A lab-controlled simulation of a letter–speech sound binding deficit in dyslexia

• We focused on initial letter–speech sound learning, using an artificial script.
• We found behavioral evidence for a letter–speech sound binding deficit in dyslexia.
• Differences between dyslexics and controls were independent of letter knowledge.
• Disrupted letter–speech sound binding immediately affected reading performance.
• Reading skills in the artificial script predicted reading skills at school.

Dyslexic and non-dyslexic readers engaged in a short training aimed at learning eight basic letter–speech sound correspondences within an artificial orthography. We examined whether a letter–speech sound binding deficit is behaviorally detectable within the initial steps of learning a novel script. Both letter knowledge and word reading ability within the artificial script were assessed. An additional goal was to investigate the influence of instructional approach on the initial learning of letter–speech sound correspondences. We assigned children from both groups to one of three different training conditions: (a) explicit instruction, (b) implicit associative learning within a computer game environment, or (c) a combination of (a) and (b) in which explicit instruction is followed by implicit learning. Our results indicated that dyslexics were outperformed by the controls on a time-pressured binding task and a word reading task within the artificial orthography, providing empirical support for the view that a letter–speech sound binding deficit is a key factor in dyslexia. A combination of explicit instruction and implicit techniques proved to be a more powerful tool in the initial teaching of letter–sound correspondences than implicit training alone.