A large number of human actives depend greatly on visual–spatial processing in mental rotation and visual search specifically. The contribution of visual–spatial processing abilities to performance in science, mathematics, engineering, and technology is so important that it even controls other relevant skills, such as memory and mathematical reasoning. Importantly, visual–spatial skills are also malleable which respond effectively to learning interventions and life experiences. In order to reliably identify age-group differences on characteristic of eye movements and priority of feature processing in visual processing, we employed a more complex comparative visual search task. Whereas the general visual search task requires participants to maintain a representation of a single target, comparative visual search task requires participants to keep two sets of items in mind at the same time. By giving abundant information about visual behavior during a significant class of cognitive conditions, comparative visual research task provided more possibilities to formulate and test new visual-spatial processing. In the present study, experimental materials consisted of 72 trials, two halves of which (24, 30, or 36 items per half) contained visual primitives of various colors (blue, green, red) and shapes (triangles, circles, squares). Two targets next to fixation were presented; the targets on one-half of the trials were the same, and on the other half of the trials, the targets differed by shape, color, or shape and color. We conducted a 2 (age: old adult, young adult) × 3 (item: 12, 15,18) × 3 (target: shape, color, shape and color) experimental design. An EyeLink 2000 eye tracker was used to ensure accurate fixation of the designated fixation location prior to each stimulus presentation and to record the location and duration of fixations. Participants performed a conjunctive search to ascertain whether the two halves were identical both quickly and accurately, and they provided no instruction regarding the visual–spatial processing strategy they were to use. The results showed that the comparative search task produced age-group decrements in eye movement measures, including longer total fixation time, more fixation counts, longer average saccade amplitude and fixation time on target stimulus for the older adults when the display size (12, 15, or 18 items per half) increases. These influences of fixation on eye movements showed quite closely with the aging effects observed in standard visual search task, which supported older adults are more likely to experience a larger decline in retinal acuity than young adults in later life. Therefore, older adults required more time than younger adults to detect, search, memory and recognize visual targets. However, Response times were shorter for redundant trials than for color-only and shape-only trials for both old and young participants. Therefore, two groups of participants had similar priority of number of dimension and characteristic of feature attribute. Given the extant evidence of strengths for holistic visual processing in older adults, the results which showed cognitive aging for older adults and similar priority of feature attribute to young adults are significant because they suggest such strengths may be task dependent.