Optical illusions in different ungulates species

doi:10.21203/rs.3.rs-3763022/v1

Download PDF

Research Article

Optical illusions in different ungulates species

https://doi.org/10.21203/rs.3.rs-3763022/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Optical illusions have been used in behavioural studies in recent years to investigate the perceptual mechanisms underlying vision in animals. Although susceptibility to some illusions has been extensively studied in some species, only two studies have yet been carried out in ungulates, suggesting that they may perceive optical illusions in a way similar to humans. Here, we used two food-choice tasks to study susceptibility to the Müller-Lyer and Delboeuf illusions in 17 captive individuals belonging to four ungulate species (Lama guanicoe, Lama glama, Ovis aries, Capra hircus). In control trials, individuals reliably preferred the longer/larger food over the shorter/smaller one. In experimental trials, individuals significantly preferred the food stick between two inward arrowheads over an identical one between two outward arrowheads in the Müller-Lyer task, and they preferred the food on the smaller circle over an identical one on the larger circle in the Delboeuf task, with no significant differences across species. Overall, these findings suggest that, in line with our predictions, ungulates are susceptible to the Müller-Lyer and the Delboeuf illusions in a similar way to humans, and indicate that the perceptual mechanisms underlying size estimation in arctiodactyla might be similar to those of other ungulate and primate species, including humans.

Susceptibility to optical illusions can result in individuals failing to accurately process visual information and misinterpreting reality in the presence of specific environmental cues (Gregory, 1997, 1998). Size illusions, in particular, occur when individuals misperceive the size of an object due to the surrounding background (Shapiro & Todorovic, 2017). One of the best-known size illusions is the Müller-Lyer (Figure 1a), which occurs when individuals perceive the same line as being longer when in proximity of two arrowheads pointing inward, as compared to outward (Gregory, 1963; Shapiro & Todorovic, 2017). This illusion is traditionally thought to occur because individuals wrongly extrapolate three-dimensional information from two-dimensional images, perceiving the line between inward-pointing arrowheads as being farther, and thus longer, than the one between outward-pointing arrowheads (Gregory, 1963, 1966), although the exact mechanisms explaining this illusion are yet discussed (Howe & Purves, 2005). Another classical size illusion is the Delboeuf illusion (Figure 1b), which occurs when individuals perceive the same circle as being larger if surrounded by a smaller than by a larger concentric circle (Coren & Girgus, 2022). This illusion is thought to occur because the smaller surrounding circle assimilates the internal circle, making it look larger, whereas the larger surrounding circle contrasts it, making it look smaller than in reality (King, 1988).

Several studies have shown that humans are widely susceptible to both the Müller-Lyer and the Delboeuf illusions (Shapiro & Todorovic, 2017), although there might be important intra-specific variation linked to cultural, genetic and environmental factors (e.g., exposure to urban environments: Deręgowski, 2017). Despite some early studies on other species, it is only more recently that researchers have started systematically investigating how species other than humans perceive size illusions (see Fujita, Nakamura, & Watanabe, 2017, for a review). As the perception of illusions occurs automatically in humans, and as humans partly share their visual system with other species, it is indeed possible that susceptibility to optical illusions is widespread across animals (Feng, Chouinard, Howell, & Bennett, 2017). Studying optical illusions in a comparative perspective, indeed, may be informative for several reasons. Comparing susceptibility to optical illusions across species, for instance, is a non-invasive way to acquire information on the similarity of their visual systems, and to infer whether these systems share a long evolutionary history in common or rather emerged multiple times as the result of convergent evolution (Fujita et al., 2017). If the same susceptibility to illusion is shared by different species, for instance, it is possible to infer that these species evolved similar mechanisms to visually perceive the world (Feng et al., 2017). In the same line, comparing susceptibility to optical illusions across species is also informative to understand the ecological and environmental conditions in which such susceptibility might emerge, and infer the adaptive significance of illusory perceptions (Fujita et al., 2017).

In species other than humans, studies have shown important variation within and across species in their susceptibility to optical illusions, with some contradictory results. Therefore, it is not yet clear to what extent different species share similar perceptual systems, and/or whether such variation rather depends on specific contextual factors (e.g., methodology, demographic characteristics of the study subjects). A recent review, for example, reports the existence of studies on optical illusions (including the Müller-Lyer, Delboeuf, Ebbinghaus, Kanisza, Zollner and Rotating snake illusions) in ten species of mammals, eight of fish, four of birds and two of reptiles (Santacà, Agrillo, & Miletto Petrazzini, 2021). For the Müller-Lyer illusion, in particular, there is evidence that several species perceive it in a way similar to humans, including rhesus macaques, Macaca mulatta (Tudusciuc & Nieder, 2010), capuchin monkeys, Cebus capucinus (Suganuma, Pessoa, Monge-Fuentes, Castro, & Tavares, 2007), horses, Equus caballus (Cappellato, Miletto Petrazzini, Bisazza, Dadda, & Agrillo, 2020; Murayama, 2012) , pigeons, Columba livia (Nakamura, Fujita, Ushitani, & Miyata, 2006) and teleost fish (Sovrano, da Pos, & Albertazzi, 2016). For the Delboeuf illusion, results are more controversial. Whereas chimpanzees, Pan troglodytes (Parrish & Beran, 2014), rhesus monkeys, Macaca mulatta, capuchin monkeys, Cebus capucinus (Parrish, Brosnan, & Beran, 2015), cats, Felis catus (Szenczi, Velázquez-López, Urrutia, Hudson, & Bánszegi, 2019) and bearded dragons, Pogona vitticeps (Santacà, Miletto Petrazzini, Agrillo, & Wilkinson, 2019) appear to perceive it in the same way as humans, other species perceive it in the opposite direction (guppies, Poecilia reticulata: (Lucon-Xiccato, Santacà, Miletto Petrazzini, Agrillo, & Dadda, 2019); other teleost fish: (Santacà, Lucon-Xiccato, & Agrillo, 2020); partially dogs, Canis lupus familiarias: Byosiere, Feng, Woodhead, et al., 2017), or do not perceive it (ring-tailed lemurs, Lemur catta: (Santaca, Regaiolli, Miletto Petrazzini, Spiezio, & Agrillo, 2017).

Clearly, some taxa appear to be especially underrepresented in research on optical illusions. To the best of our knowledge, for example, only two studies have so far been conducted in ungulates (Cappellato et al., 2020; Murayama, 2012). In one study, bottlenose dolphins (Tursiops truncatus) trained to select the larger of two circles later preferred the circle that was surrounded by six smaller rather than larger inducer circles, suggesting susceptibility to the Ebbinghaus illusion (Murayama, 2012). In the other study, horses (Equus caballus) spontaneously preferring a longer over a shorter carrot stick also preferentially selected a carrot stick between two inward-pointing arrowheads over an identical one between two outward-pointing arrowheads, suggesting susceptibility to the Müller-Lyer illusion (Cappellato et al., 2020). Therefore, ungulates appear to be a promising but yet understudied taxon to study optical illusions in a comparative perspective.

The main aim of this study was to investigate how different ungulate species (i.e., guanacos, Lama guanicoe, llamas, Lama glama, Skudde sheep, Ovis aries, and Damara goats, Capra hircus) perceive two optical size illusions: the Müller-Lyer and the Delboeuf illusions. In ungulates, eyes are positioned on the side of the head, providing them with a wide field of view to detect predators (Sugnaseelan, Prescott, Broom, Wathes, & Phillips, 2013) but likely reducing their ability to perceive depth and distance, as the overlap between the visual fields of both eyes is limited (Fowler, 2011). Although visual acuity may vary across species (Carroll, Murphy, Neitz, Hoeve, & Neitz, 2001), vision is considered the dominant sense in ungulates (Fletcher & Lindsay, 1968; Lindsay & Fletcher, 1968), playing a crucial role in environmental perception (Baldwin, 1979, 1981), individual recognition (Davis, Norris, & Taylor, 1998; Lickliter & Heron, 1984; Taylor & Davis, 1998) and selection of ecological resources (Arnold, 1966; Bazely & Ensor, 1989). In ungulates, vision is indeed well-suited for detecting movement, identifying objects (Caro, 1994; Hirata, Arimoto, Hattori, & Anzai, 2019) and distinguishing shapes and patterns (Baldwin, 1981; Blakeman & Friend, 1986; Roitberg & Franz, 2004; Schaeffer & Sikes, 1971).

Although different ecological characteristics might be linked to the emergence of differences in visual systems (Wasserman, Lazareva, & Shimizu, 2012), in this study we hypothesized that, if their visual systems share a long common evolutionary history, ungulate species should be susceptible to optical illusions, as already shown in dolphins and horses (Cappellato et al., 2020; Murayama, 2012). In particular, we predicted that guanacos, llamas, sheep and goats would be similarly susceptible to both the Müller-Lyer and the Delboeuf illusions, preferring the food stick between two inward arrowheads over an identical one between two outward arrowheads in the Müller-Lyer task, and preferring the food surrounded by a smaller circle over an identical one surrounded by a larger circle in the Delboeuf task.

Study subjects. We tested 17 subjects belonging to 4 ungulate species (Table 1), including 5 guanacos (Lama guanicoe), 3 llamas (Lama glama), 5 Skudde sheep (Ovis aries) and 4 Damara goats (Capra hircus). All subjects were housed with conspecifics at the zoo of Leipzig, in Germany, and were individually recognizable thanks to variation in their morphological features (e.g., height, size, fur colour). Study subjects included both males and females, and were all adults (i.e., older than one year), except for one sheep younger than one, who was only tested in the first task. The daily diet of all species included hay, which was available ad libitum, and fresh vegetables. None of the study subjects had ever been tested in an optical illusion task before, although all species had occasionally participated in enrichment activities or in other non-invasive experimental tasks (Caicoya et al., 2023; Schaffer et al., 2020; Schaffer et al., 2021).

Materials and procedures. We administered two tasks: one for the Müller-Lyer illusion and one for the Delboeuf illusion. To facilitate comparisons with previous studies, we followed the procedures used in literature with other non-human species (i.e., Müller-Lyer task: (Cappellato et al., 2020) Delboeuf task: (Bánszegi, Szenczi, Urrutia, Martínez-Byer, & Hudson, 2021) . The tasks were carried out in the outdoor facilities of each species, between 7.30 A.M. and 10.30 A.M., without changing the daily routine of the study subjects. To avoid providing inadvertent cues, the experimenter (i.e., the first author) always wore sunglasses during the tasks. The general procedure for both tasks consisted in the experimenter presenting one cardboard to the study subject, either by attaching it to the fence, inside the enclosure, thanks to a wooden frame (for the guanacos, who were tested in a facility surrounded by 2-meter-high mesh), or by supporting it with both hands (for the other species, whose facilities were surrounded by a lower fence). To avoid separating individuals during the tasks, we waited for the study subject to be alone in proximity of the experimenter and threw a small piece of food at approximately one meter from the cardboard, so that after retrieving the food the subject faced the cardboard frontally. The subject could then approach the cardboard and touch one of the stimuli attached to the cardboard (see below). As soon as the subject chose one stimulus by touching it with the muzzle or lips, the experimenter allowed the subject to eat the chosen stimulus, while moving the other out of reach. As stimuli, we used familiar food that subjects highly liked and usually received in small quantities. For the Müller-Lyer illusion, we used carrots for all species. For the Delboeuf illusion, which required stimuli with a larger diameter (see below), we used celery for guanacos and llamas (which we painted orange with sweet paprika, to make it visually more salient), and carrots for sheep and goats (which we cut into semicircles that we merged to create round stimuli), as sheep and goats did not like celery. All trials were video-recorded and later coded from the videos.

In the Müller-Lyer task, all the illusions were presented on black cardboards whose measure was adjusted to the size of the study species: 90x60 cm for guanacos and llamas, and 60x40 cm for sheep and goats. On the right and left halves of the cardboard, we presented two vertical carrot sticks by attaching them by means of toothpicks (Figure 1a). Since the study species have a blind central area (Sugnaseelan et al., 2013), the distance between the carrot sticks was 60 cm for guanacos and llamas (as for horses: (Cappellato et al., 2020) , and 40 cm for sheep and goats. Carrot sticks were all equally large and thick (i.e., 1 cm), but their length could vary, being either 20 or 13.5 cm for guanacos and llamas, and 13.5 or 9 cm long for sheep and goats (so that the same length ratio was maintained between longer and shorter carrot sticks for all species). Depending on the condition, white wooden sticks (7x1x1 cm for guanacos and llamas, 5x1x1 cm for sheep and goats) were arranged around the carrot pieces to form arrowheads on the cardboard (Figure 1a). All species were tested in six different conditions (Figure 1a). In the Experimental long condition, we placed two identical longer carrot sticks (i.e., 20 or 13.5 cm, depending on the species) on the cardboard: one with the two arrowheads pointing inward, one outward. The Experimental short condition was identical, except that we used two identical shorter carrot sticks (i.e., 13.5 or 9 cm, depending on the species). The experimental conditions allowed us to test whether subjects perceived the Müller-Lyer illusion, perceiving the stick between the two inward arrowheads as being longer than the other one, and thus preferentially selecting it (Gregory, 1997) . The four control conditions allowed us to control that subjects reliably selected the longer over the shorter carrot stick when arrowheads were positioned in different ways. In Control condition A, arrowheads pointed inward for both carrots; in Control condition B, they pointed outward; in Control condition C, they pointed upward, and in Control condition D, they pointed downward.

In the Delboeuf task, all the illusions were presented on grey cardboards measuring 45x17 cm. On the right and left halves of the panel, we painted two white circles that could have a diameter of either 12.5 or 9.5 cm. The distance between the centres of the plates was 30 cm. On the plates, we presented 0.3-cm-thick food circles that could have a diameter of either 4.5 or 3 cm and were attached by means of toothpicks (Figure 1b). All species were tested in three different conditions (Figure 1b). In the Experimental condition, we placed two identical larger food items (i.e., 4.5 cm) on a cardboard with a smaller (i.e. 9.5 cm) and a larger (i.e. 12.5 cm) circle. This condition allowed us to test whether subjects perceived the Delboeuf illusion, perceiving the food on the smaller circle as being larger than the other one, and thus preferentially selecting it (Coren & Girgus, 2022) . The two control conditions allowed us to control that subjects reliably distinguished food items of different sizes and consistently selected larger over smaller ones. In Control condition A, we placed a smaller (i.e., 3 cm) and a larger (i.e., 4.5 cm) food item on a cardboard with two identical larger circles (i.e., 12.5 cm); Control condition B was identical, but the smaller and larger food items were placed on a cardboard with two identical smaller circles (i.e., 9.5 cm).

Before starting testing, all study subjects went through a habituation and a pre-testing phase, to familiarize them with the general procedure and ensure that subjects spontaneously maximized food intake by selecting the larger of two quantities, respectively. In the habituation phase, subjects were presented with a 6-cm-long carrot piece attached to the black cardboard for 6 trials a day, over 2 days, following the general procedure described above. All subjects participated in the habituation and retrieved the food in all the 12 trials. In the pre-testing phase, subjects were presented with a longer (i.e., 20 or 13.5 cm, depending on the species) and a shorter carrot (i.e., 13.5 or 9 cm, depending on the species) on a black cardboard without arrowheads, and were tested until reaching criterion (i.e., selecting the longer carrot in at least 10 out of 12 consecutive trials). Subjects required on average 16±5 trials to reach criterion (all goats: 12 trials; guanacos: 19±6, range: 12-26 trials; llamas: 21±3, range: 18-25 trials; sheep: 14±3, range: 12-20 trials). After the habituation and pre-testing phases, we administered the Müller-Lyer task (i.e., 12 trials for each of the 6 conditions), presenting up to 6 trials per subject a day (i.e., one for each condition). Then, we presented the Delboeuf task (i.e., 12 trials for each of the 3 conditions), presenting up to 6 trials per subject a day (i.e. two for each condition). In both tasks, we pseudo-randomized and counterbalanced across trials the condition we administered, and the size of the food that was larger or could be perceived as being larger (never presenting it on the same side for more than twice in a row). As ungulates often show a side bias (Fourie, Berezina, Giljov, & Karenina, 2021; Leliveld, 2019), in case of two consecutive wrong choices on the same side, we administered two additional trials in which only one food item was presented on the cardboard without arrowheads, on the opposite side. As performance in the Experimental conditions could not rely on olfactory cues (as both stimuli had identical size), we included no conditions to test subjects’ use of olfactory cues.

Data coding and analyses. For each trial, we coded subject identity, condition, number of trial for each subject (i.e., 1 to 72 for the Müller-Lyer task, 1 to 36 for the Delboeuf task), side chosen (i.e., left or right) and whether the subject chose the side that was longer/larger (in the control conditions) or could appear longer/larger if perceiving the illusion (in the experimental conditions). A second observer naïve to the experimental hypothesis re-coded subjects’ choice in 190/1800 trials, from the videos. Inter-observer reliability was excellent (Cohen’s k: k=0.97, N=190, p<0.001).

We ran two generalized linear mixed models (Baayen, Davidson, & Bates, 2008) in R, using the package glmmTMB (Berry et al., 2017). To this end, we built two datasets, one for each task, entering one line for each subject and trial (N=1224 for the Müller-Lyer illusion, N=576 for the Delboeuf illusion). Our binomial response was whether the focal subject choose the side that was/looked longer (Müller-Lyer task, Model 1) or larger (Delboeuf task, Model 2). In both full models, we entered as test predictor the interaction of condition and species, and the main terms of the interaction, and as controls the number of trial and the side chosen, adding individual identity as random factor. These models thus allowed us to assess whether performance varied across conditions, in a different way across species. Full models were then compared with likelihood ratio tests to null models that were identical, but did not include test predictors (Dobson & Barnett, 2018). In case of a significant difference between full and null model, we used the drop1 function to assess which variables were significant. We checked model assumptions, including residual diagnostics and overdispersion, with the “DHARMa” package, and multicollinearity with the “performance” package (maximum variance inflation factors for both models = 1.01), and detected no issues in the models presented.

As our models evidenced no variation across conditions (see Results), we further used Wilcoxon signed-rank tests to assess whether performance in the experimental conditions differed from chance level (0.50). To this end, we built two datasets, one for each task, entering one line for each subject (N=17 for the Müller-Lyer illusion, N=16 for the Delboeuf illusion), with the mean proportion of trials in which each subject chose the side that looked longer/larger. All species were grouped together, as species had no significant effect in the models (see Results). Finally we ran a binomial test for each individual, to assess whether individual performance in the experimental conditions differed from chance.

Müller-Lyer illusion task. In the Müller-Lyer task, on average, subjects chose the carrot that was longer in 77% of the control trials, and the carrot that looked longer in 78% of the experimental trials. The full model did not significantly differ from the null model (GLMM, χ²=27.99, df=23, p=0.216), suggesting no significant effect of species and condition, neither in interaction nor as main terms, on the probability of choosing the side that was/looked longer (Table 2). As this suggests that performance was similar across conditions, and species, we run a Wilcoxon test to assess whether performance in the Experimental and Control conditions differed from chance, regardless of species. Wilcoxon tests showed that subjects performed above chance level in both Experimental conditions (both p<0.001) and in all four Control conditions (all p<0.001), preferring the side that looked/was longer in all conditions (Figure 2). Binomial tests further showed that in the Experimental conditions 11 out of 17 study subjects chose the side that looked longer significantly above chance level, whereas the other 6 subjects preferred the side that looked longer but failed to reach significance (Table 1).

Delboeuf illusion task. In the Delboeuf task, on average, subjects chose the food that was larger in 77% of the control trials, and the food that looked larger in 74% of the experimental trials. The full model did not significantly differ from the null model (GLMM, χ²=6.60, df=11, p=0.830), suggesting no significant effect of species and condition on the probability of choosing the side that was/looked larger (Table 2). As above, we thus run a Wilcoxon test to assess whether performance in the Experimental and Control conditions differed from chance. Wilcoxon tests showed that subjects performed above chance level in the Experimental condition (p<0.001) and in both Control conditions (both p<0.001), preferring the side that looked/was larger in all conditions (Figure 3). Binomial tests further showed that in the Experimental condition 4 out of 16 study subjects chose the side that looked larger significantly above chance level, 10 subjects preferred the side that looked larger but failed to reach significance, and 2 subject chose at chance levels (Table 1).

In our study, we tested the susceptibility of subjects belonging to 4 different ungulate species (i.e., guanacos, llamas, Skudde sheep and Damara goats) to the Müller-Lyer and Delboeuf illusions. Following the procedures previously used with other species (Bánszegi et al., 2021; Cappellato et al., 2020) we found that the study subjects perceived both the Müller-Lyer and the Delboeuf illusions in a similar way to humans, with no significant differences across species.

In the Müller-Lyer task, individuals reliably selected the longer over the shorter stimulus in the control conditions, spontaneously choosing the option that allowed them to maximize food intake. Similarly, they reliably selected the food piece between two inward arrowheads over an identical food piece between two outward arrowheads, suggesting that they perceived the former as being longer than the latter (Gregory, 1997) . Subjects selected the food that was/looked longer above chance level in all conditions, with no variation across species. These results are in line with literature on other species, including humans (see (Shapiro & Todorovic, 2017) , rhesus macaques (Tudusciuc & Nieder, 2010) capuchin monkeys (Suganuma et al., 2007), pigeons (Nakamura et al., 2006), teleost fish (Sovrano et al., 2016), and another ungulate species, horses (Cappellato et al., 2020; Murayama, 2012). Overall, these findings suggest that ungulates, like primates, may overall be susceptible to the Müller-Lyer illusion, although other taxon may not perceive it (e.g., dogs: (Keep, Zulch, & Wilkinson, 2018). Results were confirmed at the individual level, with all subjects preferentially selecting the food between two inward arrowheads in the Experimental conditions, and 11 out of 17 doing it significantly above chance. On a side note, susceptibility to the Müller-Lyer illusion decreased across trials (Table 1), probably because of a decrease in subjects’ motivation due to the high number of trials and the gradual emergence of side biases, which is typical of ungulates (Fourie et al., 2021).

In the Delboeuf task, we obtained very similar results. Individuals reliably selected the larger over the smaller food in the control conditions, and similarly preferred the food on the smaller circle over an identical one on the larger circle, suggesting that they perceived the former as being larger than the latter (Coren & Girgus, 2022) . Subjects selected the food that was/looked larger above chance in all conditions, with no variation across species. These findings suggest that ungulates are susceptible to the Delboeuf illusion, and are in line with previous findings in other species, like humans (see (Shapiro & Todorovic, 2017), chimpanzees (Parrish & Beran, 2014), rhesus monkeys, capuchin monkeys (Parrish et al., 2015), cats (Szenczi et al., 2019) and bearded dragons (Santacà et al., 2019) . Other species, in contrast, either appear to perceive this illusion in the opposite direction (guppies: (Lucon-Xiccato et al., 2019) ; other teleost fish: (Santacà et al., 2020) ; partially dogs: (Byosiere, Feng, Rutter, et al., 2017), or do not perceive it at all (ring-tailed lemurs: Santaca et al., 2017). In our study, at the individual level, only 4 out of 16 study subjects chose the food on the smaller circle significantly above chance, although no subject preferentially chose the food on the larger circle. As compared to the Müller-Lyer task, the lower number of individuals significantly preferring the food that looked larger simply depends on the lower power that we had in the second task, where we administered 12 Experimental trials per subject, instead of 24. For instance, although Frangia chose the side that looked longer/larger in 75% of the Experimental trials in both tasks, this resulted in a significant binomial test only in the Müller-Lyer task (Table 1). Therefore, overall, the results of both tasks converge in suggesting a strong susceptibility to the Müller-Lyer and Delboeuf illusions both at the group and at the individual level.

In line with our predictions, this study provides clear evidence that guanacos, llamas, sheep and goats were susceptible to two optical size illusions, in line with results from other ungulate species (Cappellato et al., 2020; Murayama, 2012). Moreover, our findings show that at least some ungulate species perceive optical illusions similarly to humans and other primates (Costa et al., 2021; Parrish & Beran, 2014). This suggests that their visual systems might share a long evolutionary history in common, as the mechanisms they have evolved for visual perception are similarly deceived by the presence of specific visual cues (e.g., arrowheads, circles; Feng et al., 2017; Fujita et al., 2017). However, only the inclusion of more species will allow to definitely exclude that these similarities are not the results of convergent evolution (see Fujita et al., 2017, for a discussion).

Our study has several limitations. First, we could only test a limited number of subjects for each species, which might have prevented us from detecting intra-specific variation in how individuals perceive illusions. Although we did not specifically expect any effect of sex on susceptibility to optical illusions, for instance, our study suffered from a strongly biased sex distribution across study species (Table 1). Second, our study only included captive individuals, who may not be good representatives of their wild counterparts. Socio-ecological constraints experienced during ontogeny and extensive exposure to objects and other human artifacts may affect the development of captive individuals (in primates, see (Boesch, 2007), and perhaps also their susceptibility to optical illusions. In cross-cultural studies on optical illusions, for instance, some researchers have suggested that the environment experienced might affect human ability to perceive optical illusions, with subjects living in a more “carpentered” world (i.e., with frequent right angles and rectangular objects) being more likely to interpret angles in illusions as projections of right angles, and thus more likely to perceive some optical illusions (Segall et al., 1966). Comparisons with wild subjects will thus be interesting to test the possible effect of environmental conditions on ungulate susceptibility to optical illusions. Similarly, future studies would benefit from the inclusion of a developmental approach, as susceptibility to illusions might change through age also in species other than humans (see e.g., Bánszegi et al., 2021, in cats).

Overall, our study provides strong evidence that different ungulate species perceive two classical optical illusions, the Müller-Lyer and Delboeuf illusions, in a way similar to humans. Over the course of millions of years, ungulates have played an essential role in human life, in agricultural settings (Pascual-Rico et al., 2021; Reimoser & Putman, 2011), for recreational purposes (Yeates & McGreevy, 2019), as companions in equine-assisted therapy (White-Lewis, 2020), and as food and economic source (Banda & Tanganyika, 2021). Understanding their perceptual systems and skills will hopefully contribute to improve their welfare and management in captive and wild settings (Held, Mendl, Laughlin, & Byrne, 2002), and uphold higher ethical standards when managing these species for human purposes.

Ethical approval. The study was approved by the welfare manager at the zoo of Leipzig, who controlled and approved all the procedures. During the task, individuals were never water or food deprived, and motivation to participate was ensured exclusively by the use of highly preferred food belonging to their natural diets. The experiments thus provided a form of enrichment for the subjects and did not present any risks or adverse effect. The study was carried out in accordance with the national regulations.

Competing interests. I declare that the authors have no competing interests to declare.

Authors' contributions. FA designed the research questions and the methods, together with CB. CB conducted the experiments, with extensive support from RH, ALC and especially AS. KL and PV helped FA with the supervision of CB. CB and FA conducted the statistical analyses and wrote the manuscript, with extensive feedback from all the other coauthors.

Funding. No funding was received for this project.

Availability of data and materials. Data will be provided upon reasonable request to the last author.

Acknowledgments. We are grateful to the keepers at the Leipzig Zoo, and especially to Janet, for endless support. We are also grateful to Nivola Coppola and Lotta Bräuer for coding for inter-observer reliability.

Arnold, G. W. (1966). The special senses in grazing animals. I. Sight and dietary habits in sheep. Australian Journal of Agricultural Research, 17(4), 521. https://doi.org/10.1071/AR9660521
Baayen, R. H., Davidson, D. J., & Bates, D. M. (2008). Mixed-effects modeling with crossed random effects for subjects and items. Journal of Memory and Language, 59(4), 390–412. https://doi.org/10.1016/j.jml.2007.12.005
Baldwin, B. A. (1979). Operant studies on shape discrimination in goats. Physiology & Behavior, 23(3), 455–459. https://doi.org/10.1016/0031-9384(79)90043-X
Baldwin, B. A. (1981). Shape discrimination in sheep and calves. Animal Behaviour, 29(3), 830–834. https://doi.org/10.1016/S0003-3472(81)80017-6
Banda, L. J., & Tanganyika, J. (2021). Livestock provide more than food in smallholder production systems of developing countries. Animal Frontiers: The Review Magazine of Animal Agriculture, 11(2), 7–14. https://doi.org/10.1093/af/vfab001
Bánszegi, O. [O.], Szenczi, P. [P.], Urrutia, A. [A.], Martínez-Byer, S., & Hudson, R. [R.] (2021). Visual discrimination of size and perception of the Delboeuf illusion in the domestic cat (Felis silvestris catus): A developmental disjunction? Journal of Comparative Psychology (Washington, D.C. : 1983), 135(4), 505–515. https://doi.org/10.1037/com0000288
Bazely, D. R., & Ensor, C. V. (1989). Discrimination learning in sheep with cues varying in brightness and hue. Applied Animal Behaviour Science, 23(4), 293–299. https://doi.org/10.1016/0168-1591(89)90098-1
Berry, R. B., Brooks, R., Gamaldo, C., Harding, S. M., Lloyd, R. M., Quan, S. F., . . . Vaughn, B. V. (2017). Aasm Scoring Manual Updates for 2017 (Version 2.4). Journal of Clinical Sleep Medicine : JCSM : Official Publication of the American Academy of Sleep Medicine, 13(5), 665–666. https://doi.org/10.5664/jcsm.6576
Blakeman, N. E., & Friend, T. H. (1986). Visual discrimination at varying distances in spanish goats. Applied Animal Behaviour Science, 16(3), 279–283. https://doi.org/10.1016/0168-1591(86)90120-6
Boesch, C. (2007). What makes us human (Homo sapiens)? The challenge of cognitive cross-species comparison. Journal of Comparative Psychology (Washington, D.C. : 1983), 121(3), 227–240. https://doi.org/10.1037/0735-7036.121.3.227
Byosiere, S.‑, Feng,L. C., Rutter, N. J. R., Woodhead, J. K., Chouinard, P. A., & Howell, T. J. (2017). Do Dogs See the Ponzo illusion? Animal Behavior and Cognition, 4(4), 396–412. https://doi.org/10.26451/abc/.04.04.01.2017
Byosiere, S.‑, Feng,L. C., Woodhead, J. K., Rutter, N. J., Chouinard, P. A., Howell, T. J., & Bennett, P. C. (2017). Visual perception in domestic dogs: Susceptibility to the Ebbinghaus-Titchener and Delboeuf illusions. Animal Cognition, 20(3), 435–448. https://doi.org/10.1007/s10071-016-1067-1
Caicoya, A. L., Schaffer, A., Holland, R., Fersen, L. von, Colell, M., & Amici, F. (2023). Innovation across 13 ungulate species: Problem solvers are less integrated in the social group and less neophobic. Biological Sciences, 290(1996), 20222384. https://doi.org/10.1098/rspb.2022.2384
Cappellato, A., Miletto Petrazzini, M. E., Bisazza, A., Dadda, M., & Agrillo, C. (2020). Susceptibility to Size Visual Illusions in a Non-Primate Mammal (Equus caballus). Animals : An Open Access Journal from MDPI, 10(9). https://doi.org/10.3390/ani10091673
Caro, T. M. (1994). Ungulate Antipredator Behaviour: Preliminary and Comparative Data From African Bovids. Behaviour, 128(3-4), 189–228. https://doi.org/10.1163/156853994X00262
Carroll, J., Murphy, C. J., Neitz, M., Hoeve, J. N., & Neitz, J. (2001). Photopigment basis for dichromatic color vision in the horse. Journal of Vision, 1(2), 80–87. https://doi.org/10.1167/1.2.2
Coren, S., & Girgus, J. (2022). Seeing is Deceiving: The Psychology of Visual Illusions (1st edition). London: Routledge.
Costa, A. L. L., Silva, R. C., Coelho-Cordeiro, P. H., da Silveira, F. S., Barros, M., Caixeta, F. V., & Maior, R. S. (2021). The Perception of the Müller-Lyer Visual Illusion in Schizophrenics and Non-human Primates: A Translational Approach. Frontiers in Behavioral Neuroscience, 15, 641776. https://doi.org/10.3389/fnbeh.2021.641776
Davis, H. [H.], Norris, C., & Taylor, A. (1998). Wether ewe know me or not: The discrimination of individual humans by sheep. Behavioural Processes, 43(1), 27–32. https://doi.org/10.1016/S0376-6357(97)00082-X
Deręgowski, J. B. (2017). Cross-Cultural Studies of Illusions. In A. G. Shapiro & D. Todorovic (Eds.), The Oxford Compendium of Visual Illusions (pp. 38–53). Oxford University PressNew York. https://doi.org/10.1093/acprof:oso/9780199794607.003.0002
Dobson, A., & Barnett, A. G. (2018). An Introduction to Generalized Linear Models, Fourth Edition. Chapman and Hall/CRC. https://doi.org/10.1201/9781315182780
Feng, L. C., Chouinard, P. A., Howell, T. J., & Bennett, P. C. (2017). Why do animals differ in their susceptibility to geometrical illusions? Psychonomic Bulletin & Review, 24(2), 262–276. https://doi.org/10.3758/s13423-016-1133-3
Fletcher, I. C., & Lindsay, D. R. (1968). Sensory involvement in the mating behaviour of domestic sheep. Animal Behaviour, 16(4), 410–414. https://doi.org/10.1016/0003-3472(68)90032-8
Fourie, B., Berezina, E., Giljov, A., & Karenina, K. (2021). Visual lateralization in artiodactyls: A brief summary of research and new evidence on saiga antelope. Laterality, 26(1-2), 106–129. https://doi.org/10.1080/1357650X.2020.1852245
Fowler, M. E. (2011). Medicine and Surgery of Camelids (3rd ed.). Somerset: Wiley.
Fujita, K., Nakamura, N., & Watanabe, S. (2017). Visual Illusion in a Comparative Perspective. In A. G. Shapiro (Ed.), The Oxford Compendium of Visual Illusions (pp. 54–63). Oxford: Oxford University Press. https://doi.org/10.1093/acprof:oso/9780199794607.003.0003
Gregory (1963). Distorsion of visual space as inappropriate constance scaling. Nature, 199(4894), 678–680. https://doi.org/10.1038/199678a0
Gregory (1966). Optical illusions. Nature, 209(5020), 328. https://doi.org/10.1038/209328a0
Gregory, R. L. (1997). Knowledge in perception and illusion. Philosophical Transactions of the Royal Society B: Biological Sciences, 352(1358), 1121–1127. https://doi.org/10.1098/rstb.1997.0095
Gregory, R. L. (1998). Eye and Brain: The Psychology of Seeing - Fifth Edition. Princeton, NJ: Princeton University Press.
Held, S., Mendl, M., Laughlin, K., & Byrne, R. W. (2002). Cognition studies with pigs: Livestock cognition and its implication for production. Journal of Animal Science, 80(E-suppl_1), E10-E17. https://doi.org/10.2527/animalsci2002.0021881200800ES10003x
Hirata, M., Arimoto, C., Hattori, N., & Anzai, H. (2019). Can cattle visually discriminate between green and dead forages at a short distance while moving in the field? Animal Cognition, 22(5), 707–718. https://doi.org/10.1007/s10071-019-01268-z
Howe, C. Q., & Purves, D. (2005). The Müller-Lyer illusion explained by the statistics of image-source relationships. Proceedings of the National Academy of Sciences of the United States of America, 102(4), 1234–1239. https://doi.org/10.1073/pnas.0409314102
Keep, B., Zulch, H. E., & Wilkinson, A. (2018). Truth is in the eye of the beholder: Perception of the Müller-Lyer illusion in dogs. Learning & Behavior, 46(4), 501–512. https://doi.org/10.3758/s13420-018-0344-z
King, D. L. (1988). Assimilation is due to one perceived whole and contrast is due to two perceived wholes. New Ideas in Psychology, 6(3), 277–288. https://doi.org/10.1016/0732-118X(88)90039-6
Leliveld, L. M. C. (2019). From Science to Practice: A Review of Laterality Research on Ungulate Livestock. Symmetry, 11(9), 1157. https://doi.org/10.3390/sym11091157
Lickliter, R. E., & Heron, J. R. (1984). Recognition of mother by newborn goats. Applied Animal Behaviour Science, 12(1-2), 187–192. https://doi.org/10.1016/0168-1591(84)90109-6
Lindsay, D. R., & Fletcher, I. C. (1968). Sensory involvement in the recognition of lambs by their dams. Animal Behaviour, 16(4), 415–417. https://doi.org/10.1016/0003-3472(68)90033-X
Lucon-Xiccato, T., Santacà, M., Miletto Petrazzini, M. E., Agrillo, C., & Dadda, M. (2019). Guppies, Poecilia reticulata, perceive a reversed Delboeuf illusion. Animal Cognition, 22(3), 291–303. https://doi.org/10.1007/s10071-019-01237-6
Murayama, T. (2012). Relative Size Discrimination and Perception of the Ebbinghaus Illusion in a Bottlenose Dolphin (Tursiops truncatus). Aquatic Mammals, 38(4), 333–342. https://doi.org/10.1578/AM.38.4.2012.333
Nakamura, N., Fujita, K., Ushitani, T., & Miyata, H. (2006). Perception of the standard and the reversed Müller-Lyer figures in pigeons (Columba livia) and humans (Homo sapiens). Journal of Comparative Psychology (Washington, D.C. : 1983), 120(3), 252–261. https://doi.org/10.1037/0735-7036.120.3.252
Parrish, A. E., & Beran, M. J. (2014). When less is more: Like humans, chimpanzees (Pan troglodytes) misperceive food amounts based on plate size. Animal Cognition, 17(2), 427–434. https://doi.org/10.1007/s10071-013-0674-3
Parrish, A. E., Brosnan, S. F., & Beran, M. J. (2015). Do you see what I see? A comparative investigation of the Delboeuf illusion in humans (Homo sapiens), rhesus monkeys (Macaca mulatta), and capuchin monkeys (Cebus apella). Journal of Experimental Psychology. Animal Learning and Cognition, 41(4), 395–405. https://doi.org/10.1037/xan0000078
Pascual-Rico, R., Morales-Reyes, Z., Aguilera-Alcalá, N., Olszańska, A., Sebastián-González, E., Naidoo, R., . . . Sánchez-Zapata, J. A. (2021). Usually hated, sometimes loved: A review of wild ungulates' contributions to people. The Science of the Total Environment, 801, 149652. https://doi.org/10.1016/j.scitotenv.2021.149652
Reimoser, F., & Putman, R. (2011). Impacts of wild ungulates on vegetation: costs and benefits. In Ungulate Management in Europe (1st ed., pp. 144–191). Cambridge University Press. https://doi.org/10.1017/cbo9780511974137.007
Roitberg, E., & Franz, H. (2004). Oddity learning by African dwarf goats (Capra hircus). Animal Cognition, 7(1), 61–67. https://doi.org/10.1007/s10071-003-0190-y
Santaca, M., Regaiolli, B., Miletto Petrazzini, M. E., Spiezio, C., & Agrillo, C. (2017). Preliminary study to investigate the Delboeuf illusion in ring-tailed lemurs (Lemur catta): Methodological Challenges. Animal Behavior and Cognition, 4(3), 365–377. https://doi.org/10.26451/abc.04.03.13.2017
Santacà, M., Agrillo, C., & Miletto Petrazzini, M. E. (2021). The Challenge of Illusory Perception of Animals: The Impact of Methodological Variability in Cross-Species Investigation. Animals, 11(6). https://doi.org/10.3390/ani11061618
Santacà, M., Lucon-Xiccato, T., & Agrillo, C. (2020). The Delboeuf illusion's bias in food choice of teleost fishes: an interspecific study. Animal Behaviour, 164, 105–112. https://doi.org/10.1016/j.anbehav.2020.04.012
Santacà, M., Miletto Petrazzini, M. E., Agrillo, C., & Wilkinson, A. (2019). Can reptiles perceive visual illusions? Delboeuf illusion in red-footed tortoise (Chelonoidis carbonaria) and bearded dragon (Pogona vitticeps). Journal of Comparative Psychology, 133(4), 419–427. https://doi.org/10.1037/com0000176
Schaeffer, R. G., & Sikes, J. D. (1971). Discrimination learning in dairy calves. Journal of Dairy Science, 54(6), 893–896. https://doi.org/10.3168/jds.S0022-0302(71)85937-4
Schaffer, A., Caicoya, A. L., Colell, M., Holland, R., Ensenyat, C., & Amici, F. (2020). Gaze Following in Ungulates: Domesticated and Non-domesticated Species Follow the Gaze of Both Humans and Conspecifics in an Experimental Context. Frontiers in Psychology, 11, 604904. https://doi.org/10.3389/fpsyg.2020.604904
Schaffer, A., Caicoya, A. L., Colell, M., Holland, R., Fersen, L. von, Widdig, A., & Amici, F. (2021). Neophobia in 10 ungulate species-a comparative approach. Behavioral Ecology and Sociobiology, 75(7), 102. https://doi.org/10.1007/s00265-021-03041-0
Segall, M., Campbell, D., & Herskovits, M. (1966). The Influence of Culture on Visual Perception. Art Education, 23(6), 30. https://doi.org/10.2307/3191488
Shapiro, A. G., & Todorovic, D. (Eds.) (2017). The Oxford Compendium of Visual Illusions. Oxford University PressNew York. https://doi.org/10.1093/acprof:oso/9780199794607.001.0001
Sovrano, V. A., da Pos, O., & Albertazzi, L. (2016). The Müller-Lyer illusion in the teleost fish Xenotoca eiseni. Animal Cognition, 19(1), 123–132. https://doi.org/10.1007/s10071-015-0917-6
Suganuma, E., Pessoa, V. F., Monge-Fuentes, V., Castro, B. M., & Tavares, M. C. H. (2007). Perception of the Müller-Lyer illusion in capuchin monkeys (Cebus apella). Behavioural Brain Research, 182(1), 67–72. https://doi.org/10.1016/j.bbr.2007.05.014
Sugnaseelan, S., Prescott, N. B., Broom, D. M., Wathes, C. M., & Phillips, C. J. (2013). Visual discrimination learning and spatial acuity in sheep. Applied Animal Behaviour Science, 147(1-2), 104–111. https://doi.org/10.1016/j.applanim.2013.04.012
Szenczi, P. [Péter], Velázquez-López, Z. I., Urrutia, A. [Andrea], Hudson, R. [Robyn], & Bánszegi, O. [Oxána] (2019). Perception of the Delboeuf illusion by the adult domestic cat (Felis silvestris catus) in comparison with other mammals. Journal of Comparative Psychology, 133(2), 223–232. https://doi.org/10.1037/com0000152
Taylor, A. A., & Davis, H. [Hank] (1998). Individual humans as discriminative stimuli for cattle (Bos taurus). Applied Animal Behaviour Science, 58(1-2), 13–21. https://doi.org/10.1016/S0168-1591(97)00061-0
Tudusciuc, O., & Nieder, A. (2010). Comparison of length judgments and the Müller-Lyer illusion in monkeys and humans. Experimental Brain Research, 207(3-4), 221–231. https://doi.org/10.1007/s00221-010-2452-7
Wasserman, E. A., Lazareva, O. F., & Shimizu, T. (2012). How Animals See the World: Comparative Behavior, Biology, and Evolution of Vision. New York: Oxford University Press. https://doi.org/10.1093/acprof:oso/9780195334654.001.0001
White-Lewis, S. (2020). Equine-assisted therapies using horses as healers: A concept analysis. Nursing Open, 7(1), 58–67. https://doi.org/10.1002/nop2.377
Yeates, J., & McGreevy, P. (2019). Ungulates (Ungulata ). In J. Yeates (Ed.), Companion animal care and welfare: The UFAW companion animal handbook (pp. 249–265). Hoboken, NJ: John Wiley & Sons, Inc. https://doi.org/10.1002/9781119333708.ch12

Table 1. For the Müller-Lyer and Delboeuf tasks, study subjects, species, sex, performance in the experimental conditions (i.e. number of correct trials/number of trials) and p values of the corresponding binomial test (marked with an asterisk if significant).

Species	Subjects	Sex	Müller-Lyer task		Delboeuf task
Species	Subjects	Sex	Performance	P	Performance	P
Guanacos	Phibie	Female	22/24	<0.001*	10/12	0.039*
	Lolita	Female	21/24	<0.001*	9/12	0.146
	Rike	Female	21/24	<0.001*	9/12	0.146
	Lissitha	Female	20/24	0.002*	10/12	0.039*
	Maike	Female	17/24	0.064	9/12	0.146
Llamas	Sanchio	Male	17/24	0.064	6/12	0.5
	Krumel	Male	15/24	0.308	9/12	0.146
	Flax	Male	18/24	0.023*	6/12	0.5
Goats	Bacca	Female	17/24	0.064	10/12	0.039*
	Frangia	Female	18/24	0.023*	9/12	0.146
	Nina	Female	21/24	<0.001*	9/12	0.146
	Zampa	Female	19/24	0.007*	10/12	0.039*
Sheep	Trilli	Female	13/24	0.839	9/12	0.146
	Bianca	Female	19/24	0.007*	9/12	0.146
	Lady	Female	17/24	0.064	9/12	0.146
	Goccia	Female	20/24	0.002*	9/12	0.146
	Fiocco	Male	21/24	<0.001*	-	-

Table 2. For both models, estimates, standard errors (SE), confidence intervals (CIs), likelihood ratio tests (LRT), degrees of freedom (df), and p-values for each test predictor (marked with an asterisk if significant) and control (in italics); reference categories are in parentheses.

Models, predictors and controls	Estimate	SE	2.5% to 97.5% CI	df	*LRT*	P
Model 1: Müller-Lyer task
Intercept	1.38	0.26	0.88 to 1.88	-	-	-
Condition (Control B)	-0.06	0.23	-0.52 to 0.39	5	4.20	0.521
Condition (Control C)	0.28	0.24	-0.19 to 0.76
Condition (Control D)	0.10	0.24	-0.36 to 0.56
Condition (Experimental long)	-0.03	0.23	-0.49 to 0.43
Condition (Experimental short)	0.29	0.24	-0.19 to 0.77
Species (guanacos)	-0.04	0.19	-0.41 to 0.34	3	4.61	0.202
Species (llamas)	-0.22	0.21	-0.63 to 0.20
Species (sheep)	0.24	0.20	-0.14 to 0.63
Trial number	-0.05	0.02	-0.09 to -0.01	1	6.77	0.009
Choice.side	0.23	0.14	-0.04 to 0.50	1	2.72	0.099
Model 2: Delboeuf task
Intercept	0.80	0.31	0.19 to 1.41	-	-	-
Condition (Control B)	0.14	0.25	-0.34 to 0.62	2	1.03	0.596
Condition (Experimental)	-0.10	0.24	-0.57 to 0.36	2	1.03	0.596
Species (guanacos)	0.14	0.27	-0.39 to 0.66	3	1.95	0.584
Species (llamas)	-0.24	0.29	-0.80 to 0.33
Species (sheep)	0.10	0.28	-0.46 to 0.65
Trial number	0.03	0.03	-0.02 to 0.09	1	1.42	0.234
Choice.side	0.24	0.20	-0.15 to 0.64	1	1.48	0.224

No competing interests reported.

Download PDF

Editorial decision: Revision requested
05 Feb, 2024
Reviews received at journal
31 Jan, 2024
Reviewers agreed at journal
12 Jan, 2024
Reviewers invited by journal
08 Jan, 2024
Editor assigned by journal
03 Jan, 2024
Submission checks completed at journal
16 Dec, 2023
First submitted to journal
16 Dec, 2023

You are reading this latest preprint version

Optical illusions in different ungulates species

Status:

Version 1

Abstract

Figures

Introduction

Methods

Results

Discussion

Declarations

References

Tables

Additional Declarations

Status:

Version 1