If horses had toes: demonstrating mirror self recognition at group level in Equus caballus. Paolo Baragli, Chiara Scopa, Veronica Maglieri & Elisabetta Palagi. Animal Cognition, Mar 13 2021. https://rd.springer.com/article/10.1007/s10071-021-01502-7
Rolf Degen's take: "Our results suggest the presence of mirror self-recognition in horses"
Abstract: Mirror self-recognition (MSR), investigated in primates and recently in non-primate species, is considered a measure of self-awareness. Nowadays, the only reliable test for investigating MSR potential skills consists in the untrained response to a visual body mark detected using a reflective surface. Here, we report the first evidence of MSR at group level in horses, by facing the weaknesses of methodology present in a previous pilot study. Fourteen horses were used in a 4-phases mirror test (covered mirror, open mirror, invisible mark, visible colored mark). After engaging in a series of contingency behaviors (looking behind the mirror, peek-a-boo, head and tongue movements), our horses used the mirror surface to guide their movements towards their colored cheeks, thus showing that they can recognize themselves in a mirror. The analysis at the group level, which ‘marks’ a turning point in the analytical technique of MSR exploration in non-primate species, showed that horses spent a longer time in scratching their faces when marked with the visible mark compared to the non-visible mark. This finding indicates that horses did not see the non-visible mark and that they did not touch their own face guided by the tactile sensation, suggesting the presence of MSR in horses. Although a heated debate on the binary versus gradualist model in the MSR interpretation exists, recent empirical pieces of evidence, including ours, indicate that MSR is not an all-or-nothing phenomenon that appeared once in phylogeny and that a convergent evolution mechanism can be at the basis of its presence in phylogenetically distant taxa.
Discussion
Here, we report the first evidence of mirror self-recognition at the group level in a non-primate species. Furthermore, using a larger sample size and applying a more accurate experimental procedure, the present study replicates a previous pilot study on mirror self-recognition in horses (Baragli et al. 2017).
Our horses used the mirror surface to guide their movements towards their faces previously marked, thus showing that they are able to recognize themselves in a mirror. They followed a sequence of behavioral steps towards the mirror before being marked. This is a fundamental criterion to be fulfilled before undergoing the mark test, as suggested by de Waal (2019), Gallup et al. (2002) and Gallup and Anderson (2019) in their reviews focused on the methodological issues. These steps are indicative of the cognitive processes leading animals to understand that the image reflected in the mirror is the image of self (Plotnik et al. 2006).
Firstly, we found that in presence of the reflective surface the behavior of the horses clearly differed when compared to the condition in which the surface was covered. Both selective attention and exploratory activity increased when the mirror was open, indicating the emergence of the violation of the expectancy phenomenon (Seyfarth et al. 2005; Poulin-Dubois et al. 2009; Kondo et al. 2012). Through the violation of expectancy paradigm, it has been demonstrated that horses are able to associate multiple sensory cues to recognize conspecifics and people (cross-modal recognition, Proops et al. 2009; Proops and McComb 2012). While the image in the mirror satisfied the visual criterion (there is a horse in the mirror sensu Lorenz 1974), the tactile and olfactory information did not match with the visual one (it is not a horse sensu Lorenz 1974) thus producing an incongruent set of information.
The information gathered by the selective attention and exploratory activities increased the horse’s motivation in engaging in contingency behaviors to solve such incongruency (Seyfarth et al. 2005). The so-called contingency behaviors include highly repetitive non-stereotyped or unusual movements only when animals are in front of the reflective surface, probably to verify if the movements of the image in the mirror match their own movements. When in front of the mirror, magpies moved their head or body back and forth (Prior et al. 2008), elephants displayed repetitive, non-stereotypic trunk and body movements (Plotnik et al. 2006), jackdows and crows showed “peek-a boo” movements during which the bird moved out and back in sight of the mirror (Soler et al. 2014; Vanhooland et al. 2019) and chimpanzees manipulated their lips and tongues while glancing into the mirror (Povinelli et al. 1993). Our horses engaged in contingency behaviors similar to those reported for other species such as head movements, peek-a-boo, and tongue protrusion almost exclusively in presence of the reflective surface (Table 3). It is possible that by slightly moving their head horses managed to avoid the blind spot characterizing their frontal view (Lansade et al. 2020) thus head movements could help verify whether the movements of the reflective image corresponds to their movements (Online Resource 6). One of the most indicative contingency behaviors reported in the literature is looking behind the mirror that is enacted to verify the possible presence of a conspecific behind the reflective surface (Pica pica, Prior et al. 2008; Equus caballus, Baragli et al. 2017; Loxodonta africana, Plotnik et al. 2006; Pan troglodytes, Gallup 1970; Povinelli et al. 1993) (Online Resource 5). Our horses showed a high inter-individual variability in performing contingency behaviors in front of the reflective surface. We suggest that the strategy employed to test the mirror function varies among subjects that engaged in one or two contingency behaviors to solve the violation of expectancy (Table 3). This means that when studying MSR we should take into account for this variability by also checking a posteriori what animals do to test their own image reflected in the mirror (unusual, repetitive non-stereotyped behaviors), thus leaving open the ethogram fixed a priori.
After solving the violation of expectancy by engaging in contingency behaviors, animals gather the necessary information to potentially pass the mark test. In this study, due to the anatomical features limiting the degree of freedom of horses to reach specific areas of their face, we considered scratching the face (Face-SCR) as an attempt to remove the mark which was placed on both cheeks (bilateral marking) (Online Resource 1 and 9–12). The analysis at a group level showed that horses spent a longer time in scratching their face when marked with the colored mark compared to the sham mark (S vs M conditions). This finding indicates that horses did not see the sham mark and that it was not the tactile sensation that induced the animal to touch its own face. The increased level of Face-SCR during the M condition suggests that by using the reflective surface the animals were able to visually perceive the colored spot on their face. The standardization of the procedure preceding the application of the mark, such as grooming on the whole body and identical shapes of the sham and colored stamps, guarantees that the use of the transparent mark worked as an effective control condition. An additional control in supporting the hypothesis that horses are able to perceive the colored spot on their face resides in the comparable levels of time spent in scratching directed to the rest of the body (Body-SCR). In the M condition, scratching appears to be highly directional towards a specific target: the colored face (Online Resource 16).
One of the novelties of the present study relies on the analysis at a group level, which ‘marks’ a turning point in the analytical technique of MSR exploration. It has been suggested that the individual variability in the MSR tests can reflect the low motivation of animals to remove the colored mark. The low motivation to react to the mark can introduce a strong individual bias in the accurate measurement of self-recognition abilities (Bard et al. 2006; Heschl and Burkart 2006). In our case, for example, four horses that did not scratch their faces in the S condition did it in the M condition but not for sufficient time to apply an individual test (expected frequencies < 5.0 s; see Table 4). The behavioral motivation of removing something from one’s own body, and to respond to the colored mark, is considered a hotspot in the discussion about the validity of the mark test for demonstrating MSR. In this perspective, the analysis at the population level provides the opportunity to employ larger samples also including the subjects showing low levels of motivation. Such individual motivation can also be affected by a series of species-specific features (e.g., anatomical difference in properly reaching the marked area, visual perception of specific colors, visual acuity, predominant sensory modality different from vision), including personality and cognitive style. Therefore, the sensory and cognitive systems, as well as the motivation to behaviorally respond to the mark, are substantial preconditions to keep in mind when we decide to test animals’ self-recognition abilities.
In conclusion, despite the strong inter-individual variability, our results suggest the presence of MSR in horses. Although the heated debate on the binary versus gradualist model in the MSR interpretation (de Waal 2019; Gallup and Anderson 2019; Brandl 2016), recent empirical pieces of evidence, including ours on horses, indicate that MSR is not an all-or-nothing phenomenon suddenly emerged in the phylogeny, but it has probably been favored by natural selection to adaptively respond to social and cognitive challenges an animal has to cope with.