On 2017 Jun 09, Lydia Maniatis commented:

The trigger for choosing this article to comment on was the reference in the title to “low-level mediation.” It reflects one of the deadly sins of contemporary vision science, i.e. the invalid assumption that aspects of perceptual experience may be directly attributed to the activities of particular sets of neurons in particular, anatomically early (more closely linked to the retina) areas of the visual system.

One problem with such a view is that the activities of neurons in these areas, (like the retinal cells themselves, and, arguably, all the neurons in the visual system and beyond) are the physiological basis for all aspects of the percept. Thus, it doesn’t make sense to specially attribute, e.g. the movement of a grey dot to the special sensitivities of neural population x, when these underlie both the perception of the grey dot, the background, the relative position of dot and frame, and every other feature of the percept. Such claims are, in other words, arbitrary and cannot be corroborated by ascertaining what subjects are seeing.

The most popular area to invoke in this respect is, as is the case here, area V1. The reason is historical; Hubel and Wiesel’s work on the striate cortex (corresponding to V1) of cats and later monkeys. They reported on the relative firing rates of neurons to straight bars differing in orientation, length, direction of movement. These results were over-interpreted to mean that neurons in V1 were akin to specialized detectors of the particular stimuli, within the narrow set tested, which made them fire fastest. However, as Teller (1984) noted in her critique of common psychophysical linking propositions, a. we know that the neural code is not based on maximum, but relative, firing rates, b. there is an infinitely large equivalence class of stimuli that would cause these same neurons to fire.

Nevertheless, investigators keep the flawed, fragile story flickering by employing stimuli and conditions whose results may be, though loosely and permissively, interpreted according to such crude assumptions. That the story is woefully inadequate (that investigators don’t know what factors to control so as to, at least, consistently interpret results) is reflected in the perennial admission, as in this article, that psychophysical experiments get conflicting results which investigators are unable to rationalize:

“Differences in the outcomes of different psychophysical procedures have already been noted elsewhere, and perhaps deserve more attention…

“Although our manipulation of attention did not produce a directional aftereffect, Lankheet and Verstraten’s (1995) manipulation of attention did. The reason for this discrepancy remains unclear…” Morgan et al (2016, p. 2630).

“…our results (Exp 4) did not confirm the factual basis for the claim (Blaser et al , 2005) that a 90 degree probe…we cannot be certain why our results are different…Differences include the psychophysical method (2AFC rather than MSS…), the statistical methods of analysis, the use of colors that appeared equally salient to the observer….”

If the authors are saying that even the statistical methods of analysis can potentially produce opposing interpretations of data, but investigators don’t understand how or why, what can possibly be the theoretical value of their own data/interpretations? How are we supposed to evaluate these things, if theoretical control of conditions is so inadequate and interpretation of experimental results so mystifyingly wide open and susceptible to inexplicable “butterfly effects”?

On 2017 Jun 09, Lydia Maniatis commented:

The trigger for choosing this article to comment on was the reference in the title to “low-level mediation.” It reflects one of the deadly sins of contemporary vision science, i.e. the invalid assumption that aspects of perceptual experience may be directly attributed to the activities of particular sets of neurons in particular, anatomically early (more closely linked to the retina) areas of the visual system.

One problem with such a view is that the activities of neurons in these areas, (like the retinal cells themselves, and, arguably, all the neurons in the visual system and beyond) are the physiological basis for all aspects of the percept. Thus, it doesn’t make sense to specially attribute, e.g. the movement of a grey dot to the special sensitivities of neural population x, when these underlie both the perception of the grey dot, the background, the relative position of dot and frame, and every other feature of the percept. Such claims are, in other words, arbitrary and cannot be corroborated by ascertaining what subjects are seeing.

The most popular area to invoke in this respect is, as is the case here, area V1. The reason is historical; Hubel and Wiesel’s work on the striate cortex (corresponding to V1) of cats and later monkeys. They reported on the relative firing rates of neurons to straight bars differing in orientation, length, direction of movement. These results were over-interpreted to mean that neurons in V1 were akin to specialized detectors of the particular stimuli, within the narrow set tested, which made them fire fastest. However, as Teller (1984) noted in her critique of common psychophysical linking propositions, a. we know that the neural code is not based on maximum, but relative, firing rates, b. there is an infinitely large equivalence class of stimuli that would cause these same neurons to fire.

Nevertheless, investigators keep the flawed, fragile story flickering by employing stimuli and conditions whose results may be, though loosely and permissively, interpreted according to such crude assumptions. That the story is woefully inadequate (that investigators don’t know what factors to control so as to, at least, consistently interpret results) is reflected in the perennial admission, as in this article, that psychophysical experiments get conflicting results which investigators are unable to rationalize:

“Differences in the outcomes of different psychophysical procedures have already been noted elsewhere, and perhaps deserve more attention…

“Although our manipulation of attention did not produce a directional aftereffect, Lankheet and Verstraten’s (1995) manipulation of attention did. The reason for this discrepancy remains unclear…” Morgan et al (2016, p. 2630).

“…our results (Exp 4) did not confirm the factual basis for the claim (Blaser et al , 2005) that a 90 degree probe…we cannot be certain why our results are different…Differences include the psychophysical method (2AFC rather than MSS…), the statistical methods of analysis, the use of colors that appeared equally salient to the observer….”

If the authors are saying that even the statistical methods of analysis can potentially produce opposing interpretations of data, but investigators don’t understand how or why, what can possibly be the theoretical value of their own data/interpretations? How are we supposed to evaluate these things, if theoretical control of conditions is so inadequate and interpretation of experimental results so mystifyingly wide open and susceptible to inexplicable “butterfly effects”?