Literacy and phonological skills: a reciprocal relationship?

Literacy and phonological skills: a reciprocal relationship?

First Author: Anna Cunningham -- Coventry University
Keywords: Literacy, Phonological processing, causality
Abstract / Summary: 

The existence of a possible reciprocal relationship between literacy and phonological skills (particularly regarding reading and phonological awareness) has important theoretical and practical implications. For example, such a relationship provides support for the orthographic influence hypothesis, and implies an effect of reading remediation on phonological abilities. However, while there is a large body of evidence showing that phonological awareness predicts reading, there is little extant research showing a convincing effect in the opposite direction. Talks 1 (Cunningham) and 2 (Malkova) present evidence that a bi-directional relationship exists between letter-knowledge and phonemic awareness in Kindergarten. Cunningham through structural equation models based on a large sample of British Kindergartners, and Malkova through an intervention study with Czech pre-schoolers. Talks 3 (Carroll) and 4 (Kolinsky) present evidence that a reciprocal relationship extends to the middle school years and adulthood. Carroll presents data from school children with phonological deficits, showing mutual predictive effects between phonological awareness and word reading accuracy. Finally, Kolinsky presents an intervention study showing that Portuguese adults who acquire literacy experience associated improvements in phonological skills. Talk 5, the Discussion (Cunningham) combines evidence from the four talks and compares with existing research to arrive at novel conclusions about the nature of reciprocity between literacy and phonological skills.

Symposium Papers: 

Decoding skill and awareness of phonemes in nonwords: an early reciprocal relationship at the letter-level only

First Author/Chair:Anna J. Cunningham -- Coventry University, UK
Additional authors/chairs: 
Caroline Witton; Joel B. Talcott; Adrian P. Burgess; Laura R. Shapiro

Purpose. Phoneme awareness is an important predictor of growth in reading skills, but few studies have examined whether the relationship is truly reciprocal, despite frequent statements in the literature that this is the case.
Method. We explored this question using a longitudinal design that measured letter-knowledge, decoding skills (nonword and regular word reading) and phoneme awareness (isolation and deletion of phonemes in nonwords) during the first two years of school. Seven hundred and seven children were tested at school entry (T1), then again at the end of the first year (T2) and again at the end of the second year (T3).
Results. Phoneme awareness predicted letter-knowledge, and letter-knowledge predicted phoneme awareness from T1-T2, indicating an early reciprocal relationship. However, while phoneme awareness predicted decoding, decoding did not predict phoneme awareness from T2-T3, implying that the relationship became uni-directional once children began to read. We will present additional analyses to elucidate the mechanisms behind the initial reciprocity, including a test of the orthographic influence hypothesis.
Conclusions. Our results highlight the role of letter learning in precipitating awareness of phonemes and suggest that phoneme awareness is a distinct process that, once precipitated, is not influenced by reading skill.

Letter knowledge training and its influence on phoneme awareness and early reading development

First Author/Chair:Gabriela Seidlová Málková -- Charles University, Czech Republic
Additional authors/chairs: 
Marketa Caravolas

Purpose: This was a 7-week intervention study aimed at training phoneme awareness (PA) and letter knowledge (LK) to preliterate children (4,5-5 years) to elucidate the interactive relationship between PA and LK and their possible impacts on early reading. In line with the theory of a reciprocally beneficial relationship between PA and LK, we predicted that additionally to the benefits at the level of each trained skill, the LK group would show impact to the growth of PA, and the PA group would also demonstrate impact to the faster learning of LK.
Method: 192 Czech preschool children were randomly assigned to: PA or LK training, or an untreated control group. All groups were pre-tested on PA, LK, reading measures and nonverbal IQ, then tested after the end of training (post-test) and at an 8 month follow-up (delayed post-test). Cohen´s d was used to describe effects of training within groups, ANCOVA analyses were used to assess differences between groups in post-test phases.
Results: In the post-test phase, the LK group demonstrated middle-strong growth of both PA and LK while the PA trained group showed effects limited to PA only. In the delayed post-test, the LK group maintained middle effects in LK measures and showed strong effects in reading while the PA group maintained effects in PA but showed also strong effects in LK. The LK group differed significantly from both other groups in LK measures in both post-test times.
Conclusions: Only LK training seems to immediately boost both LK and PA growth and gradually also helps early reading emergence. PA training has an immediate effect on PA and a delayed effect on LK.

A reciprocal relationship between reading and phonological awareness in the middle school years? Evidence from children with dyslexia and Otitis Media

First Author/Chair:Julia M. Carroll -- Coventry University, UK
Additional authors/chairs: 
Helen Breadmore

Purpose. There is extensive evidence about the relationship between phonological awareness and reading in the early school years, but much less about the middle school years, when children begin ‘reading to learn’ rather than ‘learning to read’. In addition, there is little consideration of whether these associations may differ across groups with different literacy and phonological impairments. This study investigates these issues.
Method. The study focuses on 36 children with dyslexia and 29 children with a history of transient hearing loss due to otitis media (OM) and their chronological age matched and reading age matched controls. The children completed a battery of phonological, morphological and literacy tasks at Time 1, when they were aged 8-10 years old and were retested on word reading, comprehension and phonological awareness 18 months later.
Results. At Time 1, both risk groups had phonological awareness below the level of age matched controls. The dyslexic group showed particular difficulties in tasks involving phonological manipulation, while the OM group showed particular difficulties in tasks involving segmenting and blending phonemes. This indicates some differences in the nature of the phonological awareness deficit across groups.
Across the dyslexic, OM and control groups, reading at time 1 predicted growth in phonological awareness over time. Phonological awareness also predicted growth in text reading accuracy over time. Further analyses will investigate whether these relationships differ across the different groups.
Conclusions. In the middle school years, reading skills predict growth in phonological awareness across children with different phonological impairments. There is some indication that the relationship between reading and phonological awareness may change with age.

From illiteracy to literacy in adulthood: A microlongitudinal study

First Author/Chair:Regine Kolinsky -- Université Libre de Bruxelles, Belgium
Additional authors/chairs: 
Ana Franco ; Isabel Leite; Cristina Carvalho ; Tânia Fernandes ; Axelle Calcus; Jose Morais

Purpose. Our aim was to gain more insight on literacy acquisition in adulthood, about which little is known.
Method. We designed an intensive literacy course that we applied for three ½ months to nine adult illiterate female volunteers. We tested participants five times: two times before starting the literacy classes (pre-tests), two during the literacy classes, and finally after they completed the course. We tracked the evolution of their reading and writing skills (including with eye movements recording of their reading behavior), of their metaphonological abilities, of their memory for spoken nonwords (vs. for visual objects), and of their ability to recognize speech in noisy listening conditions.
Results. Most of the participants learned to read and write, and showed associated enhancements on phonological skills, including phonological memory.
Conclusions. The data support a causal link from literacy to phonological skills.

Discussion - Is there a reciprocal relationship between literacy and phonological skills?

First Author/Chair:Discussant Anna Cunningham

Anna Cunningham will bring together the presented findings and combine with existing research to address outstanding questions in the literature.

- Does the nature of the relationship between literacy and phonological skills depend on which phonological skills are measured? For example, phonological memory and large-segment awareness develop before children/adults learn to read. Talks 3 and 4 suggest that they are stimulated further by the act of reading (therefore, both predict and are predicted by literacy). Conversely, there is debate over whether phoneme awareness can ‘precede’ literacy at all. Therefore, phoneme awareness in pre-readers cannot predict literacy, as it is not yet developed. Talk 2 supports this view by showing that PA training cannot improve literacy until LK is taught.

- Does learning to read 'cause' phoneme awareness? The two intervention studies presented (Talks 2 and 4) provide strong evidence for a causal link between these skills. Talks 1 (correlational), and 2 imply that this link happens as early as learning one's first letters.

-What implications does a causal link have for theory? For example, how many letters need to be known to ‘infer’ phoneme awareness (phonological hypothesis), and is this awareness specific at the item-level (orthographic influence hypothesis).

- Does the relationship between reading and phonological skills change with age? Talk 1 implies that once children are decoding, then reading does not predict phoneme awareness, while Talk 3 suggests that it does. This may be because intermediate readers use orthographic information more to solve phonological tasks.

- Is there sufficient evidence to state with confidence that there is a reciprocal relationship between literacy and phonological skills (as has been frequently done in the literature)? If so, what should be the direction for future research?