Training reading fluency: is it important to practice reading aloud and is generalization possible?

Outcomes of three different types of computerized training in sub-lexical items (word-initial consonant clusters) on reading speed for 39 German-speaking poor readers in Grades 2 and 3 were evaluated. A phonological-orthographic association group, a reading aloud group, and a combined group were compared in performance with an untrained control group. During short-term training, the intervention groups showed higher gains than the control group in reading speed of the trained sub-lexical items and of the words containing the trained segments. No differences were found between the intervention groups. In the development of pseudoword reading, the groups did not show differential improvements. The generalization effect to pseudoword reading was similar, whether the pseudowords contained the trained segment as a syllable or as a non-syllabic letter string. The gains induced by training were specific to the materials used in training and did not induce gains in general reading speed.