The functional dissociation of human extrastriate cortical processing streams for the perception of face identity and location was investigated in healthy men by measuring visual task-related changes in regional cerebral blood flow (rCBF) with positron emission tomography (PET) and H2(15)O. Separate scans were obtained while subjects performed face matching, location matching, or sensorimotor control tasks. The matching tasks used identical stimuli for some scans and stimuli of equivalent visual complexity for others. Face matching was associated with selective rCBF increases in the fusiform gyrus in occipital and occipitotemporal cortex bilaterally and in a right prefrontal area in the inferior frontal gyrus. Location matching was associated with selective rCBF increases in dorsal occipital, superior parietal, and intraparietal sulcus cortex bilaterally and in dorsal right premotor cortex. Decreases in rCBF, relative to the sensorimotor control task, were observed for both matching tasks in auditory, auditory association, somatosensory, and midcingulate cortex. These results suggest that, within a sensory modality, selective attention is associated with increased activity in those cortical areas that process the attended information but is not associated with decreased activity in areas that process unattended visual information. Selective attention to one sensory modality, on the other hand, is associated with decreased activity in cortical areas dedicated to processing input from other sensory modalities. Direct comparison of our results with those from other PET-rCBF studies of extrastriate cortex demonstrates agreement in the localization of cortical areas mediating face and location perception and dissociations between these areas and those mediating the perception of color and motion.

We examined age-related changes in object and spatial visual processing in two separate experiments. Regional cerebral blood flow (rCBF) was measured in young and old subjects with positron emission tomography and H,150 during tests of face matching, location matching, and a control task. The task demands in the two experiments were identical, but the stimuli in Experiment II were constructed to equalize stimulus complexity across all three tasks. The old subjects performed more slowly than the young subjects in both experiments, and showed significantly slower reaction times during location matching compared to face matching in Experiment II. Both young and old subjects showed occipitotemporal rCBF activation during face matching and occipitoparietal activation during location matching when these conditions were compared to the control task. However, in both experiments and in both tasks, young subjects showed greater activation of prestriate cortex (Brodmann's area 18), and old subjects had larger rCBF increases in occipitotemporal cortex (area 37). Areas in prefrontal cortex, as well as in inferior and medial parietal cortex, were more activated in the old subjects during location matching in both experiments. These results demonstrate that reliable age-related changes during visual processing can be found in rCBF patterns, suggesting more efficient use of occipital visual areas by younger subjects and more reliance by older subjects on one or more cortical networks, particularly for spatial vision, perhaps to compensate for reduced processing efficiency of occipital cortex. Both the differentially increased reaction times and the more widespread prefrontal activation in the old subjects during location matching suggest that spatial vision may be affected to a greater degree by aging than is object vision.

No abstract

The participation of the medial temporal cortex and other cerebral structures in the memory impairment that accompanies aging was examined by means of positron emission tomography. Cerebral blood flow (rCBF) was measured during encoding and recognition of faces. Young people showed increased rCBF in the right hippocampus and the left prefrontal and temporal cortices during encoding and in the right prefrontal and parietal cortex during recognition. Old people showed no significant activation in areas activated during encoding in young people but did show right prefrontal activation during recognition. Age-related impairments of memory may be due to a failure to encode the stimuli adequately, which is reflected in the lack of cortical and hippocampal activation during encoding.