Open Access
Article  |   February 2019
The contrast sensitivity function of a small cryptobenthic marine fish
Author Affiliations
  • Matteo Santon
    Animal Evolutionary Ecology, Institute of Evolution and Ecology, Department of Biology, Faculty of Science, University of Tübingen, Tübingen, Germany
    matteo.santon@uni-tuebingen.de
  • Thomas A. Münch
    Retinal Circuits and Optogenetics, Institute for Ophthalmic Research, Department of Ophthalmology, and Centre for Integrative Neuroscience, University of Tübingen, Tübingen, Germany
    thomas.muench@uni-tuebingen.de
  • Nico K. Michiels
    Animal Evolutionary Ecology, Institute of Evolution and Ecology, Department of Biology, Faculty of Science, University of Tübingen, Tübingen, Germany
    nico.michiels@uni-tuebingen.de
Journal of Vision February 2019, Vol.19, 1. doi:10.1167/19.2.1
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Matteo Santon, Thomas A. Münch, Nico K. Michiels; The contrast sensitivity function of a small cryptobenthic marine fish. Journal of Vision 2019;19(2):1. doi: 10.1167/19.2.1.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Spatial resolution is a key property of eyes when it comes to understanding how animals' visual signals are perceived. This property can be robustly estimated by measuring the contrast sensitivity as a function of different spatial frequencies, defined as the number of achromatic vertical bright and dark stripe pairs within one degree of visual angle. This contrast sensitivity function (CSF) has been estimated for different animal groups, but data on fish are limited to two free-swimming, freshwater species (i.e., goldfish and bluegill sunfish). In this study, we describe the CSF of a small marine cryptobenthic fish (Tripterygion delaisi) using an optokinetic reflex approach. Tripterygion delaisi features a contrast sensitivity that is as excellent as other fish species, up to 125 (reciprocal of Michelson contrast) at the optimal spatial frequency of 0.375 c/°. The maximum spatial resolution is instead relatively coarse, around 2.125 c/°. By comparing our results with acuity values derived from anatomical estimates of ganglion cells' density, we conclude that the optokinetic reflex seems to be adapted to process low spatial frequency information from stimuli in the peripheral visual field and show that small marine fish can feature excellent contrast sensitivity at optimal spatial frequency.

Introduction
Vision shows such a diversity in anatomy, physiology, and performance across animal species that we cannot make predictions on the functionality of animals' visual signals based on our own human perception (Bennett, Cuthill, & Norris, 1994; Caves, Brandley, & Johnsen, 2018; Land & Nilsson, 2012). An important start to the investigation of such signals is to measure species-specific visual properties such as spectral sensitivity and spatial resolution (Caves, Frank, & Johnsen, 2016; Olsson, Lind, Kelber, & Simmons, 2017). Such properties are traditionally estimated anatomically, but they should also be assessed behaviorally (Kelber, Vorobyev, & Osorio, 2003): whereas anatomy provides estimates of the theoretical upper limit of vision, behavioral tests should match reality more closely (Caves et al., 2018). 
Spatial resolution (Land & Nilsson, 2012) plays a critical role in defining the active space (i.e., the maximum perception distance) of a visual signal (Caves et al., 2018; Caves et al., 2016). Because this property is affected by the contrast of the signal, spatial resolution is better estimated with a contrast sensitivity function (CSF; De Valois & De Valois, 1990; Uhlrich, Essock, & Lehmkuhle, 1981). This approach is more comprehensive than measuring maximum spatial resolution only (Lind & Kelber, 2011), as it expresses contrast sensitivity as a function of spatial frequency (number of vertical bright and dark stripe pairs within one degree of visual angle). 
The CSF has been behaviorally estimated for at least two insects (Chakravarthi, Baird, Dacke, & Kelber, 2016; Srinivasan & Lehrer, 1988), seven birds (Harmening, Nikolay, Orlowski, & Wagner, 2009; Hirsch, 1982; Hodos, Ghim, Potocki, Fields, & Storm, 2002; Jarvis, Abeyesinghe, McMahon, & Wathes, 2009; Lind & Kelber, 2011; Lind, Sunesson, Mitkus, & Kelber, 2012; Reymond & Wolfe, 1981), and 13 mammals (Birch & Jacobs, 1979; Bisti & Maffei, 1974; De Valois, Morgan, & Snodderly, 1974; Hanke, Scholtyssek, Hanke, & Dehnhardt, 2011; Jacobs, 1977; Jacobs, Birch, & Blakeslee, 1982; Jacobs, Blakeslee, McCourt, & Tootell, 1980; Langston, Casagrande, & Fox, 1986; Merigan, 1976; Petry, Fox, & Casagrande, 1984). Data on fish are limited to two free-swimming, freshwater species (i.e., goldfish and bluegill sunfish; Bilotta & Powers, 1991; Northmore & Dvorak, 1979; Northmore, Oh, & Celenza, 2007). Although marine fish are a common subject of visual ecology studies, according to our knowledge a CSF has not been determined for any of them. 
Here, we describe the CSF of a small marine cryptobenthic (i.e., bottom-living and camouflaged) fish, the triplefin Tripterygion delaisi, that lives in complex coastal hard-bottom environments where interactions with other species are frequent (e.g., prey or predators) and therefore good contrast sensitivity and spatial resolution can be expected to be advantageous. 
We used an optokinetic reflex approach, which is traditionally elicited by placing an animal in the center of a rotating drum featuring a vertical grating. The optokinetic reflex consists of rotational eye movements to improve stabilization against relative motion of the environment rather than visual tracking of an individual object. It is therefore assumed to primarily depend on peripheral vision in animals that feature a fovea (Caves et al., 2018; Douglas & Djamgoz, 2012). The spatial resolution estimate obtained for T. delaisi from the CSF coincides well with previous anatomy-based resolution estimates of the peripheral retina (Fritsch, Collin, & Michiels, 2017), and the extremely fine contrast sensitivity measured is comparable to two freshwater fish species (Bilotta & Powers, 1991; Douglas & Djamgoz, 2012; Northmore & Dvorak, 1979; Northmore et al., 2007). 
Materials and methods
Model species
The triplefin Tripterygion delaisi is a small (4–5 cm) northeast Atlantic and Mediterranean cryptobenthic fish, common in rocky coastal areas between 5 m to below 20 m depth (De Jonge & Videler, 1989; Domingues, Almada, Santos, Brito, & Bernardi, 2007). Preferred prey are small crustaceans (Zander & Hagemann, 1989), which are caught with sudden strikes over distances between 1–3 cm (unpublished data). During the breeding season (March through May), individuals show a sex color dimorphism: males develop black heads and a bright yellow body; females maintain the partially translucent coloration with red irides featured by individuals of both sexes throughout the rest of the year (Bitton et al., 2017). We collected individuals close to the Station de Recherches Sous-marines et Océanographiques (STARESO) near Calvi, Corsica (France). Sampling took place under the general sampling permit of the station. Tripterygion delaisi is a non-threatened, non-protected, non-commercial, common species. 
Fish collection and housing
Animals were transported individually in plastic fish breathing bags™ (Kordon, Hayward, CA) filled with 250 ml purified seawater. At the University of Tübingen, fish were kept in individual tanks (L × W × H = 24 × 35 × 39 cm3) illuminated by weak, diffuse blue light. Coral sand covered the bottom, and a rock was provided as shelter. All aquaria were interconnected to a flow-through filtering and UV-sterilization system (23°C, salinity 35‰, pH 8.2, 12 h light/dark cycle). Water quality was checked on a weekly basis. Fish were fed with a mixture of Tetramin (Hauptfutter für alle Zierfische; Tetra GmbH, Melle, Germany) and Mysis (Einzelfuttermittel; Aki Frost GmbH, Ganderkesee, Germany) every day. Animal husbandry was carried out in accordance with German animal welfare legislation. Because the individuals were not experimentally manipulated, a formal permit was not required for this study (as confirmed by the Animal Care Officer at the Biology Department of the University of Tübingen). 
Optokinetic virtual arena
To characterize the contrast sensitivity function we observed individuals (N = 10) in the center of an “optokinetic drum” (Benkner, Mutter, Ecke, & Münch, 2013) while exposed to horizontally rotating grayscale vertical striped patterns defined by combinations of spatial frequency (cycles per degree) and contrast (Michelson) calculated as (L1L2)/(L1 + L2), where L1 and L2 are the photon radiances of the bright and dark stripes. As is true for many benthic teleosts (Fritsches & Marshall, 2002), triplefins show independent eye movement (Michiels et al., 2018). Their optokinetic reflex consists of tracking behavior (at least one eye or the body following the rotation direction of the pattern), often mixed with opposing saccades (sudden eye movement in the opposite direction of the pattern movement; Supplementary Movie S1). 
The virtual arena (OptoDrum software; Striatech UG, Tübingen, Germany; L × W × H: 53 × 53 × 30 cm3) consisted of four 23.8 in. LCD monitors (EIZO EV2450) set to DICOM presentation mode. Individuals were placed in a transparent glass cylinder (D × H: 6.5 × 10 cm2) with a black bottom, filled with home tank water and positioned on an elevated circular platform (Figure 1). The bottom and the top of the arena consisted of two mirrors to ensure that the fish would see the pattern even if looking up or down (Figure 1). Circular holes in the mirrors (D = 10 cm) were used for inserting the platform, and to position a Canon EOS 7D (Canon Inc., Tokyo, Japan) with a 100 mm macro lens above the setup to record the fish. The width of the horizontally moving vertical stripes on the screens was electronically widened toward the corners to maintain the same angular resolution from the center of the setup. For the fish in the setup, this generates the optical illusion of being in the center of a cylindrical drum (Figure 2). Overall brightness in the setup was kept constant across all stimuli at the value obtained when all four screens were set at 50% homogeneous gray. Michelson contrasts between the stripes ranged from of 0.8% to 99.7%. These values represent the smallest and largest contrasts the setup could generate. The total photon radiance (integrated from 380 to 780 nm) of a polytetrafluoroethylene (PTFE) white reflectance standard (Lake Photonics, Uhldingen-Mühlhofe, Germany) placed flat on the platform surrounded by uniform 50% gray screens was measured from above using a calibrated SpectraScan PR 670 spectroradiometer (Photo Research, Syracuse, NY) and was 1.13 × 1017 photons s−1 sr−1 m−2 nm−1, similar to the total radiance measured in a comparable way in the shade around 20 m depth in the field where this species occurs (Harant et al., 2018). The OptoDrum software (Striatech) allows the free adjustment of the perceived width (spatial frequency), contrast, rotation direction (left or right) and angular speed of the striped pattern. 
Figure 1
 
Virtual arena inside the optokinetic drum. Individual fish were placed in a central transparent glass cylinder positioned on an elevated circular platform. The four screens surrounding the fish displayed the grating pattern. Mirrors on the bottom and top of the setup assured the grating to be visible in a radial pattern (not visible from the camera perspective shown here). Photo credit: Matteo Santon.
Figure 1
 
Virtual arena inside the optokinetic drum. Individual fish were placed in a central transparent glass cylinder positioned on an elevated circular platform. The four screens surrounding the fish displayed the grating pattern. Mirrors on the bottom and top of the setup assured the grating to be visible in a radial pattern (not visible from the camera perspective shown here). Photo credit: Matteo Santon.
Figure 2
 
Top view of the optokinetic drum virtual arena. A fish in the center of a glass cylinder experiences the moving grating pattern on the screens as a rotating cylinder. Photo credit: Matteo Santon.
Figure 2
 
Top view of the optokinetic drum virtual arena. A fish in the center of a glass cylinder experiences the moving grating pattern on the screens as a rotating cylinder. Photo credit: Matteo Santon.
Behavioral tests
We acclimated fish with uniform 50% gray screens for 5 min. Every test consisted of several sessions (quantity depending on the fish performance), in which a fish was recorded while looking at a rotating grating. Each fish (N = 10) was tested once. Following preliminary tests, we kept the rotation at the speed of 4°/s, which was optimal for this species in this experimental setup. We randomized the order in which different spatial frequencies were displayed and alternated the rotation direction between sessions. As a transition to the next stimulus, a uniform 50% gray display was shown for 10 to 20 s. We live-evaluated and recorded a fish's response to spatial frequencies from 0.125 c/° to the individual perception threshold in steps of 0.250 c/°. Within each spatial frequency step, we gradually reduced the contrast of the stripes starting from the maximum of 99.7% down to the contrast value at which the fish did not show a response anymore. To minimize the number of sessions and therefore reduce potential stress generation in the fish, the descending contrast steps displayed were slightly different for each individual in accordance with previous fish responses at the different spatial frequencies tested. The contrast sensitivity limit at any given spatial frequency was determined as the weakest contrast still eliciting a response. As a control, we recorded the response to 50% gray uniform screens for three minutes. All videos were zoomed in to such an extent that only the snout and eyes of the fish were visible, but not the stimulus (Supplementary Movie S1). Run duration was decided by the experimenter based on the live image and ranged from 1 min (if a fish showed immediate, unequivocal reflexes) to maximum 3 min (when a fish showed weak or no reaction). Fish that showed tracking behavior at least once or several saccadic reflexes during the entire live run were considered to have perceived the stimulus. To confirm these live assessments, a second observer naïve to the experiment evaluated the recordings in random order. The second observer assessed if the fish was showing an optokinetic reflex, and also in which direction. To minimize false positives, a fish was considered “non-responding” when reflex-like behavior was detected but its direction was not in accordance with the rotation direction of the grating. Also, to further reduce the risk of false positives, only the sessions in which the first observer (live assessment) and the second observer (unbiased assessment) noted a response were counted as positives. 
Data analysis
Behavioral data were analyzed using generalized linear mixed effects models (gamma distribution, link = log) with the glmmTMB package (Brooks et al., 2017) for R v. 3.4.3 (R Core Team, 2017). We used contrast sensitivity as response variable, the main predictor spatial frequency and rotation direction as factorial fixed components, and individual ID as random component. We performed backward model selection using the Akaike information criterion (AIC) to identify the best fitting model with the smallest number of covariates (Zuur, Ieno, Walker, Saveliev, & Smith, 2009). We only report the final reduced model and its overall goodness-of-fit (conditional R2; i.e., the proportion of variation explained by the model considering fixed and random factors [Nakagawa & Schielzeth, 2010]). Model assumptions were validated by plotting residuals versus fitted values and each covariate present in the full, non-reduced model. We used Wald z-tests to assess the significance of fixed effects. Multiple comparisons of contrast sensitivity among different spatial frequencies were computed by using 95% credible intervals (CrIs), a Bayesian analogue of confidence intervals (Bolker et al., 2009). For the response variable contrast sensitivity, we computed model-predicted means and the associated 95% CrIs from 10,000 simulations of the model, using the simulate function of the R package stats. If the mean sensitivity at one spatial frequency falls outside the credible interval of another frequency, the difference between the two groups is significant. 
To visualize the relationship between contrast sensitivity and spatial frequency on a continuous scale, we also generated an analogue model that used spatial frequency as a continuous predictor and that included a quadratic and a cubic term to compensate for non-linear patterns (tested using the gam function of the R package mgcv (Wood, 2006)). This model was only used to generate the smoothing curve in Figure 3
Figure 3
 
Contrast sensitivity function (CSF) of Tripterygion delaisi. Contrast sensitivity, expressed as the reciprocal of the Michelson contrast, as a function of spatial frequency. The optimal sensitivity was around 0.375 c/°, where seven out of 10 fish showed the maximum sensitivity of 125, which corresponds to a 0.8% Michelson contrast. Contrast sensitivity is plotted on a log-scale. Numbers at the top indicate the number of individuals (N = 10) responding to each spatial frequency. None of the individuals responded to the spatial frequency of 2.375 (not shown). Each color of the points represents a different individual (jittered for clarity). Error bars display the model-predicted group means ± 95% credible intervals. The black curve shows the relationship between contrast sensitivity and spatial frequency on a continuous scale (see Materials and methods section for details).
Figure 3
 
Contrast sensitivity function (CSF) of Tripterygion delaisi. Contrast sensitivity, expressed as the reciprocal of the Michelson contrast, as a function of spatial frequency. The optimal sensitivity was around 0.375 c/°, where seven out of 10 fish showed the maximum sensitivity of 125, which corresponds to a 0.8% Michelson contrast. Contrast sensitivity is plotted on a log-scale. Numbers at the top indicate the number of individuals (N = 10) responding to each spatial frequency. None of the individuals responded to the spatial frequency of 2.375 (not shown). Each color of the points represents a different individual (jittered for clarity). Error bars display the model-predicted group means ± 95% credible intervals. The black curve shows the relationship between contrast sensitivity and spatial frequency on a continuous scale (see Materials and methods section for details).
All data were processed using R v. 3.4.3 (R Core Team, 2017). Means are shown ± standard deviation unless specified otherwise. Contrast sensitivity values are shown as the reciprocal of the Michelson contrasts values. Raw data to perform the analyses are provided in the Appendix (Table A1). 
Results
We obtained the contrast sensitivity function of T. delaisi from 10 individuals by observing the smallest contrast that elicited the optokinetic reflex, in response to a vertical stripe pattern rotating around the fish horizontally at a speed of 4 °/s. We measured such contrast sensitivity threshold at 10 different spatial frequencies of the stripe pattern, ranging from 0.125 c/° to 2.375 c/° in equal steps of 0.250 c/°. The results are displayed as colored dots in Figure 3, where each individual is identified by a unique color. We analyzed these data by fitting a generalized linear mixed effects model (see Materials and methods for details). Model validation did not show any violation of the model assumptions. The direction of rotation of the drum was dropped during model selection. The final model only contained spatial frequency as fixed factor (generalized linear mixed effects model: R2cond: 0.86, spatial frequency: p < 0.0001). At the highest stimulus contrast (99.7% Michelson contrast, equal to a contrast sensitivity of 1), the high frequency cutoff was at 2.125 c/° (Figure 3), where three fish out of 10 still showed optokinetic reflexes. None of the fish responded to the stimulus at 2.375 c/° (not shown in Figure 3). On average, fish showed a high spatial frequency cutoff of 1.8 ± 0.3 c/°. The optimal spatial frequency eliciting an optokinetic reflex was at 0.375 c/°, with seven out of 10 fish still showing responses to a Michelson contrast as low as 0.8%, equal to a contrast sensitivity of 125 (Figure 3). At this spatial frequency, the model estimated an average contrast sensitivity around 90.33 (95% credible interval from 53.1 to 147.6; Figure 3). At the lowest experimental spatial frequency (0.125 c/°), contrast sensitivity significantly dropped to the model-estimated mean value of 31.4 (95% credible interval from 18.5 to 51.2; Figure 3). The characteristic optokinetic eye movements were also never observed with a uniform 50% gray screen control. 
Discussion
The shape of the contrast sensitivity function of T. delaisi is consistent with the one found in most species, a classic inverted U-profile (Figure 3; Uhlrich et al., 1981) and shows a maximum contrast sensitivity of 125 (0.8% Michelson contrast, lowest testable contrast) at a spatial frequency of 0.375 c/°. This maximum contrast sensitivity value is comparable to what is known from freshwater free-swimming fish species (sensitivity up to 111; Bilotta & Powers, 1991; Douglas & Djamgoz, 2012; Northmore & Dvorak, 1979; Northmore et al., 2007), and higher than in most of birds tested so far (sensitivity up to 20; Ghim & Hodos, 2006; Harmening et al., 2009; Lind & Kelber, 2011). For fish, high contrast sensitivity may help to maximize contrast perception in aquatic environments, where contrasts are usually strongly degraded because of light scatter by suspended particles (Douglas & Djamgoz, 2012). 
At higher spatial frequencies, sensitivity drops until it reaches a cutoff at 2.125 c/°. This relatively coarse spatial resolution limit suggests an adaptation of the optokinetic reflex to process low spatial frequency information from stimuli in the peripheral visual field. This statement is further supported by the fact that underwater light fields tend to act as a high spatial-frequency cutoff filter, where the light scattered by suspended particles blurs the edges of objects at distance (Douglas & Djamgoz, 2012). 
The decrease of sensitivity at low spatial frequency probably represents the center-surround receptive field properties of the retinal neurons perceiving the moving environment. 
A recent description of the retinal distribution of photoreceptors and ganglion cells in T. delaisi (Fritsch et al., 2017) estimated that its spatial resolution lies between 6.7 to 9 c/° for the fovea, and 2.4 c/° for the peripheral retina. This fits well with the spatial resolution estimated behaviorally in this study (2.125 c/°), assuming that the optokinetic reflex is based on peripheral vision. This close match suggests that the anatomical foveal acuity estimates of the same study (Fritsch et al., 2017) might also be close to reality. 
It has been already suggested that spatial resolution measures based on the optokinetic reflex in animals featuring a fovea underestimate their maximum acuity (Caves et al., 2018). To obtain a behavioral assessment of foveal spatial resolution, a different approach is required. One possibility is to induce animals to track small visual targets rather than wide-field stimuli. This can be achieved for example by training species to distinguish vertically from horizontally striped targets (Champ, Wallis, Vorobyev, Siebeck, & Marshall, 2014; Nakamura, 1968; Yamanouchi, 1956) or a specific black shape from a uniform gray background (Champ et al., 2014). 
Considering the ecology of T. delaisi, the anatomically estimated spatial resolution of 6.7 c/° (fovea) or 2.4 c/° (periphery) makes a significant difference. At a distance of 20 cm, peripheral vision would roughly allow the resolution of conspecifics, while foveal vision would reach the same limit at four times that distance (Figure 4). However, the resolving power of peripheral vision would still be sufficient to detect the outline of predators (which are about four times larger than T. delaisi) up to 80 cm (Figure 4), a distance that still allows for fast fleeing responses. This suggests that foveal vision could function to search for small prey when foraging or for intraspecific interactions, while peripheral vision could at least allow the perception of bigger predators. 
Figure 4
 
A Tripterygion delaisi's view of a conspecific and of a predator. Foveal acuity (first column) allows resolution of conspecifics (T. delaisi) at a greater distance than peripheral acuity (second column), while both acuities allow to resolve the outline of a potential predator (Serranus scriba) up to 80 cm (third and fourth column). Each image shows the spatial information content of the scene, adjusted for spatial resolution and distance. We also assume that the total length of T. delaisi in the picture is 5 cm, and the length of S. scriba is 20 cm. All manipulated images have been generated using the R package AcuityView (Caves & Johnsen, 2018). Photo credits: Matteo Santon.
Figure 4
 
A Tripterygion delaisi's view of a conspecific and of a predator. Foveal acuity (first column) allows resolution of conspecifics (T. delaisi) at a greater distance than peripheral acuity (second column), while both acuities allow to resolve the outline of a potential predator (Serranus scriba) up to 80 cm (third and fourth column). Each image shows the spatial information content of the scene, adjusted for spatial resolution and distance. We also assume that the total length of T. delaisi in the picture is 5 cm, and the length of S. scriba is 20 cm. All manipulated images have been generated using the R package AcuityView (Caves & Johnsen, 2018). Photo credits: Matteo Santon.
In conclusion, this study shows that the optokinetic reflex can be reliably used to estimate the spatial resolution of the peripheral retina and that this small marine cryptobenthic fish features excellent contrast sensitivity of up to 125. These estimates are important in planning future behavioral experiments and to inform visual models based on the vision of small marine benthic fishes. 
Acknowledgments
The authors warmly thank Valentina Richter for the evaluation of the recordings, Oeli Oelkrug for fish maintenance at the University of Tübingen, and Gregor Schulte for invaluable IT assistance. We further thank Dr. Pierre-Paul Bitton, Dr. Nils Anthes, and Dr. Martin J. How for fruitful suggestions or discussions. 
MS conceptualized the study, performed data collection, and analyzed the data. TAM provided the setup and the OptoDrum software. MS drafted the manuscript. All authors discussed the data, edited, and finalized the manuscript. 
The project was funded by the by the German Science Foundation Koselleck grant (Mi482/13-1) and the Volkswagen Foundation (Az. 89148 and Az. 91816) awarded to NKM. Development of the OptoDrum device was supported by funds from the University of Tübingen (Innovation Grant). TAM received support from the Kerstan foundation. 
Commercial relationships: none. Although TAM is co-owner of Striatech UG, who is the manufacturer of the optokinetic arena used for this study, all data collection and analysis was performed by MS. 
Corresponding author: Matteo Santon. 
Address: Animal Evolutionary Ecology, Institute of Evolution and Ecology, Department of Biology, Faculty of Science, University of Tübingen, Tübingen, Germany. 
References
Benkner, B., Mutter, M., Ecke, G., & Münch, T. A. (2013). Characterizing visual performance in mice: An objective and automated system based on the optokinetic reflex. Behavioral Neuroscience, 127 (5), 788–796.
Bennett, A. T. D., Cuthill, I. C., & Norris, K. J. (1994). Sexual selection and the mismeasure of color. The American Naturalist, 144 (5), 848–860.
Bilotta, J., & Powers, M. K. (1991). Spatial contrast sensitivity of goldfish: Mean luminance, temporal frequency, and a new psychophysical technique. Vision Research, 31 (3), 577–585, https://doi.org/10.1016/0042-6989(91)90108-H.
Birch, D., & Jacobs, G. H. (1979). Spatial contrast sensitivity in albino and pigmented rats. Vision Research, 19 (8), 933–937, https://doi.org/10.1016/0042-6989(79)90029-4.
Bisti, S., & Maffei, L. (1974). Behavioural contrast sensitivity of the cat in various visual meridians. The Journal of Physiology, 241 (1), 201–210, https://doi.org/10.1113/jphysiol.1974.sp010649.
Bitton, P.-P., Harant, U. K., Fritsch, R., Champ, C. M., Temple, S. E., & Michiels, N. K. (2017). Red fluorescence of the triplefin Tripterygion delaisi is increasingly visible against background light with increasing depth. Royal Society Open Science, 4 (3): 161009, https://doi.org/10.1098/rsos.161009.
Bolker, B. M., Brooks, M. E., Clark, C. J., Geange, S. W., Poulsen, J. R., Stevens, M. H. H., & White, J.-S. S. (2009). Generalized linear mixed models: A practical guide for ecology and evolution. Trends in Ecology & Evolution, 24 (3), 127–135, https://dx.doi.org/10.1016/j.tree.2008.10.008.
Brooks, M. E., Kristensen, K., van Benthem, K. J., Magnusson, A., Berg, C. W., Nielsen, A.,… Bolker, B. M. (2017). glmmTMB balances speed and flexibility among packages for zero-inflated generalized linear mixed modeling. The R Journal, 9 (2), 378–400.
Caves, E. M., Brandley, N. C., & Johnsen, S. (2018). Visual acuity and the evolution of signals. Trends in Ecology & Evolution, 33 (5), 358–372, https://doi.org/10.1016/j.tree.2018.03.001.
Caves, E. M., Frank, T. M., & Johnsen, S. (2016). Spectral sensitivity, spatial resolution and temporal resolution and their implications for conspecific signalling in cleaner shrimp. The Journal of Experimental Biology, 219 (4), 597–608, https://doi.org/10.1242/jeb.122275.
Caves, E. M., & Johnsen, S. (2018). AcuityView: An R package for portraying the effects of visual acuity on scenes observed by an animal. Methods in Ecology and Evolution, 9 (3), 793–797.
Chakravarthi, A., Baird, E., Dacke, M., & Kelber, A. (2016). Spatial vision in Bombus terrestris. Frontiers in Behavioral Neuroscience, 10: 17, https://doi.org/10.3389/fnbeh.2016.00017.
Champ, C., Wallis, G., Vorobyev, M., Siebeck, U., & Marshall, J. (2014). Visual acuity in a species of coral reef fish: Rhinecanthus aculeatus. Brain, Behavior and Evolution, 83 (1), 31–42.
De Jonge, J., & Videler, J. (1989). Differences between the reproductive biologies of Tripterygion tripteronotus and T. delaisi (Pisces, Perciformes, Tripterygiidae): The adaptive significance of an alternative mating strategy and a red instead of a yellow nuptial colour. Marine Biology, 100 (4), 431–437.
De Valois, R. L., & De Valois, K. (1990). Spatial vision. Oxford, UK: Oxford University Press.
De Valois, R. L., Morgan, H., & Snodderly, D. M. (1974). Psychophysical studies of monkey Vision-III. Spatial luminance contrast sensitivity tests of macaque and human observers. Vision Research, 14 (1), 75–81, https://doi.org/10.1016/0042-6989(74)90118-7.
Domingues, V. S., Almada, V. C., Santos, R. S., Brito, A., & Bernardi, G. (2007). Phylogeography and evolution of the triplefin Tripterygion delaisi (Pisces, Blennioidei). Marine Biology, 150 (3), 509–519.
Douglas, R., & Djamgoz, M. (2012). The visual system of fish. Berlin, Germany: Springer Science+Business Media.
Fritsch, R., Collin, S. P., & Michiels, N. K. (2017). Anatomical analysis of the retinal specializations to a crypto-benthic, micro-predatory lifestyle in the Mediterranean triplefin blenny Tripterygion delaisi. Frontiers in Neuroanatomy, 11: 122, https://doi.org/10.3389/fnana.2017.00122.
Fritsches, K. A., & Marshall, N. J. (2002). Independent and conjugate eye movements during optokinesis in teleost fish. Journal of Experimental Biology, 205 (9), 1241–1252.
Ghim, M. M., & Hodos, W. (2006). Spatial contrast sensitivity of birds. Journal of Comparative Physiology A, 192 (5), 523–534, https://doi.org/10.1007/s00359-005-0090-5.
Hanke, F. D., Scholtyssek, C., Hanke, W., & Dehnhardt, G. (2011). Contrast sensitivity in a harbor seal (Phoca vitulina). Journal of Comparative Physiology A, 197 (2), 203–210, https://doi.org/10.1007/s00359-010-0600-y.
Harant, U. K., Santon, M., Bitton, P. P., Wehrberger, F., Griessler, T., Meadows, M. G.,… Michiels, N. K. (2018). Do the fluorescent red eyes of the marine fish Tripterygion delaisi stand out? In situ and in vivo measurements at two depths. Ecology and Evolution, 8 (9), 4685–4694, https://doi.org/10.1002/ece3.4025.
Harmening, W. M., Nikolay, P., Orlowski, J., & Wagner, H. (2009). Spatial contrast sensitivity and grating acuity of barn owls. Journal of Vision, 9 (7): 13, 1–12, https://doi.org/10.1167/9.7.13. [PubMed] [Article]
Hirsch, J. (1982). Falcon visual sensitivity to grating contrast. Nature, 300, 57–58, https://doi.org/10.1038/300057a0.
Hodos, W., Ghim, M. M., Potocki, A., Fields, J. N., & Storm, T. (2002). Contrast sensitivity in pigeons: A comparison of behavioral and pattern ERG methods. Documenta Ophthalmologica, 104 (1), 107–118, https://doi.org/10.1023/a:1014427615636.
Jacobs, G. H. (1977). Visual capacities of the owl monkey (Aotus trivirgatus)—I. Spectral sensitivity and color vision. Vision Research, 17 (7), 811–820.
Jacobs, G. H., Birch, D. G., & Blakeslee, B. (1982). Visual acuity and spatial contrast sensitivity in tree squirrels. Behavioural Processes, 7 (4), 367–375, https://doi.org/10.1016/0376-6357(82)90008-0.
Jacobs, G. H., Blakeslee, B., McCourt, M. E., & Tootell, R. B. H. (1980). Visual sensitivity of ground squirrels to spatial and temporal luminance variations. Journal of Comparative Physiology, 136 (4), 291–299, https://doi.org/10.1007/bf00657349.
Jarvis, J. R., Abeyesinghe, S. M., McMahon, C. E., & Wathes, C. M. (2009). Measuring and modelling the spatial contrast sensitivity of the chicken (Gallus g. domesticus). Vision Research, 49 (11), 1448–1454, https://doi.org/10.1016/j.visres.2009.02.019.
Kelber, A., Vorobyev, M., & Osorio, D. (2003). Animal colour vision—Behavioural tests and physiological concepts. Biological Reviews, 78 (1), 81–118, https://doi.org/10.1017/S1464793102005985.
Land, M. F., & Nilsson, D. E. (2012). Animal eyes (2nd ed.). Oxford, UK: Oxford University Press.
Langston, A., Casagrande, V. A., & Fox, R. (1986). Spatial resolution of the galago. Vision Research, 26 (5), 791–796, https://doi.org/10.1016/0042-6989(86)90094-5.
Lind, O., & Kelber, A. (2011). The spatial tuning of achromatic and chromatic vision in budgerigars. Journal of Vision, 11 (7): 2, 1–9, https://doi.org/10.1167/11.7.2. [PubMed] [Article]
Lind, O., Sunesson, T., Mitkus, M., & Kelber, A. (2012). Luminance-dependence of spatial vision in budgerigars (Melopsittacus undulatus) and Bourke's parrots (Neopsephotus bourkii). Journal of Comparative Physiology A, 198 (1), 69–77, https://doi.org/10.1007/s00359-011-0689-7.
Merigan, W. H. (1976). The contrast sensitivity of the squirrel monkey (Saimiri sciureus). Vision Research, 16 (4), 375–379, https://doi.org/10.1016/0042-6989(76)90199-1.
Michiels, N. K., Seeburger, V. C., Kalb, N., Meadows, M. G., Anthes, N., Mailli, A. A., & Jack, C. B. (2018). Controlled iris radiance in a diurnal fish looking at prey. Royal Society Open Science, 5: 170838, https://doi.org/10.1098/rsos.170838.
Nakagawa, S., & Schielzeth, H. (2010). Repeatability for Gaussian and non-Gaussian data: A practical guide for biologists. Biological Reviews, 85 (4), 935–956.
Nakamura, E. L. (1968). Visual acuity of two tunas, Katsuwonus pelamis and Euthynnus affinis. Copeia, 1, 41–49.
Northmore, D. P. M., & Dvorak, C. A. (1979). Contrast sensitivity and acuity of the goldfish. Vision Research, 19 (3), 255–261, https://doi.org/10.1016/0042-6989(79)90171-8.
Northmore, D. P. M., Oh, D. J., & Celenza, M. A. (2007). Acuity and contrast sensitivity of the bluegill sunfish and how they change during optic nerve regeneration. Visual Neuroscience, 24 (3), 319–331, https://doi.org/10.1017/S0952523807070307.
Olsson, P., Lind, O., Kelber, A., & Simmons, L. (2017). Chromatic and achromatic vision: Parameter choice and limitations for reliable model predictions. Behavioral Ecology, 29 (2), 273–282.
Petry, H. M., Fox, R., & Casagrande, V. A. (1984). Spatial contrast sensitivity of the tree shrew. Vision Research, 24 (9), 1037–1042, https://doi.org/10.1016/0042-6989(84)90080-4.
R Core Team. (2017). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing, https://www.R-project.org/.
Reymond, L., & Wolfe, J. (1981). Behavioural determination of the contrast sensitivity function of the eagle Aquila audax. Vision Research, 21 (2), 263–271, https://doi.org/10.1016/0042-6989(81)90120-6.
Srinivasan, M. V., & Lehrer, M. (1988). Spatial acuity of honeybee vision and its spectral properties. Journal of Comparative Physiology A, 162 (2), 159–172, https://doi.org/10.1007/bf00606081.
Uhlrich, D. J., Essock, E. A., & Lehmkuhle, S. (1981). Cross-species correspondence of spatial contrast sensitivity functions. Behavioural Brain Research, 2 (3), 291–299, https://doi.org/10.1016/0166-4328(81)90013-9.
Wood, S. N. (2006). Generalized additive models: An introduction with R. Boca Raton, FL: Chapman & Hall/CRC Press.
Yamanouchi, T. (1956). The visual acuity of the coral fish Microcanthus strigatus (Cuvier & Valenciennes). Publication Seto Marine Biology Laboratory, 5, 133–156.
Zander, C. D., & Hagemann, T. (1989). Feeding ecology of littoral gobiid and blennioid fishes of the Banyuls area (Mediterranean Sea). III. Seasonal variations. Scientia Marina, 53 (2), 441–449.
Zuur, A., Ieno, E. N., Walker, N., Saveliev, A. A., & Smith, G. M. (2009). Mixed effects models and extensions in ecology with R. New York, NY: Springer.
Supplementary material
Supplementary Movie S1. The optokinetic reflex in Tripterygion delaisi can be seen as tracking behavior (at least one eye or the body following the rotation direction of the pattern), often mixed with opposing saccades (sudden eye movement in the opposite direction of the pattern). The grating displayed in this video is rotating left, has a spatial frequency of 0.125 c/° and a Michelson contrast of 25%. 
Appendix
Table A1
 
Experimental data. contrast_sensitivity, indicates the weakest contrast (expressed as the reciprocal of the Michelson contrast) still eliciting a response in a triplefin (ID_triplefin) for a specific stimulus spatial resolution (stimulus_resolution) and angular speed (stimulus_speed); stimulus_rotation, indicates the rotation of the striped pattern used for each session shown in this table.
Table A1
 
Experimental data. contrast_sensitivity, indicates the weakest contrast (expressed as the reciprocal of the Michelson contrast) still eliciting a response in a triplefin (ID_triplefin) for a specific stimulus spatial resolution (stimulus_resolution) and angular speed (stimulus_speed); stimulus_rotation, indicates the rotation of the striped pattern used for each session shown in this table.
Figure 1
 
Virtual arena inside the optokinetic drum. Individual fish were placed in a central transparent glass cylinder positioned on an elevated circular platform. The four screens surrounding the fish displayed the grating pattern. Mirrors on the bottom and top of the setup assured the grating to be visible in a radial pattern (not visible from the camera perspective shown here). Photo credit: Matteo Santon.
Figure 1
 
Virtual arena inside the optokinetic drum. Individual fish were placed in a central transparent glass cylinder positioned on an elevated circular platform. The four screens surrounding the fish displayed the grating pattern. Mirrors on the bottom and top of the setup assured the grating to be visible in a radial pattern (not visible from the camera perspective shown here). Photo credit: Matteo Santon.
Figure 2
 
Top view of the optokinetic drum virtual arena. A fish in the center of a glass cylinder experiences the moving grating pattern on the screens as a rotating cylinder. Photo credit: Matteo Santon.
Figure 2
 
Top view of the optokinetic drum virtual arena. A fish in the center of a glass cylinder experiences the moving grating pattern on the screens as a rotating cylinder. Photo credit: Matteo Santon.
Figure 3
 
Contrast sensitivity function (CSF) of Tripterygion delaisi. Contrast sensitivity, expressed as the reciprocal of the Michelson contrast, as a function of spatial frequency. The optimal sensitivity was around 0.375 c/°, where seven out of 10 fish showed the maximum sensitivity of 125, which corresponds to a 0.8% Michelson contrast. Contrast sensitivity is plotted on a log-scale. Numbers at the top indicate the number of individuals (N = 10) responding to each spatial frequency. None of the individuals responded to the spatial frequency of 2.375 (not shown). Each color of the points represents a different individual (jittered for clarity). Error bars display the model-predicted group means ± 95% credible intervals. The black curve shows the relationship between contrast sensitivity and spatial frequency on a continuous scale (see Materials and methods section for details).
Figure 3
 
Contrast sensitivity function (CSF) of Tripterygion delaisi. Contrast sensitivity, expressed as the reciprocal of the Michelson contrast, as a function of spatial frequency. The optimal sensitivity was around 0.375 c/°, where seven out of 10 fish showed the maximum sensitivity of 125, which corresponds to a 0.8% Michelson contrast. Contrast sensitivity is plotted on a log-scale. Numbers at the top indicate the number of individuals (N = 10) responding to each spatial frequency. None of the individuals responded to the spatial frequency of 2.375 (not shown). Each color of the points represents a different individual (jittered for clarity). Error bars display the model-predicted group means ± 95% credible intervals. The black curve shows the relationship between contrast sensitivity and spatial frequency on a continuous scale (see Materials and methods section for details).
Figure 4
 
A Tripterygion delaisi's view of a conspecific and of a predator. Foveal acuity (first column) allows resolution of conspecifics (T. delaisi) at a greater distance than peripheral acuity (second column), while both acuities allow to resolve the outline of a potential predator (Serranus scriba) up to 80 cm (third and fourth column). Each image shows the spatial information content of the scene, adjusted for spatial resolution and distance. We also assume that the total length of T. delaisi in the picture is 5 cm, and the length of S. scriba is 20 cm. All manipulated images have been generated using the R package AcuityView (Caves & Johnsen, 2018). Photo credits: Matteo Santon.
Figure 4
 
A Tripterygion delaisi's view of a conspecific and of a predator. Foveal acuity (first column) allows resolution of conspecifics (T. delaisi) at a greater distance than peripheral acuity (second column), while both acuities allow to resolve the outline of a potential predator (Serranus scriba) up to 80 cm (third and fourth column). Each image shows the spatial information content of the scene, adjusted for spatial resolution and distance. We also assume that the total length of T. delaisi in the picture is 5 cm, and the length of S. scriba is 20 cm. All manipulated images have been generated using the R package AcuityView (Caves & Johnsen, 2018). Photo credits: Matteo Santon.
Table A1
 
Experimental data. contrast_sensitivity, indicates the weakest contrast (expressed as the reciprocal of the Michelson contrast) still eliciting a response in a triplefin (ID_triplefin) for a specific stimulus spatial resolution (stimulus_resolution) and angular speed (stimulus_speed); stimulus_rotation, indicates the rotation of the striped pattern used for each session shown in this table.
Table A1
 
Experimental data. contrast_sensitivity, indicates the weakest contrast (expressed as the reciprocal of the Michelson contrast) still eliciting a response in a triplefin (ID_triplefin) for a specific stimulus spatial resolution (stimulus_resolution) and angular speed (stimulus_speed); stimulus_rotation, indicates the rotation of the striped pattern used for each session shown in this table.
Supplement 1
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×