Three experiments investigated the identification or localization of a letter that was displaced from the fixation point by 1 degree-3 degrees. The subject's task was to identify a fixated letter and identify (Experiment 1) or localize (Experiments 2 and 3) the displaced letter. On uncued trials, the displaced letter could appear at any of eight locations on any of three rings surrounding the fixated letter; on cued trials, the ring containing the displaced letter was specified. The results indicated that cuing improved subjects' identification and localization of the displaced letter. Invalid cuing (Experiment 3) produced costs comparable in magnitude to the benefits. The distance of the target from the cued ring determined cost, but costs were unaffected by the appearance of a target within the presumed beam of attention. It was proposed that attention should be viewed as a general, rather than feature-specific, resource that can be voluntarily allocated to multiple regions of the visual field.