Mirror test

A baboon looking in a mirror

The mirror test – sometimes called the mark test, mirror self-recognition test (MSR), red spot technique, or rouge test – is a behavioural technique developed in 1970 by psychologist Gordon Gallup Jr. as an attempt to determine whether an animal possesses the ability of visual self-recognition. The MSR test is the traditional method for attempting to measure self-awareness. However, there has been agreement that animals can be self-aware in ways not measured by the mirror test, such as distinguishing between their own and others' songs and scents. On the other hand, animals that can pass the MSR do not necessarily have self-awareness.

In the classic MSR test, an animal is anaesthetised and then marked (e.g., painted, or a sticker attached) on an area of the body the animal cannot normally see. When the animal recovers from the anesthetic, it is given access to a mirror. If the animal then touches or investigates the mark, it is taken as an indication that the animal perceives the reflected image as itself, rather than of another animal.

Very few species have passed the MSR test. As of 2015, only great apes (including humans), a single Asiatic elephant, dolphins, orcas, and the Eurasian magpie have passed the MSR test. A wide range of species have been reported to fail the test, including several species of monkey, giant pandas, sea lions, and dogs.

Method and history

The inspiration for the mirror test comes from an anecdote about Charles Darwin and a captive orangutan. While visiting the London Zoo in 1838, Darwin observed an orangutan, named Jenny, throwing a tantrum after being teased with an apple by her keeper. This started him thinking about the subjective experience of an orangutan. He also watched Jenny gaze into a mirror and noted the possibility that she recognised herself in the reflection.

In 1970, Gordon Gallup, Jr., experimentally investigated the possibility of self-recognition with two male and two female wild pre-adolescent chimpanzees (Pan troglodytes), none of which had presumably seen a mirror previously. Each chimpanzee was put into a room by itself for two days. Next, a full-length mirror was placed in the room for a total of 80 hours at periodically decreasing distances. A multitude of behaviours was recorded upon introducing the mirrors to the chimpanzees. Initially, the chimpanzees made threatening gestures at their own images, ostensibly seeing their own reflections as threatening. Eventually, the chimps used their own reflections for self-directed responding behaviours, such as grooming parts of their body previously not observed without a mirror, picking their noses, making faces, and blowing bubbles at their own reflections.

Gallup expanded the study by manipulating the chimpanzees' appearance and observing their reaction to their reflection in the mirror. Gallup anaesthetised the chimpanzees and then painted a red alcohol-soluble dye on the eyebrow ridge and on the top half of the opposite ear. When the dye dried, it had virtually no olfactory or tactile cues. Gallup then returned the chimpanzees to the cage (with the mirror removed) and allowed them to regain full consciousness. He then recorded the frequency with which the chimpanzees spontaneously touched the marked areas of skin. After 30 minutes, the mirror was re-introduced into the room and the frequency of touching the marked areas again determined. The frequency of touching increased to 4–10 with the mirror present, compared to only 1 when the mirror had been removed. The chimpanzees sometimes inspected their fingers visually or olfactorily after touching the marks. Other mark-directed behaviour includes turning and adjusting of the body to better view the mark in the mirror, or tactile examination of the mark with an appendage while viewing the mirror.

An important aspect of the classical mark-test (or rouge test) is that the mark/dye is non-tactile, preventing attention being drawn to the marking through additional perceptual cues (somesthesis). For this reason, animals in the majority of classical tests are anesthetised. Some tests use a tactile marker. If the creature stares unusually long at the part of its body with the mark or tries to rub it off, then it is said to pass the test.

Animals that are considered to be able to recognise themselves in a mirror typically progress through four stages of behaviour when facing a mirror:

1. social responses
2. physical inspection (e.g. looking behind the mirror)
3. repetitive mirror-testing behaviour
4. realisation of seeing themselves

Gallup conducted a follow-up study in which two chimpanzees with no prior experience of a mirror were put under anesthesia, marked and observed. After recovery, they made no mark-directed behaviours either before or after being provided with a mirror.

The rouge test was also done by Michael Lewis and Jeanne Brooks-Gunn in 1979 for the purpose of self-recognition with human mothers and their children.

Animals that have passed

European magpies have demonstrated mirror self recognition

A large number of studies using a wide range of species have investigated the occurrence of spontaneous, mark-directed behaviour when given a mirror, as originally proposed by Gallup. Most marked animals given a mirror initially respond with social behaviour, such as aggressive displays, and continue to do so during repeated testing. Only a small number of species have touched or directed behaviour toward the mark, thereby passing the classic MSR test.

Findings in MSR studies are not always conclusive. Even in chimpanzees, the species most studied and with the most convincing findings, clear-cut evidence of self-recognition is not obtained in all individuals tested. Prevalence is about 75% in young adults and considerably less in young and aging individuals.

Until the 2008 study on magpies, self-recognition was thought to reside in the neocortex area of the brain. However, this brain region is absent in non-mammals. Self-recognition may be a case of convergent evolution, where similar evolutionary pressures result in similar behaviours or traits, although species arrive at them via different routes, and the underlying mechanism may be different.

Mammals

Cetaceans

• Bottlenose dolphin (Tursiops truncatus): Researchers in a study on two male bottlenose dolphins observed their reactions to mirrors after having a mark placed on them. Reactions such as decreased delay in approaching the mirror, repetitious head circling and close viewing of the eye or genital region which had been marked, were reported as evidence of MSR in these species.
• Killer whale (Orcinus orca): Killer whales and false killer whales (Pseudorca crassidens) may be able to recognise themselves in mirrors.

Primates

• Bonobo (Pan paniscus)
• Bornean orangutan (Pongo pygmaeus): However, mirror tests with a juvenile (2-year-old), male orangutan failed to reveal self-recognition.
• Chimpanzee (Pan troglodytes): However, mirror tests with a juvenile (11 months old) male chimpanzee failed to reveal self-recognition. Two young chimpanzees showed retention of MSR after one year without access to mirrors.
• Human (Homo sapiens): Typically, humans begin to show self-recognition in the mirror test when they are about 18 months old, or in what psychoanalysts call the "mirror stage". A 2010 cross-cultural study observed variations in the presence of self-oriented behaviors exhibited by children (ranging from 18 to 55 months old) from non-Western rural communities and Western urban and rural communities when each were given the mark test. Their results indicate a distinction between cultures and communities. They found that children from Western communities showed earlier signs of self-oriented behaviors toward the mark when given the mirror mark test. Whereas, there was an absence of this behavior in children from non-Western communities. Such results do not suggest a delayed development in cognition in the latter group, but rather the potential of how differences in parenting styles (as influenced by culture) impact the way children express self-concept. "In relation to the mark test used in the present studies, we think that compliance norms shape the way children manifest self-recognition, specifically by not touching the mark. This is in sharp contrast with the independence and self-initiative that tends to be encouraged and nurtured in the industrial West, especially in the middle- and upper-classes of the majority cultures." Ultimately, this study questions the universal validity of the mirror mark test as an accurate measurement of self-concept .

Proboscidea

• Asian elephant (Elephas maximus): In a study performed in 2006 three female Asian elephants were exposed to a large mirror to investigate their responses. Visible marks and invisible sham-marks were applied to the elephants' heads to test whether they would pass the MSR test. One of the elephants showed mark-directed behaviour, though the other two did not. An earlier study failed to find MSR in two Asian elephants; it was claimed this was because the mirror was too small. The study was conducted with the Wildlife Conservation Society (WCS) using elephants at the Bronx Zoo in New York. All three Asian elephants in the study were standing in front of a 2.5 m-by-2.5 m mirror—they inspected the rear and brought food close to the mirror for consumption. Evidence of elephant self-awareness was shown when one (and only one) elephant, Happy, repeatedly touched a painted X on her head with her trunk, a mark which could only be seen in the mirror. Happy ignored another mark made with colorless paint that was also on her forehead to ensure she was not merely reacting to a smell or feeling. Frans De Waal, who ran the study, stated, "These parallels between humans and elephants suggest a convergent cognitive evolution possibly related to complex society and cooperation."

Birds

Video of the responses of a European magpie in a MSR test. The magpie repeatedly attempts to remove the marks.

• Eurasian magpie (Pica pica): The Eurasian magpie is the first non-mammal to have passed the mirror test. Researchers applied a small red, yellow or black sticker to the throat of five Eurasian magpies, where they could be seen by the bird only by using a mirror. The birds were then given a mirror. The feel of the sticker on their throats did not seem to alarm the magpies. However, when the birds with coloured stickers glimpsed themselves in the mirror, they scratched at their throats—a clear indication that they recognised the image in the mirror as their own. Those that received a black sticker, invisible against the black neck feathers, did not react.

• Pigeons can pass the mirror test after training in the prerequisite behaviors.

Ants

• In a Belgian study from 2015, 23 out of 24 adult ants scratched at small blue dots painted on their clypeus (part of their "face") when they were able to see the dot in a mirror. According to the purported results, the ants were individually tested and were from three species, Myrmica sabuleti, Myrmica rubra and Myrmica ruginodis. None of the ants scratched the clypeus when they had no mirror to see the dot. None tried to scratch the blue dot on the mirror. When they had a mirror and a brown dot similar to their own color, only one of thirty ants scratched the brown dot; researchers said she was darker than average so the dot was visible. They also reacted to the mirror itself. Even without dots, 30 out of 30 ants touched the mirror with legs, antennae and mouths, while 0 of 30 ants touched a clear glass divider, with ants on the other side. Ants a few days old did not react to the dots. These three species have limited eyesight, with 109–169 facets per eye, and the authors suggest doing tests on ants with more facets (some have 3,000) and on bees.

Fish

• Cleaner Wrasse have become the first fish ever to pass the mirror test. They can recognize themselves in a mirror, according to a study done in 2019. The Cleaner Wrasse, or Labroides dimidiatus, is a tiny tropical reef fish who cleans other fish. When put through the mirror test, the Cleaner Wrasse showed all the behaviors of passing through the phases of the test. When provided with a colored tag in a modified mark test, the fish could see that in the mirror and attempted to scrape off this tag. They did this by scraping their bodies on the side of the mirror. Additionally, the fish did not try to remove the colored tag in the absence of a mirror, nor did they try to remove a transparent tag. This is considered a hallmark of cognition, and the Cleaner Wrasse remains the only fish to pass the mirror test. The experiment done by Masanori Kohda, et al. included many different aspects. Kohda and his team of scientists caught ten wild Cleaner Wrasses and put them into individual tanks, which were outfitted with mirrors. The fish were observed at first becoming aggressive with their reflections in the tanks, suggesting they may have viewed their reflection as another Cleaner Wrasse in its space. Eventually, the fish began to approach its own reflection in different ways, such as swimming towards it upside down. This is known as "contingency testing," or directly interacting with its own reflection in order to determine if it is themselves or another fish. After the fish became acquainted with the mirrors, the researchers injected a benign brown gel into the skin of the fish. Some of these injections were in places that the fish would not have been able to see if they were not by a mirror, such as their throats, which is an important thing to note. The fish probably identified the colored marks as a parasite, and began using whatever surfaces were around to scrape it off of them. This suggests that the fish recognized the fish with the mark in the mirror as themselves. It is also important to note that when the fish had the colored mark but no mirror, they did not try to scrape it off. The same happened when the fish were injected with a transparent gel. The fish also looked in the mirror before and after scraping their throats as if to get a better look. However, there is some doubt and controversy when it comes to this finding. Gordon Gallup, who invented the mirror test, is an evolutionary psychologist at the State University of New York at Albany, and he says the Cleaner Wrasses' behavior can be attributed to something other than recognizing itself in a mirror. Gallup has argued that a Cleaner Wrasse's job in life is to be aware of ectoparasites on the bodies of other fish, and therefore it would be hyper-aware of the fake parasite that it noticed in the mirror, perhaps seeing it as a parasite that it needed to clean off of a different fish. Gallup also argues that the Cleaner Wrasse's strange positioning towards the mirror (swimming upside down and looking at itself from different angles) may be because it is learning how to manipulate the "other fish" that it wants to clean. He says that the scraping behavior of the fish is most likely the fish trying to call attention to the "other fish" that it has a parasite on its body. However, Kohda, who ran the study, says that because the fish checked itself in the mirror before and after the scraping, this means that the fish has self awareness and recognizes that its reflection belongs to its own body.

It is also worth noting that the Cleaner Wrasses, when tested, spent a large amount of time with the mirror when they were first getting acquainted to it, without any training. The most important thing is that the Cleaner Wrasses performed scraping behavior with the colored mark, and they did not perform the same scraping behavior without the colored mark in the presence of the mirror, nor when they were with the mirror and had a transparent mark.

Animals that have failed

A range of species have been exposed to mirrors. Although these might have failed the classic MSR test, they have sometimes shown mirror-related behaviour:

Mammals

• Sea lions (Zalophus californianus)
• Giant panda (Ailuropoda melanoleuca): In one study, 34 captive giant pandas of a wide range of ages were tested. None of the pandas responded to the mark and many reacted aggressively towards the mirror, causing the researchers to consider the pandas viewed their reflection as a conspecific.

Primates

• Gibbon (g. Hylobates, Symphalangus and Nomascus)
• Stump-tailed macaque (Macaca arctoides)
• Crab-eating macaque (Macaca fascicularis)
• Rhesus monkey (Macaca mulatta): It has been reported that rhesus monkeys exhibit other behaviours in response to a mirror which indicate self-recognition.
• Black-and-white colobus monkey (Colobus guereza)
• Capuchin monkey (Cebus apella)
• Hamadryas baboon (Papio hamadryas)
• Cotton-top tamarin (Saguinus oedipus)

Birds

• Grey parrot
• New Caledonian crow
• Jackdaw
• Great tit (Parus major)

Fish

• Tanganyikan cichild, or Daffodil cichlid (Neolamprologus pulcher) is another fish that has failed the mirror test, according to a study done in 2017. These fish are typically regarded as socially intelligent, and can recognize conspecifics in their social groups. Therefore, they would theoretically make a good candidate for the mirror test. However, they ended up failing. Similar to the Cleaner Wrasse, the Tanganyikan cichlid first exhibited signs of aggression towards the mirrored image. After a colored mark was injected, the researchers found no increased scraping or trying to remove the mark, and the cichlids did not observe the side with the mark any longer than it would have otherwise. This demonstrates a lack of contingency checking, and means that the Tanganyikan cichlid did not pass the mirror test.

Octopuses

• Octopuses oriented towards their image in a mirror, but there is no difference in their behaviour in this condition, compared with a view of other octopuses.

Animals that may pass

Gorillas

Findings for gorillas are mixed. At least four studies have reported that gorillas failed the MSR test. It has been suggested that the gorilla may be the only great ape "which lacks the conceptual ability necessary for self-recognition". Other studies have found more positive results, but have tested gorillas with extensive human contact, and required modification of the test by habituating the gorillas to the mirror and not using anaesthetic. Koko reportedly passed the MSR test, this was without anaesthetic. In gorillas, protracted eye contact is an aggressive gesture and they may therefore fail the mirror test because they deliberately avoid making eye contact with their reflections. This could also explain why only gorillas with extensive human interaction and a certain degree of separation from other gorillas and usual gorilla behaviour are more predisposed to passing the test.

Fish

Two captive giant manta rays showed frequent, unusual and repetitive movements in front of a mirror suggested contingency checking. They also showed unusual self-directed behaviours when exposed to the mirror. Manta rays have the largest brains when it comes to fish. In 2016, Dr. Csilla Ari tested captive manta rays at the Atlantis Aquarium in the Bahamas by exposing them to a mirror. The manta rays appeared to be extremely interested in the mirror. They would behave strangely in front the mirror, including doing flips and moving their fins. They would also blow bubbles. They did not interact with the reflection as if it were another manta ray; they didn't try to socialize with it. However, only an actual mirror test can determine if they actually recognize their own reflection, or if they are just demonstrating exploratory behavior. A classic mirror test has yet to be done on manta rays.

Another fish that may pass the mirror test is an archerfish, or Toxotes chatareus. A study in 2016 showed that archerfish can discriminate between human faces. Researchers showed this by testing the archerfish, who spit a stream of water at an image of a face when they recognized it. The archerfish would be trained to expect food when it spat at a certain image. When the archerfish was showed images of other human faces, the fish did not spit. They only spit for the image that they recognized. Archerfish normally, in the wild, use their spitting streams to knock down prey from above into the water below. The study showed that archerfish could be trained to recognize a 3-D image of one face compared to an image of a different face, and would spit at the face when they recognized it. The archerfish were even able to continue recognizing the image of the face even when it was rotated 30, 60, and 90 degrees. This is an impressive task for a fish, so the archerfish could be worth testing in the mirror task, since it has already succeeded in a different visual task.

Other uses for mirrors

Primates, other than the great apes, have so far universally failed the mirror test. However, mirror tests with three species of gibbon (Hylobates syndactylus, H.gabriellae, H. leucogenys) have shown convincing evidence of self-recognition despite the fact that the animals failed the standard version of the mirror test.

Rhesus macaques have failed the MSR test, but use mirrors to study otherwise-hidden parts of their bodies, such as their genitals and the implants in their heads. It has been suggested this demonstrates at least a partial self-awareness, although this is disputed.

Pigs can use visual information seen in a mirror to find food, and show evidence of self-recognition when presented with their reflection. In an experiment, 7 of the 8 pigs tested were able to find a bowl of food hidden behind a wall and revealed using a mirror. The eighth pig looked behind the mirror for the food. BBC Earth also showed the foodbowl test, and the "matching shapes to holes" test, in the Extraordinary Animals series.

B. F. Skinner found that Pigeons are capable of passing a highly modified mirror test after extensive training. In the experiment, a pigeon was trained to look in a mirror to find a response key behind it, which the pigeon then turned to peck to obtain food. Thus, the pigeon learned to use a mirror to find critical elements of its environment. Next, the pigeon was trained to peck at dots placed on its feathers; food was, again, the consequence of touching the dot. The latter training was accomplished in the absence of the mirror. The final test was placing a small bib on the pigeon—enough to cover a dot placed on its lower belly. A control period without the mirror present yielded no pecking at the dot. When the mirror was revealed, the pigeon became active, looked in the mirror and then tried to peck on the dot under the bib. However, untrained pigeons have never passed the mirror test.

Manta rays repeatedly swim in front of the mirror, turning over to show their undersides and moving their fins. When in front of the mirror, they blow bubbles, an unusual behaviour. They do not try to socially interact with the mirror image, suggesting that they recognise that the mirror image is not another ray. However, a classic mirror test using marks on the rays’ bodies has yet to be done.

Robots

In 2012, early steps were taken to make a robot pass the mirror test.

Criticism

The MSR test has been criticised for several reasons, in particular, because it may result in findings that are false negatives.

The MSR test may be of limited value when applied to species that primarily use senses other than vision. For example, dogs mainly use olfaction and audition; vision is used only third. It is suggested this is why dogs fail the MSR test. With this in mind, the biologist Marc Bekoff developed a scent-based paradigm using dog urine to test self-recognition in canines. He tested his own dog, but his results were inconclusive. Dog cognition researcher Alexandra Horowitz formalized Bekoff's idea in a controlled experiment, reported on in 2016 and published in 2017. She compared the dogs' behavior when examining their own and others' odors, and also when examining their own odor with an added smell "mark" analogous to the visual mark in MSR tests. These subjects not only discriminated their own odor from that of other dogs, as Bekoff had found, but also spent more time investigating their own odor "image" when it was modified, as subjects who pass the MSR test do. A 2016 study suggested an ethological approach, the "Sniff test of self-recognition (STSR)" which may shed light on different ways of checking for self-recognition.

Another concern with the MSR test is that some species quickly respond aggressively to their mirror reflection as if it were a threatening conspecific thereby preventing the animal to calmly consider what the reflection actually represents. It has been suggested this is the reason why gorillas and monkeys fail the MSR test.

In a MSR test, animals may not recognise the mark as abnormal, or, may not be sufficiently motivated to react to it. However, this does not mean they are unable to recognise themselves. For example, in a MSR test conducted on three elephants, only one elephant passed the test but the two elephants that failed still demonstrated behaviours that can be interpreted as self-recognition. The researchers commented that the elephants might not have touched the mark because it was not important enough to them. Similarly, lesser apes infrequently engage in self-grooming, which may explain their failure to touch a mark on their head in the mirror test.

Frans de Waal, a biologist and primatologist at Emory University, has stated that self-awareness is not binary, and the mirror test should not be relied upon as a sole indicator of self-awareness, though it is a good test to have. Different animals adapt to the mirror in different ways.

Finally, it should be noted that there has been controversy over whether self-recognition implies self-awareness. The ant researchers state that many ants, from three species, pass the mirror test, but the researchers do not know that they have self-awareness. Dogs recognize their own scent as different from others' scents, but fail the mirror test.

Rouge test

A human child exploring his reflection

The rouge test is a version of the mirror test used with human children. Using rouge makeup, an experimenter surreptitiously places a dot on the face of the child. The child is then placed in front of a mirror and their reactions are monitored; depending on the child's development, distinct categories of responses are demonstrated. This test is widely cited as the primary measure for mirror self-recognition in human children.

Developmental reactions

From the age of 6 to 12 months, the child typically sees a "sociable playmate" in the mirror's reflection. Self-admiring and embarrassment usually begin at 12 months, and at 14 to 20 months most children demonstrate avoidance behaviours. Finally, at 18 months half of children recognise the reflection in the mirror as their own and by 20 to 24 months self-recognition climbs to 65%. Children do so by evincing mark-directed behaviour; they touch their own nose or try to wipe the mark off.

It appears that self-recognition in mirrors is independent of familiarity with reflecting surfaces. In some cases the rouge test has been shown to have differing results, depending on sociocultural orientation. For example, a Cameroonian Nso sample of infants 18 to 20 months of age had an extremely low amount of self-recognition outcomes at 3.2%. The study also found two strong predictors of self-recognition: object stimulation (maternal effort of attracting the attention of the infant to an object either person touched) and mutual eye contact. A strong correlation between self-concept and object permanence have also been demonstrated using the rouge test.

Implications

The rouge test is a measure of self-concept; the child who touches the rouge on his own nose upon looking into a mirror demonstrates the basic ability to understand self-awareness. Animals, young children, and people who have their sight restored after being blind from birth, sometimes react to their reflection in the mirror as though it were another individual.

Theorists have remarked on the significance of this period in a child's life. For example, psychoanalyst Jacques Lacan used a similar test in marking the mirror stage when growing up. Current views of the self in psychology position the self as playing an integral part in human motivation, cognition, affect, and social identity.

Methodological flaws

There is some debate as to the interpretation of the results of the mirror test, and researchers in one study have identified some potential problems with the test as a means of gauging self-awareness in young children and animals.

Proposing that a self-recognising child or animal may not demonstrate mark-directed behaviour because they are not motivated to clean up their faces, thus providing incorrect results, the study compared results of the standard rouge test methodology against a modified version of the test.

In the classic test, the experimenter first played with the children, making sure that they looked in the mirror at least three times. Then, the rouge test was performed using a dot of rouge below the child's right eye. For their modified testing, the experimenter introduced a doll with a rouge spot under its eye and asked the child to help clean the doll. The experimenter would ask up to three times before cleaning the doll themselves. The doll was then put away, and the mirror test performed using a rouge dot on the child's face. These modifications were shown to increase the number of self-recognisers.

The results uncovered by this study at least suggest some issues with the classic mirror test; primarily, that it assumes that children will recognise the dot of rouge as abnormal and attempt to examine or remove it. The classic test may have produced false negatives, because the child's recognition of the dot did not lead to them cleaning it. In their modified test, in which the doll was cleaned first, they found a stronger relationship between cleaning the doll's face and the child cleaning its own face. The demonstration with the doll, postulated to demonstrate to the children what to do, may lead to more reliable confirmation of self-recognition.

On a more general level, it remains debatable whether recognition of one's mirror image implies self-awareness. Likewise, the converse may also be false—one may hold self-awareness, but not present a positive result in a mirror test.

See also

• Animal consciousness
• Cognitive tests
• Narcissus (mythology)
• Self-agency

External links

• List of animals who passed the mirror test and promising candidates on animalcognition.org
• The World First Self-Aware Robot and the Success of Mirror Image Cognition (Lecture at the Karlsruhe University and the Munich University, Germany), 8 November 2006.
• Elephants pass mirror test of self-awareness (The Guardian)
• Elephants' jumbo mirror ability (BBC News)
• Elephant study published in Proceedings of National Academy of Sciences, USA
• Elephants see themselves in the mirror (Newscientist.com with video)
• Can a robot pass the mirror test? – Raúl Arrabales Moreno, 2010-01-08