Abstract

This research explored the cognitive nature of the RAN (Rapid Automatized Naming) task, a test widely used to assess reading development. It addressed automaticity- and attention-based processing and their relative contribution to RAN task performance to better understand why the RAN task has the predictive value for reading development. Study 1 (N=68) utilized two different indices of automatic stimulus recognition and an index of attention control as predictors of naming speed on the four original versions of the RAN task. The study found little support for an automaticity-based account of RAN task performance, but did support an attention-based account. Symbolic and non-symbolic RAN subtasks differed in terms of the role played by automatic and attention-based factors, and in terms of their correlations with reading speed. Study 2 (N=16) used ten modified versions of the RAN task that manipulated attention and memory demands. Naming speed was sensitive to attentional demands and to stimulus familiarity, but not to factors of long-term memory retrieval. Study 3 (N=97) provided additional information on the roles played by automatic and attention-based processing in RAN task performance, using new measures of these constructs. Attention came out as explaining a large proportion of the variance in naming speed; skill in automatic stimulus detection and in lexical access efficiency did not. Working memory was strongly associated with RAN task performance. Finally, a meta-analysis on a representative sample of research data (65 studies reporting 530 coefficients of correlation between RAN tasks performance and different measures of reading, N=8555) revealed the average point estimates were r + = .345 and r + = .398, for cross-sectional and longitudinal research designs respectively. The moderator analyses showed that reading skills more closely associated with RAN task performance required expertise with printed text and depend on applying rules and building and managing associations. These regularities are largely consistent with the results of the three experimental studies. Overall, these indicated that attention-based factors rather than automaticity underlie naming speed as measured by the RAN tasks, and these mechanisms presumably link RAN to reading performance. Implications for further research and educational practices are discussed.