There is great variation in eye morphology across the animal kingdom. Variation in external eye appearance has been inspected in many different animal taxa, but still little is known about its putative adaptive value. Most efforts to study the adaptive value have been directed towards understanding primate variation in eye appearance (Mayhew & Gómez, 2015; Caspar et al., 2021; Perea-García et al., 2022, Mearing et al., 2022), mainly because of the proposal that in contrast to the eyes of other primates, human eyes are uniquely adapted for social functions (Kobayashi & Kohshima, 1997). Nevertheless, studies on the topic have also examined such broadly distributed groups as fish (Volpato et al., 2003), frogs (Amat et al., 2013), and birds (Craig and Hulley, 2004; Davidson et al., 2014, Delhey, 2015; 2018; 2019; Yorzinski et al., 2022). So far, the literature on the adaptive significance of external eye appearance has distinguished two alternative, but not mutually exclusive hypotheses that suggest either a social or an ecological function, in both primates (Kobayashi and Kohshima, 1997; 2001; Mearing et al., 2022; Perea-García et al., 2022) and birds (or “signaling” and “survival” functions in Corbett et al., 2022).
In primates, the most exhaustively explored functions of external eye appearance are social in nature. Kobayashi & Kohshima (1997; 2001) evaluated the external eye morphology of an extensive range of primate species, proposing that primate eyes could be categorized as "cryptic" (gaze camouflage hypothesis) or "conspicuous" (gaze signaling hypothesis). These authors concluded that the human eye is unique in its features as it allows easy detection of gaze direction and as such could have a communicative function. It has, for example, been hypothesized that ocular conspicuity may enhance interindividual coordination by easing the perception of eye gaze, which can be used as a spatial pointer (Tomasello et al., 2007). Supporting social functions, recent studies show that modern-day humans differentially perceive individuals based on changes in eye appearance (Wacewicz et al., 2022), possibly via mechanisms rooted in early development (Wolf et al., 2023). Thus, rather than facilitating eye-gaze following, these studies support the notion that the external appearance of eyes can be an honest indicator of temperament.
Whereas Kobayashi and Kohshima (1997; 2001) focused on social functions as drivers of variation in external eye appearance, they specifically rejected ecological alternatives such as the anti-glare theory (Duke-Elder, 1985), which proposes that pigmentation patterns in and around the eye serve to enhance visual functions in uneven lighting. More recently, researchers have revisited these potential ecological functions of external eye appearance in primates (Perea-García et al. 2021), with results providing support for ecological functions (Perea-García et al., 2022). At the infraorder level (anthropoid primates), iris color was found to change depending on the distance of the species’ distribution from the equator (Perea-García et al., 2022). Fur, hair and skin can provide protection against ultraviolet radiation, especially as they become darker closer to the equator (Caro, 2005; Kamilar & Bradley, 2011). Species living closer to the equator thus also have more pigmented conjunctivae, supporting the photoprotective functions of the external eye appearance (Perea-García et al., 2022). In addition to the link between iris coloration and latitude, the shape of the eye aperture seems to follow allometric scaling rules (Kobayashi and Kohshima, 1997; 2001; Perea-García et al., 2022); i.e., the amount of exposed eyeballs (measured as Sclera Size Index or SSI; Kobayashi & Kohshima, 1997) increases with increasing body weight (Kobayashi & Kohshima, 1997). As bodies become greater, the eyeballs become relatively lighter than the head. With big heads, it is therefore more economical to move the eyeballs rather than the head. To our knowledge, however, eye fissures have not been inspected with regard to body mass outside primates. While the allometric scaling relationship between eye and head mass should hold across vertebrate taxa, this may be less so in birds, because the skeletons of birds tend to be lighter than those of primates (Van Tyne & Berger 1966; Villee et al., 1973; Hickman et al., 1974). Supporting this hypothesized relaxed scaling between eye and head mass is the observation that, even though birds are capable of moving their eyeballs, they also rely on head movements to reorient their attention (Tyrrell et al., 2015).
As in primates, some evidence supports a role of external eye appearance with regard to ecological functions in birds. In a recent review of both avian eye morphology and the literature related to its evolutionary function, Corbett et al. (2022) summarize the proposed association between iris color and vision, camouflage, habitat, and foraging strategies (termed “survival-based” functions). For example, Passarotto et al. (2018) proposed camouflage-related hypotheses based on their finding that nocturnal owls tend to have darker irises compared to other owl species with diurnal and crepuscular habits in order to avoid detection by prey while hunting. In a similar manner, Davidson and colleagues (2016) found that birds with non-cavity nests evolved to have darker eye features compared to cavity nesters, who had more conspicuous eyes. The authors explained that having bright eyes could have drawbacks for non-cavity nesters as they could draw attention and increase (nest-)predation risk. Alternative results, however, suggest that dark pigmentation near the pupil has a role in reducing glare and improving vision in Turnix species, thus suggesting anti-glare functions for eye appearance (Gutiérrez‑Expósito, 2019).
Notwithstanding the studies on potential ecological drivers of external eye appearance, perhaps most attention to the variation in eye coloration in bird species has, like in primates, been directed to proposed social functions. Gaze-following, which is the ability to monitor the gaze directions of other individuals (Emery et al., 1997), is not unique to primates, and birds are known to be sensitive to humans’, predators and conspecifics’ gaze (Schloegl et al., 2007; von Bayern and Emery, 2009; Carter et al., 2008; Yorzinski et al., 2022). Corbett et al. (2022), therefore, categorized the social functions related to variations in avian external eye appearance as ‘signaling’. This category includes functional explanations related to nest defense, competition, flock vigilance, and breeding (Corbett et al., 2022). Previous research found for example that jackdaws, cavity-nesting birds with brightly colored eyes, compete for their nests but they also avoid conflicts if the nest is occupied by another jackdaw. Based on that, Davidson et al. (2014) suggested that conspicuous eyes can be seen outside of the nest, so it is a useful signal to prevent conflict. Adding to the ways in which birds may use eyes in conspecific signaling, a growing number of studies report that birds can voluntarily change their pupil size during social interaction associated with agonistic behavior, happiness, or excitement (Marcuk et al., 2020; Moore et al., 2022). One study shows that, in Japanese quails, increases in pupil size can be observed during pleasant behavioral sequences like dustbathing (Bertin et al., 2018). Unlike the smooth muscle fibers in the iris of mammals (Huber et al., 2007), those of birds are predominantly striated (Coli et al., 2016), suggesting that avian pupil size changes can be under voluntary control. Numerous lay reports also note the propensity of Psittacidae (parrots) to change pupil size rapidly as part of visually striking displays, known as “eye pinning” or “eye flashing”, which is also associated with agonistic behavior, happiness or excitement (Hardy, 1963; Smith 1973; Marcuk et al., 2020; Moore et al., 2022). In short, regardless of the voluntary or involuntary nature of pupil size changes in birds, these appear to comply with communicative functions for the external eye appearance either as part of displays or by enabling the inference of internal states and emotions.
Among birds, parrots (Psittaciformes) are an interesting taxonomic group to test hypotheses about the relative contribution of social and environmental selection pressures in shaping the appearance of the eye. Most of the species within the Psittaciformes order (i.e., parrots) do not have visible sclera, however, iris coloration is extremely diverse from bright to dark colors, with species showing stark contrast between the inner (proximal to the pupil) and outer (distal to the pupil) portion of the iris, such as, for example, a gray inner part of the iris and red outer part of the iris (top right square of Fig. 1, Trichoglossus haematodus), or a dark green inner part of the iris and bright yellow outer part of the iris (mid top square of Fig. 1, Deroptyus accipitrinus). In addition, parrots have: I) a wide range of body sizes, from 14 grams (Micropsitta pusio) to 2250 grams (Strigops habroptilus: Dunning, 2007) in our sample, which makes them an interesting system to explore the relationship between body weight and exposure of the eyeball, as was found in primates (Kobayashi and Kohshima, 1997), even after correcting for phylogenetic relatedness (Perea-García et al., 2022). II) Parrots are geographically widespread and can live in different places around the world, such as the Austral conure (living in the southern hemisphere below 54S (Tierra del Fuego) and the Alexandrine Parakeet (Psittacula eupatria) occurring northerly at around 33N (Afghanistan; Juniper & Parr, 1998). Similarly, they occur at different altitudes, with some (e.g. Psilopsiagon aurifrons) being sighted even at 4500 meters. As such parrots allow for testing hypotheses that suggest ecological functions for external eye appearance. Finally, III) there is enormous variation in social systems within this taxonomic group, specifically with regard to breeding systems, sexual dimorphism and flock size, allowing the examination of social hypotheses. Therefore, in the present study, we investigate the potential effect of allometric scaling rules, and variables of ecological and social nature on the external eye appearance of Psittaciformes.
Hypotheses
With regards to ecological factors and allometric scaling rules, we first expected heavier parrots to have more exposed eyeballs (Sclera Surface Index: SSI). While we expect SSI to increase with mass, this relationship may be less pronounced than in primates due to the lighter skeleton of birds (Van Tyne & Berger, 1966; Villee et al., 1973; Hickman et al., 1974).
Specifically with regards to photopic ecological factors, we investigated whether external eye appearance is related to the ambient lighting typical of species’ ranges by using latitude and maximum flying altitude as proxies for typical amounts of light. Latitude and altitude influence the amount of UV light, with greater amounts of UV radiation found closer to the equator (Schmucki & Philipona, 2002; Grigalavicius et al., 2015; Wald, 2018) and at higher altitudes (Blumtaler et al., 1997). Pigmentation can absorb excess UV radiation, thus preventing damage to tissue and it may also regulate the amount and quality of light entering the retina, enhancing visual perception (Miller & Sanghvi, 1990; Gutiérrez‑Expósito, 2019). Therefore, we expected to find darker irises closer to the equator and at higher altitudes (cf. Perea-García et al., 2022). Because the thickness of the troposphere differs depending on latitude (being thickest at the equator and thinning with distance from it), we also added an interaction term in our model.
Regarding social functions, we investigated potential links between species’ breeding system, flock size and conspicuity of the eye, expecting that cooperatively breeding species would display greater differences in brightness between parts of the eye. Contrast between pupil and iris could enhance the perception of changes in pupil size, which has been shown to regulate processes of physiological synchrony in humans and other apes (Kret et al., 2014), and to mediate the perception of trust in humans (Kret, Fischer & De Dreu, 2015). During social dominance interactions and courtship displays, parrots are engaged in constriction of the pupil which is called ‘eye-pinning’ (Hardy, 1963; Smith, 1973; Craig, 2022). One study shows that rapid pupil constriction occurs in synchrony with vocalizations in a parrot (Amazona ochrocephala panamensis), suggesting that rapid changes in pupil size may be part of a multimodal complex of audiovisual displays (Gregory & Hopkins, 1974).
We also tested whether chromatic salience of the eyes of sexually dimorphic species was greater than species where both sexes have a similar physical appearance. Although not as common as differences in plumage, sexual dimorphism in eye color has been reported for birds (Corbett et al., 2022), and among parrots, the iris does vary between the sexes in some species (e.g., Cacatua species: Smith, 1973). We expect the inner and outer part of the irises to be more different in hue in sexually dimorphic species as this difference could act as an attention-getter and/or hybrid trait to prevent introgression in closely related species.
Lastly, we investigated whether greater differences in brightness between the pupil and the inner part of the iris, and between the inner and the outer part of the iris were related to flock size. This contrast could help in assessing flight direction of neighboring individuals, to facilitate flocking behavior. This would be an extension of the proposed link between ocular conspicuity and spatial coordination between conspecifics in e.g. primates (Tomasello et al., 2007).