On 2017 Jun 09, Lydia Maniatis commented:

The idea that one can define a spatial extent (an aperture) for which perception of faces is equivalent to perception without such a constraint, is not credible, and the conclusions not interpretable; the conditions under which it was defined here make it all the more uninterpretable. The large difference local details of stimulation can make on the organization of the resulting percept, and the contingency of the effect of that detail on the all of the other local details and their relationship to the entire collection, are facts not compatible with simply additive, spatially-defined predictions.

In addition, the experiments are highly confounded and the authors address these confounds post hoc, using a hodge podge of analytical tools that entail many untested, untestable, and very likely false, assumptions. Among these is the "Random Field Theory" used to assess similarity between images of faces, which involves a pixel-by-pixel definition of similarity that, given the nature of perception and particularly face perception, is not credible. (It is certain that two faces might have a strong family resemblance which would not be correlated with a pixel-by-pixel definition. Similarity is a notoriously difficult concept and in practice comparisons are highly selective).

Given the post hoc ad hoc style of the analysis and weak theoretical framework, it is virtually certain that these results could not be replicated using a different sample of images understood as "faces" (the choice of the sample images has no more specificity than that) but otherwise identical procedures and analysis. This, the authors make pretty clear:

"Following this idea, we emphasize that the Facespan should not be considered as an absolute quantity. Inasmuch as the perceptual span for reading is not absolute, but instead flexible, the Facespan reported here should be considered as an average benchmark obtained under the aforementioned specific viewing conditions and task."

Obtained - and re-obtainable? It seems contradictory to suggest that a flexible outcome, the parameters mediating whose flexibility are undetermined, can serve as a benchmark, average or otherwise. We're left with a vague, less than credible, not-really-quantified concept - the Facespan.

On 2017 Jun 09, Lydia Maniatis commented:

The idea that one can define a spatial extent (an aperture) for which perception of faces is equivalent to perception without such a constraint, is not credible, and the conclusions not interpretable; the conditions under which it was defined here make it all the more uninterpretable. The large difference local details of stimulation can make on the organization of the resulting percept, and the contingency of the effect of that detail on the all of the other local details and their relationship to the entire collection, are facts not compatible with simply additive, spatially-defined predictions.

In addition, the experiments are highly confounded and the authors address these confounds post hoc, using a hodge podge of analytical tools that entail many untested, untestable, and very likely false, assumptions. Among these is the "Random Field Theory" used to assess similarity between images of faces, which involves a pixel-by-pixel definition of similarity that, given the nature of perception and particularly face perception, is not credible. (It is certain that two faces might have a strong family resemblance which would not be correlated with a pixel-by-pixel definition. Similarity is a notoriously difficult concept and in practice comparisons are highly selective).

Given the post hoc ad hoc style of the analysis and weak theoretical framework, it is virtually certain that these results could not be replicated using a different sample of images understood as "faces" (the choice of the sample images has no more specificity than that) but otherwise identical procedures and analysis. This, the authors make pretty clear:

"Following this idea, we emphasize that the Facespan should not be considered as an absolute quantity. Inasmuch as the perceptual span for reading is not absolute, but instead flexible, the Facespan reported here should be considered as an average benchmark obtained under the aforementioned specific viewing conditions and task."

Obtained - and re-obtainable? It seems contradictory to suggest that a flexible outcome, the parameters mediating whose flexibility are undetermined, can serve as a benchmark, average or otherwise. We're left with a vague, less than credible, not-really-quantified concept - the Facespan.