When parietal-damaged patients fail to report a contralesional stimulus because of extinction, is this because the stimulus is not perceived, or because it is perceived but cannot reach conscious awareness? VOLPE et al. [10] reported an intriguing study that seemed to locate the problem at least partly in the transfer of information to conscious awareness. They showed patients with extinction pairs of stimuli, one in each hemifield. Although patients were predictably poor at reporting the identity of the contralesional stimulus, they were able to make accurate same/different judgements comparing the two stimuli. This was interpreted as evidence that both stimuli were perceived. In the present paper, we point out that the dissociation between identification and same/different matching could also be due to the possibility that less visual information about the contralesional stimulus is necessary to make a same/different judgement than to identify the stimulus, and that chance performance is considerably higher in the first than in the second type of task. In Experiment 1, we verified this by degrading one side of a stimulus display and "replicating" the dissociation with normal subjects. We also equated the amount of visual information needed for the two tasks by yoking the stimulus pairs on "different" trials of the same/different matching task with the choice pairs on a forced choice identification task. Under these conditions, the dissociation vanished. In Experiment 2, we administered these tasks to three parietal-damaged patients with extinction. When the original method was used, same/different matching was better than identification of the contralesional stimulus. With the forced choice identification method, the dissociation again vanished.