Characterizing the mechanisms behind improvements in visual sensitivity during childhood

The human visual system is not an ideal transmitter of information. A number of separate, quantifiable factors, such as internal noise (Barlow, 1956; Pelli, 1981), have been introduced to characterize what limits our visual sensitivity and how it changes as a result of attention (Lu & Dosher, 1998), training/learning (Li & Levi, 2004), or aging (Betts, Sekuler, & Bennett, 2007). In the current study, we used external noise to model the mechanisms underlying improvements in sensitivity to contrast during childhood. We measured the contrast thresholds of 5-year-olds, 7-year-olds, 9-year-olds, and adults (n = 20/age) in a two-alternative forced-choice orientation discrimination task using the quick-TvC method that adaptively varies the contrast of the signal at a number of levels of external noise (Lesmes, Jeon, Lu, & Dosher, 2006). Overall, contrast thresholds decreased over a wide range of external noise levels as age increased (mean optimal contrast: 8.7%, 5.1%, 4%, and 3.3% for 5-year-olds, 7-year-olds, 9-year-olds, and adults, respectively). A perceptual template model based on Dosher and Lu (1999) provided an excellent fit (r2 = 0.985) to the developmental changes in contrast thresholds at different levels of external noise and performance. The model suggested that a mixture of mechanisms underlie the changes: the improvements in contrast thresholds across ages were best modelled by a combination of reductions in internal additive noise (Aa), reductions in internal multiplicative noise (Am), and improved external noise exclusion (Af). Between 5 and 7 years of age, there were 40%, 70%, and 45% reductions in Aa, Am, and Af, respectively. The modelled improvements likely reflect developmental changes at cortical levels, rather than changes of front-end structural properties (Kiorpes, Tang, Hawken, & Movshon, 2003).