On 2016 Apr 10, Lydia Maniatis commented:

For the reasons described below, this introductory sentence has no content: "Noise can be used to characterize visual processing in noiseless conditions. For instance, contrast detection threshold in absence of noise is limited by both internal noise and the ability of detecting the signal embedded in noise, namely, calculation efficiency, which is inversely proportional to the smallest signal-to-(internal) noise ratio required to detect the signal."

The stimulus consists of photons hitting the retina. Is this the "signal" according to these authors? No. The "signal" is a product of perceptual organisation to which the stimulus may give rise. This signal (e.g. a letter of the alphabet) has been designated a priori by the authors. If the stimulus is created by laying out a perceptible letter shape, and then adding dots overlying it, the additional dots will be called the "noise" even though they are a perfectly good "signal." The dots, called "external noise" will be supposed to represent "noise" and to constitute some kind of parallel to supposed "internal noise" that is supposed to mediate the percept (though to my knowledge no empirical tests of this assumption have been performed). The arbitrariness of the signal/noise dichotomy may be appreciated if we imagine that the letters themselves are composed of the same type of dots? Would we say the "signal" is composed of "noise?"

What if, instead of dots, the "added noise" consisted of lines?

Unless we specify that "noise" will always be defined as a specific structure, e.g. pixels of x visual angle and y luminance structure, then the effect of the arbitrarily defined "noise" on the arbitrarily designated "signal" will always depend on the physical characteristics of both and how they interact to influence the perceptual organisation of the stimulus. As has been demonstrated many times (e.g. by Gottschald, Kanizsa, Wertheimer), in some "noisy" conditions the "signal" may not be detectable at all, even if its contrast is high.

Thus, referring to "noise" and "signal" in the abstract, without reference to structure, is meaningless. Therefore, the introductory sentence quoted above has no content.

QED.

On 2016 Apr 10, Lydia Maniatis commented:

For the reasons described below, this introductory sentence has no content: "Noise can be used to characterize visual processing in noiseless conditions. For instance, contrast detection threshold in absence of noise is limited by both internal noise and the ability of detecting the signal embedded in noise, namely, calculation efficiency, which is inversely proportional to the smallest signal-to-(internal) noise ratio required to detect the signal."

The stimulus consists of photons hitting the retina. Is this the "signal" according to these authors? No. The "signal" is a product of perceptual organisation to which the stimulus may give rise. This signal (e.g. a letter of the alphabet) has been designated a priori by the authors. If the stimulus is created by laying out a perceptible letter shape, and then adding dots overlying it, the additional dots will be called the "noise" even though they are a perfectly good "signal." The dots, called "external noise" will be supposed to represent "noise" and to constitute some kind of parallel to supposed "internal noise" that is supposed to mediate the percept (though to my knowledge no empirical tests of this assumption have been performed). The arbitrariness of the signal/noise dichotomy may be appreciated if we imagine that the letters themselves are composed of the same type of dots? Would we say the "signal" is composed of "noise?"

What if, instead of dots, the "added noise" consisted of lines?

Unless we specify that "noise" will always be defined as a specific structure, e.g. pixels of x visual angle and y luminance structure, then the effect of the arbitrarily defined "noise" on the arbitrarily designated "signal" will always depend on the physical characteristics of both and how they interact to influence the perceptual organisation of the stimulus. As has been demonstrated many times (e.g. by Gottschald, Kanizsa, Wertheimer), in some "noisy" conditions the "signal" may not be detectable at all, even if its contrast is high.

Thus, referring to "noise" and "signal" in the abstract, without reference to structure, is meaningless. Therefore, the introductory sentence quoted above has no content.

QED.