The Visual Acuity of Rats in Touchscreen Setups

Round 1

Reviewer 1 Report

The purpose of this study was to examine visual acuity in rats using touchscreen technology. The researchers used a pairwise visual (orientation) discrimination task with oblique sinusoidal gratings as stimuli. Visual acuity was tested by increasing the number of cycles/screen and further extrapolated acuity in c/degree with different viewing distances in order to translate vision research using other methodologies and aid in future touchscreen research experimental design. The results are of interest to the wider touchscreen community which seeks to improve cognitive translation from rodents into the clinic as well as vision scientists.

Major Concerns:

Page 2, lines 55-58. If the researchers are using oblique stimuli, why would they assume a theoretical decision boundary of 1 c/degree instead of a lower acuity of 0.7 c/degree which is reflected in the literature the authors have cited? Please justify. Figure 2: I think that for A) a box plot (dots for each of the 8 rats and n of sessions to criterion on y axis) with the mean and SD would provide better information graphically compared with the current bar graph. The x-axis labels are confusing and need to be amended. For % correct responses (B), I think that a line graph showing the mean and SEM for the final two sessions would be best (i.e., two dots on the graph for each of the two days). I make these suggestions based on that data that the authors emphasize in the text. Discussion Section: Motivation issues are mentioned. 1) The authors should look at the latency to reward magazine which provides some insights into motivation. For example, did the rats with lower acuity have longer latencies? If so, it could suggest that motivation is confounding the experiment. Furthermore, the researchers should clarify whether the rats always completed 100 within 60 minutes. If they didn’t, were they less likely to complete the trials as the discriminations got harder. Note the authors could also look at whether the number of correction trials increased along with task difficulty (which would be expected). Discussion Section 2: The authors should comment on whether prior discrimination experience can impact their study. For instance, if animals received the harder stimulus sets first, would they discriminate as well as the incremental method used here?

Minor Concerns:

Page 2, lines 35-38: “Typical stimulus parameters include spatial frequency (SF), orientation, contrast ([11]), velocity and direction of motion. Most of these can be easily defined on the screen, but the position of the animals head cannot be controlled, which may distort the parameters at the moment of decision or response.” This is also true of non-touchscreen vision tasks where an animal is freely moving (not head fixed).  The way this paragraph is arranged it appears as though this statement is specifically problematic for touchscreen technology alone. Please amend. Page 2, lines 78-79: change 45deg or 135deg to “compared with” or “versus”. It’s unclear as it’s written whether some animals are given 45deg stimuli only and other 135deg stimuli, but they are the choices the animals are given. To further drive this point home you could add a sentence saying that for half the animals, the 45deg was reward (S+) and the 135deg resulted in a time out (S-). There is a sentence on page 3 (lines 83-84) which you could bring above mentioning what happens when a stimulus is correct or incorrect. You can further include this in Figure 1 (not necessary, but may further clarify). Page 4, line 116: “the performance shown in figure 3 was averaged across less animals.” Change “less” to “fewer”. Figure 3. Some of the data points represent more than one session for the same animal. Could you please indicate this error bars or provide a range of how many sessions each point contains? Figure 4. The animal represented by the purple doesn’t go below 70% in this graph. Does that mean that this animal had a performance score above 70% for all c/screen tested? Figure 6. I believe that the 44c/screen and 15c/screen labels are mixed? Page7, lines 172-173: “As discussed above”. Remove as it was not. It’s discussed at length below in two separate sections. I suggest considering restructuring the discussion to increase the flow. Page 7, line 182: “touchscreens contains much more trials per daily session” should be changed to “many more” Page 7 line 189: Rats are not actually required to touch the screen for a response to be recorded. IR beams located a few mm in front of the screen detect responses. Though it may be a small difference, it may make a difference to the animals and is factually incorrect. Page 8, line 196: “In other words,” instead of “Or phrased differently” Page 8, line 221-222: This sentence is awkward and needs to be re-worked.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

This is an interesting methodological study. One contribution is to show the effectiveness of using touch-screens, with freely moving rats, to perform visual acuity tasks. Another is to address the question at to the likely distance at which judgements were made. There is inherently an ambiguity about distance and the retinal spatial frequency, since it is not know at what distance the judgement is made. I have a few comments on how I think the manuscript can be clarified and its contribution explored.

What were the luminance and contrast of the stimuli? These are likely to affect contrast sensitivity and acuity.

Oblique gratings were used - is there an oblique effect in rats (e.g. is sensitivity lower here than for horizontal and vertical stimuli, at these high frequencies) and do comparisons with previous acuity value take this into account?

In the results, you assume possible viewing distances. There are however 2 ways to address this. Either you assumed the viewing distance and calculate acuity, or you assume acuity and calculate viewing distance. It is this second which I assumed from the introduction. I think it would be helpful set this out more clear on page 4. Although you do address this point, as the effective viewing distance appears too far.

This ambiguity exists due to the constraints of your experimental design, which is one that is appropriate to mimic the intended applications. I can see that it is impossible to know that point at which the rat 'decided' which response to give. However, it would be helpful to consider, in the discussion, possible alternative designs.Would it for example be possible to test in a situation where the rats' movements were more constrained, so that is was only possible to see the images from a particular distance. Another alternative would be to track their location, such that images were only presented when they were within a range of distances. These are just suggestions, I appreciate that this is not the desired use-case for this technology. However, if it were possible to get a more precise measure of actual retinal acuity for you stimuli, you could be more confident in fixing this value and working back to infer the decision distance in a free moving situation.

A mention ethical concerns above as there is not mention of study approval and these details should be included.

minor grammar point- p7line 188 should be "animals' behavior"

Author Response

Please see the attachment

Author Response File: Author Response.pdf