Purchase this article with an account.
Eugenie Roudaia, Patrick J. Bennett, Allison B. Sekuler; The effect of aging on contour integration. Journal of Vision 2007;7(9):604. doi: https://doi.org/10.1167/7.9.604.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Form perception requires the integration of local contour information (Hess et al, 2003). For example, Saarinen & Levi (2001) measured contrast thresholds for identifying the location of the gap in a C-shaped contour comprising Gabor patches. Observers required less contrast to perform the task for Gabors oriented tangentially to the contour than those oriented orthogonally or in mixed orientations, suggesting that young observers integrate information across local orientation elements. Despite our vast knowledge about the contour integration process in young observers, we know virtually nothing about contour integration in older observers. Previous research (Govenlock et al., VSS 2006) showed that the basic mechanisms underlying orientation discrimination are intact in older age, but older observers are impaired at curvature discrimination (Legault et al., VSS 2006). Does this age-related deficit stem from an inability to integrate local orientation information across space? We compared performance of 8 older (∼ 68 years) and 8 younger (∼ 21 years) adults on a variant of Saarinen & Levi's task. Observers viewed C shapes comprising 16 Gabor patches, with the gap in the C facing one of four directions. The orientation of the Gabors varied across blocks: all tangential to the C, all orthogonal, or mixed; stimuli in those conditions were always presented in positive cosine phase. In a fourth condition, Gabors were oriented tangentially, but alternated positive and negative cosine phase. Two interleaved staircases manipulated Gabor contrast to estimate contrast thresholds for correct gap localization. As in Saarienen & Levi, younger observers required significantly less contrast to perceive the contour in the tangential condition than in other conditions. As a group, older observers' thresholds were higher than those of the younger group. However, older observers showed no advantage for tangentially oriented contours. These results support the idea that contour integration processes are impaired in older observers.
This PDF is available to Subscribers Only