Many visual capabilities deteriorate with age, including sensitivity to motion and direction-discrimination. We examined how age affects the relation between motion detection and perceived direction.
Thirty-nine observers with normal vision, aged 20-81 yrs, viewed random dot cinematograms at four stimulus durations ranging from 75 to 460 ms. On signal trials, the direction of each of the 300 dots was randomly selected from a uniform distribution of directions +/- 125 deg about the mean direction. On noise trials, the range of directions was 360 deg, and the stimulus evoked no percept of global flow.
After each stimulus, observers made two different judgements. First, they used a mouse-controlled cursor to indicate the perceived direction of global flow; then, they used a six-point scale to rate their certainty that the stimulus had been signal rather than noise.
Sensitivity to global flow was indexed by
- cumulating responses from the six-point rating scale,
- generating receiver operating characteristics, and then
- calculating the fraction of a unit square that lies beneath the receiver operating characteristic.
Measures of sensitivity to global flow were computed for subjects in each of five age-groups and for four stimulus durations. Analysis showed that motion detectability varied significantly across age groups, (p<.01), and there was a significant interaction between age group and duration (p<.05). Post-hoc tests showed that sensitivity was significantly lower in the 70-81 yr olds, especially at the shortest durations.
Accuracy of judgments of global flow's direction was indexed by the absolute difference between the actual mean direction of flow and the subject's judged direction of flow. Accuracy of judging the direction of global flow differed significantly across age-groups (p<.01), and there was a significant interaction between age group and duration (p<.01). Post-hoc tests showed that accuracy was significantly lower in the 70-81 yr olds, especially at the shortest durations.
People over 70 yrs are less sensitive to motion flow than are younger people, and they also make larger errors when attempting to identify the direction of global flow. These age-differences diminish when longer stimulus durations allow observers more time to process stimulus information. We are currently determining whether standard models of motion perception (e.g., Sekuler et al. 1996) can account for these age-related changes.