Fighting With Your Own Reflection: Can Animals Perceive The Differences Between Natural And Fake Opponents?

While standing in front of a mirror, it is easy for humans to recognize that the image in the mirror is nothing but a reflection. However, for animals lacking self-recognition, it is difficult to tell whether the one behind the mirror is real or fake.

Indeed, most studies assume that animals that lack self-recognition would consider the mirror image as a real conspecific and, therefore, should respond to the image in ways that mimic interactions with a live, unfamiliar individual. However, this assumption has been questioned because behavioral responses toward a mirror image and toward an actual opponent are not always correlated. Additionally, interactions with a mirror image and an actual opponent elicit different hormonal responses and different patterns of neural activity in the brain.

These inconsistencies in behavioral and physiological responses to image opponents and real conspecifics may result from the orientations that animals adopt during aggressive contests, in particular during “lateral displays.” During a social interaction, individuals often display one side of their body to exhibit physical traits to an opponent, which allows animals to exchange information, such as size and competitive ability, before escalating to high-intensity combat.

Lateral display is essential in many fish species because it provides an opportunity for individuals to use the lateral line system (a string of sensory receptors extending the length of the body) to assess opponent information without engaging in physical contact. Individuals exhibit lateral displays in two basic ways, either with their heads aligned in the same direction (head-head lateral display), or in opposite directions (head-tail lateral display).

A regular mirror forces animal to engage in head-head lateral displays, which is not the norm for many species, including the vast majority of fishes. When exposed to regular mirrors, species that prefer head-tail lateral display are thus unable to acquire information about the opponent in ways that simulate the dynamics of natural contests, and this might explain why the stimuli elicit very different behavioral, physiological, and neurobiological responses.

A recent study, therefore, presented a new design, the “nonreversing mirror,” which allows animals that typically use head-tail lateral displays during contests to experience a true reflection. A regular mirror design generates a left-right reversing image of the same size as the testing animals in front of it, while a nonreversing mirror, which is made by connecting two regular first-surface mirrors at a 90-degree angle, generates a nonreversing image (or so-called “true reflection”) of the same size as the testing subject. This new design allows animals that prefer head-tail orientation to experience a more appropriate type of lateral display during interactions with the mirror image.

This research team used mangrove rivulus fish, Kryptolebias marmoratus, as a study organism to investigate behavioral differences between individuals exposed to regular and nonreversing mirrors and to examine which test best predicted aggressive performance toward an actual opponent. Their results demonstrate that animals behave differently when exposed to regular and nonreversing mirrors as aggression-eliciting stimuli. Rivulus fish preferred using right lateral display in both regular and nonreversing mirror tests, while those in the nonreversing mirror tests delivered more attacks to the opponent than those in the regular mirror test.

Interestingly, fish in the regular mirror tests switched their orientation more frequently than those in nonreversing mirror tests, suggesting that animals were attempting to achieve the appropriate displays, which was ultimately impossible. Asymmetries between two fish in their number of attacks toward the nonreversing mirror, but not toward the regular mirror, was strongly predictive of contest intensity and contest duration in real fights. Also, only the number of attacks directed towards the nonreversing mirror was predictive of which individual would win the real fight.

Compared with the regular mirror test, the nonreversing mirror test elicits behavior that better corresponds with rivulus’ performance during real interactions. These results demonstrate that regular mirrors and nonreversing mirrors likely convey different types of information to the focal animal. The nonreversing mirror provides a complementary method to study social interactions in a standardized fashion where animals can achieve appropriate display orientations that facilitate opponent assessment, and that more reliably predict actual contest behavior. This, in turn, could have broad implications for research ranging from behavioral neuroendocrinology to behavioral ecology and evolutionary biology.

These findings are described in the article entitled, Nonreversing mirrors elicit behaviour that more accurately predicts performance against live opponents, recently published in the journal Animal Behaviour. This work was conducted by Cheng-Yu Li, Caitlin Curtis, and Ryan Earley from the University of Alabama, Tuscaloosa.