Four Days of Visual Contrast Deprivation Reveals Limits of Neuronal Adaptation

Abstract

Sensory systems continuously adjust their function to match changes in the environment. Such adaptation produces large perceptual effects, and its pervasiveness makes it a key part of understanding cortical function generally. In visual contrast adaptation, for example, brief exposure to vertical stripes can dramatically alter the apparent orientation and intensity of similarly oriented patterns (e.g., [4-7]). However, many environmental changes are long lasting. How does the visual system adjust to such challenges? Most past work on contrast adaptation has adapted subjects for just a few minutes. Only a few studies have examined durations greater than 1 hr, and none have exceeded 1 day. Here, we measured perceptual effects of adaptation in humans who viewed a world lacking vertical information for 4 days continuously. As expected, adaptation increased in magnitude during the first day, but it then showed a drop in strength. The decrease in adaptation is surprising because the adapting environment remained constant, and in short-term work, adaptation always strengthens or at least is maintained under such conditions. It indicates that the classical effects of contrast adaptation, which arise largely in primary visual cortex, are not maintained after approximately 1 day. Results from day 2 through day 4 further showed that slower adaptive processes can overcome this limit. Because adaptation is generally beneficial overall, its limits argue that the brain is sensitive to costs that arise when the neural code changes. These costs may determine when and how cortex can alter its function.

Plasticity, and Its Limits, in Adult Human Primary Visual Cortex

“…Importantly, the decline in adaptation strength observed by Haak et al (2014b) was then followed by an increase in adaptation during subsequent days, indicating that a second, more slowly acting adaptive mechanism was able to overcome the costs of the initial adjustments in neuronal responsiveness. It is likely that this second, slower form of adaptation reflects a process more similar to 'perceptual learning', during which the visual system typically adjusts the neural codes in later rather than in earlier visual areas (see e.g., Hochstein and Ahissar, 2002;Ahissar and Hochstein, 2004).…”

“…Indeed, Haak et al (2014b) observed that the tilt-aftereffect, an illusion thought to be due to the coding-catastrophe, began to decline toward the end of the experiment. Here, we put forward the hypothesis that the same principles may also apply to visual processing in the face of retinal lesions.…”

“…Using immersive virtual reality, Haak et al (2014b) exposed a group of young adults to a world with only very little vertical visual contrast energy for four days continuously, in an attempt to mimic classic selective rearing experiments (Hirsch and Spinelli, 1970;Blakemore and Cooper, 1970) in adult humans. Just as staring at a waterfall for a prolonged period of time changes the response gains of motion-sensitive neurons, the prolonged viewing of a world with relatively little vertical contrast will cause adjustments to the responsiveness of orientation-selective cells in primary visual cortex (e.g., Graham, 1989;Maffei et al, 1973;Movshon and Lennie, 1973;Ohzawa et al, 1985;Dragoi et al, 2000).…”

“…Just as staring at a waterfall for a prolonged period of time changes the response gains of motion-sensitive neurons, the prolonged viewing of a world with relatively little vertical contrast will cause adjustments to the responsiveness of orientation-selective cells in primary visual cortex (e.g., Graham, 1989;Maffei et al, 1973;Movshon and Lennie, 1973;Ohzawa et al, 1985;Dragoi et al, 2000). Monitoring for the perceptual consequences of these changes in the responsiveness of orientation-selective neurons, Haak et al (2014b) found that adaptation increased in magnitude during the first day, but then decreased, despite the sustained presence of the adapting environment. Thus, it appears that there are factors that prevent visual neuroplasticity, in the form of neuronal adaptation, from being sustained over the long-term.…”

“…Thus, it appears that there are factors that prevent visual neuroplasticity, in the form of neuronal adaptation, from being sustained over the long-term. Haak et al (2014b) concluded that if neuronal adaptation does in fact optimize vision, then the decline in adaptation strength must be due to costs that outweighed its benefits. An obvious candidate cost is the 'c ding catastr phe', where changes in the firing of neurons responsible for early visual processing are mistaken for 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 stimulus-changes by neurons that are responsible for subsequent, higher-level stages of visual processing (Schwartz et al, 2007;Series et al, 2009;Druv and Carandini, 2014;Patterson et al, 2014).…”