing. Secondly, we propose that consonant RAN assessed in first grade will be a stronger predictor for later reading outcomes than consonant RAN assessed in kindergarten. As first graders begin to read they are more extensively exposed to consonants. Consequently consonant names become more familiar and can be rapidly accessed and retrieved.
To address these hypotheses we assessed color, vowel, consonant and digit RAN performances and reading outcomes in two groups of children at two measurement waves. By adding vowel, consonant and digit RAN measures to the standard color RAN, we attempted to estimate the unique contributions of distinct alphanumeric RAN measures to different reading outcomes in first and second-grade readers.
Section snippets
Participants
The initial sample comprised 93 children. Five participants dropped out of the study due to missing data. The final sample included 88 typically developing children (46 girls; mean age = 77,15 months) from the last year of kindergarten (n = 33; mean age = 71.63 months; Group 1) and first grade (n = 55; mean age = 82.67 months; Group 2) and was followed over a period of 6 months. The present study has been carried out in accordance with appropriate ethical principles and standards. In accordance with the
Principal component analysis, descriptive results and correlational analyses
Principal component analysis based on Eigenvalue produced a single RAN factor explaining 76.67% of the variance. Principal component analysis (PCA) based on the extraction of 2 factors with Varimax rotation, explained 86.47% of the variance. PCA yielded a first alphanumeric factor with the highest loading on vowel RAN and a second non-alphanumeric factor with the highest loading on color RAN.
The descriptive results and Pearson bivariate correlations for RAN and reading accuracy measures for
Discussion
The main goal of the present study was to investigate the specific contribution of three different alphanumeric RAN measures on different reading outcomes in first and second graders. Findings from previous research are mixed regarding the independent contribution of RAN tasks to reading outcomes. As outlined in the introduction, we studied two main research hypotheses and below we discuss further our findings.
First, we investigated the ability of vowel RAN to predict reading outcomes relative
Conclusions
In sum, the present findings indicate that RAN tasks distinctively contributed to reading outcomes in first and second grades. Our results emphasize that vowel RAN was a strong and unique predictor for reading accuracy in first grade. Vowel RAN and digit RAN were both significant and unique predictors for reading speed in second grade. The current findings underline that vowel RAN is a promising predictor in young children for early reading outcomes when most of the consonants still need to be
Acknowledgments
We would like to thank all the children, their parents and their teachers for participating in the resent study and Mrs. Lise Dubouchet-Lombart and Mrs. Bénédicte Debats-Legrand for helping with data collection. We want to thank Dr. Salvador Rivas and Jennifer Jenkins for their editorial support. We would also like to thank 2 anonymous reviewers and the editor for their valuable comments on an earlier draft of the present manuscript.
References (60)
- et al.
Rapid “automatized” naming of pictures, objects, colors, and letters, and numbers by normal children
Cortex
(1974) - et al.
Is rapid automatized naming related to reading and mathematics for the same reason(s)? A follow-up study from kindergarten to Grade 1
Journal of Experimental Child Psychology
(2013) - et al.
On letter frequency effects
Acta Psychologica
(2011) - et al.
Structural generalizations over consonants and vowels in 11-month-old infants
Cognition
(2010) - et al.
Development of serial processing in reading and rapid naming
Journal of Experimental Child Psychology
(2013) Does reading make you smarter? Literacy and the development of verbal intelligence
Introduction to multivariate regression analysis
Hippokratia
(2010)- et al.
Rapid automatized naming and reading performance: A meta-analysis
Journal of Education & Psychology
(2015) - et al.
Component processes subserving rapid automatized naming in dyslexic and non-dyslexic readers
Dyslexia
(2011) - et al.
Visual rapid naming and phonological abilities: Different subtypes in dyslexic children
International Journal of Psychology
(2010)
EDA: Evaluation des fonctions cognitives et apprentissages (Assessment battery of cognitive functions and learning processes)
Theoretical links among naming speed, precise timing mechanisms and orthographic skill in dyslexia
Reading and Writing: An Interdisciplinary Journal
Predicting individual differences in learning to read
A structural equation model using fluency-based early literacy measures to predict emerging reading ability in kindergarten
Remedial and Special Education
The acquisition of language
Brain activation for consonants and vowels
Cerebral Cortex
Are vowels and consonants processed differently? Event-related potential evidence with a delayed letter paradigm
Journal of Cognitive Neuroscience
The influence of item composition on RAN letter performance in first-grade children
Journal of Special Education
What discrete and serial rapid automatized naming can reveal about reading
Scientific Studies of Reading
Specific contributions of phonological abilities to early reading acquisition: Results from a Dutch latent variable longitudinal study
Journal of Educational Psychology
Normal vowel development
Systematic phonics instruction helps students learn to read: Evidence from the National Reading Panel's meta-analysis
Review of Educational Research
Letter names, letter sound and phonological awareness: An examination of kindergarten children across letters and of letters across children
Reading and Writing
Predicting reading and spelling difficulties in transparent and opaque orthographies: A comparison between Scandinavian and U.S./Australian children
Dyslexia
Phonological working memory: A critical building block for reading development and vocabulary acquisition?
European Journal of Psychology of Education
RAN components and reading development from Grade 3 to Grade 5: What underlies their relationship?
Scientific Studies of Reading
Cognitive predictors of word and pseudoword reading in Spanish first-grade children
Frontiers in Psychology
Infant speech: Equations for consonant-vowel ratios
The Journal of Speech Disorders
Infant speech: Vowel and consonant frequency
The Journal of Speech Disorders
Reading performance is predicted by more than phonological processing
Frontiers in Psychology
Cited by (6)
Rapid automatized naming skills of children with intellectual disability
2021, HeliyonCitation Excerpt :It should be noted that the difference between TD children and those with ID was not significant for vowel naming, but this was due more to the low skills of the young TD children than to preserved skills among children with ID. Capital vowels were chosen because they are easier to produce and more familiar to young children (Hornung et al., 2017b). Such items were probably also easier to name for the children with ID than consonants or even lower-case vowels.
The impact of explicitly teaching word segmentation using a visual vowel hand sign system to help at-risk children learn to read and spell English: A proof-of-concept study
2023, Australian Journal of Learning Difficulties