Even frequent and expected words are not identified without spatial attention

Previous studies have disagreed about the extent to which people extract meaning from words presented outside the focus of spatial attention. The present study examined a possible explanation for such discrepancies inspired by attenuation theory: Unattended words can be read more automatically when they are expected within a given context (e.g., due to frequent repetition). We presented a brief prime word in lowercase, followed by a target word in uppercase. Participants indicated whether the target word belonged to a particular category (e.g., "sports"). When we used a visual cue to draw attention to the location of the prime word, it produced substantial priming effects on target responses (i.e., especially fast responses when the prime and target words were identical or from the same category). When prime words were not attended, however, they produced no priming effects. This finding replicated even when there were only four words, each repeated 160 times during the experiment. It appears that very little word processing is possible without spatial attention, even for words that are expected and frequently presented.