A number of researchers have suggested that deficient visual attention may play a causal role in dyslexia. However, traditional methods for investigating this assertion have been limited by the conflation of sensory and attentional factors and the inability to isolate large attentional effects. In this study, we sought to overcome these problems by combining spatial cuing with a visual search task measuring psychophysical thresholds. In normal readers, uncued search performance was characterized by a strong dependence on the number of elements in the stimulus array. Cuing the location of the target removed much of this effect, suggesting attentional facilitation of performance. Although dyslexic participants' performance in uncued search was nearly identical to that of normal readers, all dyslexic participants failed to gain the same effect of cuing that normal readers did. However, dyslexic participants did not differ from normal readers on tests of magnocellular function, suggesting that this spatial-cuing deficit is not merely a secondary consequence of magnocellular dysfunction.