Early self-teaching: Testing the nature of orthographic learning and learning transfer in beginning readers

The current study aimed to clarify the nature of orthographic learning during independent reading (i.e., self-teaching) among beginning readers. The most prominent theory addressing learning of word-specific orthographic forms, the self-teaching hypothesis, predicts that beginning reading is beginning orthographic learning. And yet, empirical evidence to date has focused on older children. Here we test the extent to which beginning readers learn new words during self-teaching experiences, whether they transfer that learning to their subsequent processing of related words, and the role of decoding in both processes. In this study, children in Grades 1 and 2 read simple nonwords (e.g., lurb) embedded in short stories adapted to be appropriate for early readers. Children then completed orthographic choice tasks to test both their learning of those words and the transfer of learning to novel words that are either morphologically or orthographically related (e.g., lurber and lurble, respectively). Results indicated that children in Grades 1 and 2 learned the spelling patterns of novel words. Further, they were able to transfer that learning to their processing of the novel related words; however, only orthographically related words showed clear evidence of spelling-specific transfer. Notably, only children in Grade 2 were able to do retain their learning three days later. Finally, results indicated that accurate phonological decoding is not required for learning to occur, although it may facilitate learning for children in Grade 2. Taken together, these findings help to better understand the nature of self-teaching in beginning readers, informing future research and educational practices.