Evidence for perceptual discrimination between stimuli of different magnitudes based on Weber’s law have been detected in several different taxa, including visual discrimination (see specific comment #4 below) However, detecting this pattern in the swordtail X. helleri is important for increasing our understanding of selection on males due mate preferences, as well as selection on female mate preferences. The pattern detected in this study suggests that selection on male size will be diminished among larger males. It also suggests that it would be interesting to examine if this pattern varies across females depending on the costs/benefits of assessing male size. Given the extensive information about mate preference, this system provides an excellent opportunity to use this phenomenon to better understand preference for male size in this species.

I have one primary comment in addition to one concern. First, the authors might consider how they place their study in the context of a signaler’s quality. While I generally agree that many mate preferences are adaptive, thus often based on traits of mate quality, the results presented in this study do not necessarily rely on this type of true communication (benefits to signalers and receivers). Instead, cases where receivers are responding due to their sensory bias, or to a “cue” that has not evolved to be a true signal (benefits of the signaler), and could also follow Weber’s law. Indeed, it is not known why swordtail females prefer larger males, and in some studies sensory bias has been implicated (no need to invoke mate quality). Instead, I would suggest that the results from this study suggest that examining variation in this phenomenon would provide a valuable tool for studying adaptive mate preferences for male size in X. helleri

Second, as has been pointed out in several different studies (e.g. Nachev V, Stich KP, Winter Y 2013), that it is important to realize that behavioral studies estimate “…discrimination performance rather than the capacity for perceptual discrimination. Animals might perceive differences between the available options but distribute their visits more evenly between the alternatives, regardless of expected value.” This limitation needs to be discussed.

I have a few more specific comments

Specific Comments

1) Line 39. The word “something” here seems too vague. As you are discussing information theory and communication here, why not say “information”. However, as I noted above, not all male traits and female preferences for these traits have evolved to be “signals” and “responses” (i.e. in some cases the male traits are cues and the in others the female preferences are biases). Therefore, it might be better to keep your introduction more general.

Suggestion. Change the first two sentences to something like…”Given that there is signal variation, that may or may not be correlated with information about the receiver, it is often assumed that the receivers perceive this variation….”

2) Line 74. This would be the place to introduce the idea that we do not really know the benefits of preferring larger males, and it has been suggested that it could be a sensory bias. I think it would be interesting to point this out as the pattern you detected could be used to test this idea to see if there has been selection on females to discriminate differently depending on aspects of the females condition, for example.

3) Line 92. Do you mean “size” rather than “quality” here? I think adding in quality is not a good idea, as it is unclear if larger males are higher “quality” mates in some respect. Actually, it seems this sentence could end after “might weaken preference for the larger male”

4) Line 148. This statement (“Our results are the first demonstration of proportional processing of a visual signal”) does not appear to be true. For example:
Cammaerts, M. C., & Cammaerts, R. (2020). Weber’s law applies to the ants’ visual perception. Journal of Biology and Life Sciences, 11(2), 36-61.

5) Methods: What was the order in which females were tested, and did this influenced your results? So, for example, if females tested with video animations become less interested, as the males do not respond to the females, their preferences may become weaker in later tests. Information about the order in which females were tested is needed, as well as analysis to show how this influenced their strength of preference for the larger males.

In many ways the authors have improved the MS and have made an effort to address most of my concerns (both major and minor). However, I am disappointed in the authors response in relation to how unique their study is in relation to past work. I have tried to better explain my point below. In particular, I am concerned about one summary statement (Lines 169-170), and potentially missing relevant citations.

While the authors mention stimulus area in relation to visual modalities for studies of Weber’s law, (lines75-76) studies of discriminating between numbers are missing. Interesting, as the authors use numbers of items in Figure 1 to explain Weber’s law

This quote comes from the paper about weber’s law and its application to the ant’s visual perception that I mentioned in my previous review.

“Regarding the ability in discriminating numbers, the validity of Weber’s law has been demonstrated at a behavioral and a neural level, in humans, monkeys and birds (Dehaene, 2003; Nieder & Miller, 2003; Jordan & Brannon, 2006; Cantlon & Brannon, 2006; Merten & Nieder, 2008; Ditz & Nieder, 2016).”

Now consider the statement on lines 169-170: “Our results are the first demonstration of proportional processing of visual stimuli in a context that is ecologically relevant to the viewer and which elicit an untrained response”.

So, the authors of the swordtail study are implying that getting “more” of something in the number studies cited above are not ecologically relevant to the viewer??

I would say instead that the results of the proportional processing detected in the swordtail
study has implications for selection on the signaler… this could be the point made in relation to how this study is unique.

In addition, they also implying that all the studies based on discriminating numbers were trained responses?? I do not know this literature well, but they should check the literature more carefully to be sure that is the case before stating this here.