Abstract
At least three studies (Bennett et al., 1999; Spérenza et al., 2001; Pardhan, 2004) have used external noise paradigms to investigate the cause of contrast sensitivity losses due to healthy aging. These studies have used noise that was spatiotemporally localized on the target. Allard and Cavanagh (2011) have recently shown that the processing strategy can change with localized noise thereby violating the noise-invariant assumption, which compromises the application of external noise paradigms. The goal of the present study was to reassess the cause of age-related contrast sensitivity losses using external noise that is spatiotemporally broad (i.e. full-screen, continuously displayed dynamic noise). Contrast thresholds were measured for two age groups, young (n = 20, mean = 24 years) and older adults (n = 20, mean = 69 years), for 3 spatial frequencies (1, 3 and 9 cpd) and 3 noise conditions (noise-free, local noise and broad noise). At the lowest spatial frequency, the results drastically differed depending on the noise condition: age-related contrast sensitivity losses were attributed to the internal equivalent noise when using broad noise (i.e. age did not affect contrast thresholds in broad noise) and, consistent with previous studies, due to calculation efficiency with local noise (i.e. similar age-related contrast threshold effects in noise-free and local noise). At the two highest spatial frequencies, the results were similar with local and broad noise: the sensitivity loss was mainly due to lower calculation efficiency, consistent with 2 previous studies. These results show that the interpretation of external noise paradigms can drastically differ depending on the noise type suggesting that external nose paradigms should use spatiotemporally broad external noise, like internal noise, to avoid triggering a processing strategy change. Contrary to all previous studies, we conclude that healthy aging does not affect the calculation efficiency of detection processing at low spatial frequencies.
Meeting abstract presented at VSS 2013