Introduction: Advances in Early Detection of Reading Risk

Abstract

As we write the prologue to this special issue of the Journal of Learning Disabilities, “Advances in Early Detection of Reading Risk,” the U.S. National Early Literacy Panel (NELP) has recently released its report Developing Early Literacy (National Early Literacy Panel, 2008). A chapter of the report is devoted to reviewing studies that attempt to identify the most accurate preschool and kindergarten predictors of later outcomes in reading and spelling. The report is affirmative of an emerging consensus that the very early status of skills directly related to literacy—alphabet knowledge, phonological awareness, rapid automatic naming (RAN), phonological memory, and early writing—is highly correlated to early progress in reading, spelling, and writing. Yet the report also highlights how far we are from fully understanding the early development of literacy. As the authors point out, despite the scope of the analysis being from birth to 5 years old, the initial age of children in most of the studies available was 4 years of age or older and outcome variables generally did not go beyond second grade, with many studies restricting their focus to within-kindergarten relationships. In this way, the findings are highly influenced by proximity effects; namely, skills measurable immediately at the onset of literacy instruction yield the highest correlations to early literacy through both their cognitive and temporal proximity to early reading measures. Another somewhat surprising finding was the inconsistent predictive value of oral language once alphabetic knowledge and phonological awareness were controlled and further the observation that more complex oral language skills, such as grammar, definitional vocabulary, and listening comprehension, had stronger relationships with both decoding and reading comprehension compared to basic measures of vocabulary production/ understanding. The report calls for studies that help elucidate a more nuanced and dynamic understanding of the role of oral language in literacy development. In the collection of articles here, we set out to add meaningfully to the body of evidence summarized by the NELP report in several key ways. The first contribution is adding to the number of studies that go beyond the highly studied kindergarten to second grade period. Understanding associations between success in literacy acquisition and skills measured significantly before literacy emerges may be a more complex task than finding relationships with contiguous skills; however, early intervention really cannot be early enough unless we better understand trajectories of risk observable within the earliest stages of language learning. In “An Exploratory Study of the Development of Early Syllable Structure in Reading-Impaired Children,” Lambrecht Smith, Roberts, Locke, and Tozer report speech production data from the first longitudinal study to look in depth at babbling development between the ages of 8 and 19 months in a sample of children at genetic risk of reading difficulties. The findings are suggestive of phonological vulnerability in at-risk children from English-speaking families even from 8 months of age, with children later identified as reading disabled having a lower proportion of canonical babbling (containing true consonant and vowel combinations) as well as reduced syllable complexity within canonical babbling. In “Language Development, Literacy Skills and Predictive Connections to Reading in Finnish Children With and Without Familial Risk of Dyslexia,” Torppa, Lyytinen, Erskine, Eklund, and Lyytinen also expand our knowledge of very early predictors of literacy skills, within the context of the highly transparent Finnish orthography. In another longitudinal study of infants at genetic risk of reading difficulties, the study characterizes language skills from 11⁄2 years old onward. At 2 years of age, group differences and predictive relationships to reading difficulties at the end of second grade were found within the domains of expressive language and maximum sentence length (although in contrast to Lambrecht Smith et al., this study did not look at speech measures directly). From 21⁄2 years old, all measures except expressive language differentiated the group with dyslexia from the typically reading control group. The study also found that while significant group differences in receptive language were observed at these early ages, the predictive power of the measures in relation to reading was indirect. Receptive language measured at 2 to 21⁄2 years contributed most to subsequent development, however this was via inflectional morphology. Torppa et al. also conclude that “predictors