Modeling relations among rapid automatized naming, processing speed, and reading fluency in early reading development

Modeling relations among rapid automatized naming, processing speed, and reading fluency in early reading development

First Author: Silvia Siu-Yin Clement-Lam -- Northwestern University
Additional authors/chairs: 
Ola Ozernov-Palchik; Sara D. Beach; Nadine Gaab; John Gabrieli; Elizabeth S. Norton
Abstract / Summary: 

Purpose: Measures of rapid automatized naming (RAN) are correlated with lower-level nonverbal processing speed measures, but little is known about how changes in these automaticity-related skills affect early reading development. This study tested the nature of the relationship between RAN and processing speed, and their relations with reading fluency from kindergarten to grade 2.

Method: Cross-lagged path analyses were conducted with data from N=162 children. Measures included RAN (Objects and Letters), processing speed (Wechsler Coding), and word reading fluency (TOWRE Sight Word Efficiency) from kindergarten, grade 1, and grade 2.

Results: We first tested a multivariate panel model including RAN, processing speed, and reading fluency without cross-lagged effects; the model fit was poor. Then, we tested a model with cross-lagged effects and the model fit was satisfactory (x2=21.62, CFI=0.98, RMSEA=0.08, SRMR=0.05). Although kindergarten and grade 1 RAN measures significantly predicted later processing speed, this relationship was not bidirectional, as early processing speed did not have an impact on subsequent RAN abilities. Early RAN abilities predicted later reading skills; however, early (grade 1) word reading fluency did not explain later RAN skills. These findings are consistent with recent research (Peterson et al., 2017) suggesting that RAN skills in older children are independent of early literacy performance.

Conclusions: The present study suggests a potential directional influence of early RAN on later processing speed and reading skills. The results highlight the uniqueness of RAN beyond the influence of basic processing speed and its critical role in predicting children’s later reading skills.