In an experimental paradigm adapted from Hari (1995), forty observers listened via headphones to 8 binaural clicks: 4 left-ear leading followed by 4 right-ear leading with either 38 or 140 ms interstimulus intervals (ISIs). Concurrently, they viewed either foveal or peripheral visual stimuli designed to activate either the parvocellular or magnocellular pathway. They then reported the perceived location of each click-pair. Our results replicated Hari's finding that observers mistake the perceived location of short ISI click-pairs more often than long. That is, when ISIs were short, the sounds seemed to play across the inside of the head in a phenomenon called illusory directional hearing. However, when click-pairs were accompanied by peripheral visual stimuli that activated the magnocellular pathway, observers were more accurate than when there were no visual stimuli. Conversely, parvocellular-activating foveal visual stimuli produced more illusory hearing than when there were no visual stimuli. These findings suggest that activating the slow sustained parvocellular system may result in a longer processing window. Thirty dyslexic observers who repeated the experimental paradigm had an even longer processing window than control observers indicating that dyslexics may have a magnocellular system deficit. (Contains 4 figures.)