Abstract
The mechanisms responsible for age-related declines in face perception are not well understood. Using the classification image (CI) method with sub-sampled faces, Creighton et al. (VSS 2014) showed that, relative to younger adults, older adults made less efficient use of informative face regions. However, many older adults showed no obvious structure in their CIs, suggesting increased internal noise may play a role in decreased performance. To address this issue, we assessed the relative contributions of additive internal noise and efficiency on age-related differences in face discrimination using the equivalent input noise paradigm. Contrast thresholds were measured in a 2 alternative forced-choice face discrimination task in 17 older adults and 14 younger adults with faces presented in varying levels of external noise, and the threshold-vs-noise curves were used to estimate efficiency and equivalent input noise. To facilitate comparison with previous work (e.g., Gold et al., Cog Sci, 2004), thresholds were measured with full faces and sub-sampled faces. For full faces, we observed age-related decreases in efficiency, but no evidence of increased equivalent input noise. For sub-sampled faces, relative to younger adults, older adults showed decreased efficiency and increased equivalent input noise. These results are consistent with those of Creighton et al. (VSS 2014), and support the notion that older adults' lack of obvious structure in their CIs was due partly to a decreased ability to extract diagnostic information in faces and partly to increased additive internal noise. We currently are testing additional subjects who participated in the original CI study to identify the role of individual differences in comparing results across the two methods. In sum, our findings suggest that age-related declines in face discrimination are attributable both to decreased efficiency and, at least in some conditions, to increased additive internal noise.
Meeting abstract presented at VSS 2017