On 2016 Sep 26, Lydia Maniatis commented:

Pelli and Bex (2013) make the following assertion:

"Neurally, Campbell and Robson (1968) revealed the presence of multiple channels in vision, each selective to a different band of spatial frequencies. This greatly increased interest in measuring the CSF. Today, the set of thresholds as a function of spatial frequency is usually fit with a contrast sensitivity function (Watson, 2000)."

The claim regarding Campbell and Robson (1968) is false.

It has been made previously by Pelli and coauthors, (and others), e.g. by Solomon and Pelli (1994). In a comment on that article (https://pubpeer.com/publications/6EA25EFC758D6B5C990CFDFD5899BD) I quote Robson and Campbell (1968), as follows:

"Thus, it seems that we cannot satisfactorily model the over-all visual system by a simple peak detector following a spatial filter...As a modification of this theory, we may assume...Thus, we may suppose that the visual system behaves not as a single detector mechanism preceded by a single broadband spatial filter but as a number of independent detector mechanisms each preceded by a relatively narrowband filter 'tuned' to a different frequency...Such a model could account for our findings...

In other words, Robson and Cambell's (1968) results did not bear out their original predictions, and they discuss purely speculative alternatives to account for their data. Between 1968 and 2013, no firmer references apparently became available.

Nevertheless, as is evident from the second part of Pelli and Bex (2013) statement, measurements based on Robson and Campbell's (1968) uncorroborated assumptions have become very popular. These measurements involve "fitting" a "constrast sensitivity function." As Pelli and Bex (2013) note, this fitting involves a minimum of four free parameters. What does this mean?

According to Wikipedia, a free parameter "is a variable in a mathematical model which cannot be predicted precisely or constrained by the model..." The mathematician and physicist Jon von Neumann is quoted as saying that "With four parameters I can fit an elephant, and with five I can make him wiggle his trunk." Investigators at the Max Planck institute have, in fact, fit an elephant with four free parameters (https://publications.mpi-cbg.de/Mayer_2010_4314.pdf).

What four free parameters means, in short, is that the model doesn't have to predict anything. The model assumptions (e.g., "the presence of multiple channels in vision, each selective to a different band of spatial frequencies") can be false, but still be fit to the (qualitatively generic) shape of the data.

The authors are advocating a method to be used to generate a certain score for individuals. But given the situation described above, we really can't say what that score means, in any theoretical sense.

On 2016 Sep 26, Lydia Maniatis commented:

Pelli and Bex (2013) make the following assertion:

"Neurally, Campbell and Robson (1968) revealed the presence of multiple channels in vision, each selective to a different band of spatial frequencies. This greatly increased interest in measuring the CSF. Today, the set of thresholds as a function of spatial frequency is usually fit with a contrast sensitivity function (Watson, 2000)."

The claim regarding Campbell and Robson (1968) is false.

It has been made previously by Pelli and coauthors, (and others), e.g. by Solomon and Pelli (1994). In a comment on that article (https://pubpeer.com/publications/6EA25EFC758D6B5C990CFDFD5899BD) I quote Robson and Campbell (1968), as follows:

"Thus, it seems that we cannot satisfactorily model the over-all visual system by a simple peak detector following a spatial filter...As a modification of this theory, we may assume...Thus, we may suppose that the visual system behaves not as a single detector mechanism preceded by a single broadband spatial filter but as a number of independent detector mechanisms each preceded by a relatively narrowband filter 'tuned' to a different frequency...Such a model could account for our findings...

In other words, Robson and Cambell's (1968) results did not bear out their original predictions, and they discuss purely speculative alternatives to account for their data. Between 1968 and 2013, no firmer references apparently became available.

Nevertheless, as is evident from the second part of Pelli and Bex (2013) statement, measurements based on Robson and Campbell's (1968) uncorroborated assumptions have become very popular. These measurements involve "fitting" a "constrast sensitivity function." As Pelli and Bex (2013) note, this fitting involves a minimum of four free parameters. What does this mean?

According to Wikipedia, a free parameter "is a variable in a mathematical model which cannot be predicted precisely or constrained by the model..." The mathematician and physicist Jon von Neumann is quoted as saying that "With four parameters I can fit an elephant, and with five I can make him wiggle his trunk." Investigators at the Max Planck institute have, in fact, fit an elephant with four free parameters (https://publications.mpi-cbg.de/Mayer_2010_4314.pdf).

What four free parameters means, in short, is that the model doesn't have to predict anything. The model assumptions (e.g., "the presence of multiple channels in vision, each selective to a different band of spatial frequencies") can be false, but still be fit to the (qualitatively generic) shape of the data.

The authors are advocating a method to be used to generate a certain score for individuals. But given the situation described above, we really can't say what that score means, in any theoretical sense.