<p>The mirror test is among the most famous experiments in animal psychology. One can learn a lot simply by observing animals in front of mirrors. The test is used to see if animals can recognize themselves when they look in a mirror. What’s such a big deal about an animal seeing itself in the mirror? Well, to <em>see</em> yourself - that is, to recognize that it is <em>you</em> in the mirror - you have to be aware that <em>you</em> are an entity. <em>You</em> are an entity with a body, just like those around you.</p>
<p>This might seem trivial, but I don’t think it’s necessarily true that many animals are aware. If you interviewed a lizard, would it be able to tell you it’s a lizard? My guess is that it could not. Lizards don’t speak English so we can’t interview them directly. But by observing them, I think we get at something fundamental that we would otherwise miss. Despite all of its flaws, I don’t know of a better way to get a sense of an animal’s mind than watching its behavior. And the most interesting single observation I have seen is observing an animal’s behavior in front of a mirror.</p>
<p>The mirror test was devised by Gordon Gallup who placed a red dye on chimpanzees. The dye had no scent and could not be felt, so it could only be detected by sight. He placed it above their eye and on their ear - places they couldn’t see without using the mirror. Tests generally place an invisible dye on the other side of the face to distinguish an animal using its reflection from one inspecting where it was touched.</p>
<p>There are obvious limitations to this test. There’s no doubt that an animal can still be self-aware and still “fail” a mirror test. The main problem is that while seeing some behaviors in a mirror definitely shows self-awareness (e.g. an animal using their reflection to clean their teeth), the reverse is not necessarily true. Animals that do not recognize themselves in a mirror do not, necessarily, lack self-awareness. There are two main limitations to this test that I see:</p>
<ol>
<li>
<p>If an animal recognizes itself but doesn’t display any behavior that we associate with self-recognition, the mirror test will fail.</p>
</li>
<li>
<p>An animal could also be self-aware yet not recognize itself. This is particularly possible if an animal is not a particularly visual animal. In these cases, the test is likely not to work.</p>
</li>
</ol>
<p>However, I want to push back on these limitations. After watching many animals perform this test, I have found that the resulting behaviors fall into relatively distinct groups. For those that recognize themselves, we see a lot of self-investigation. And why not? It makes sense that an animal would take advantage of a rare chance to view itself. For example, chimpanzees like to look in their mouths and at their teeth. Most animals who pass the test seem interested in the mirror when they first realize what they are looking at.</p>
<p>For those that don’t recognize themselves, the behaviors usually fall into either fighting/intimidating, playing, or ignoring. Although it’s hard to glean much from an animal ignoring a mirror, I think we can make some determinations from the other two behaviors. I remember living in a house where cardinals would see their reflections on a specific window and attack it. Sometimes a cardinal would do this every evening for over a week. To me, this is pretty clear evidence that these cardinals never realized they were looking at a reflection of themselves. I don’t think it’s likely that an animal could recognize itself yet still be scared of or attempt to intimidate the reflection. To me, it’s pretty clear those animals are not aware that it <em>is</em> a reflection.</p>
<p>I will admit it’s possible to draw too many conclusions from these tests, especially due to our natural tendency to anthropomorphize animals. However, I fear that if we’re too skeptical of our ability to interpret these results there is a risk of ignoring a great deal of insight here. As <a href="https://www.nature.com/articles/419255a">researcher Marc Bekoff said</a>, “We must be careful neither to imbue animals with unknown cognitive capacities nor to rob them of skills that they might possess.”</p>
<p>The mirror test is easy to criticize, but as far as tests of self-awareness go, I think it’s the best we have.</p>
<p>I am on the lookout for animals using their reflection from water or ice as a natural mirror, but I haven’t seen that yet.</p>
<h3 id="jaguars">Jaguars</h3>
<p><img src="https://github.com/jss367/antools/blob/gh-pages-2.3.4/assets/images/mirrors/jaguar_mirror.gif?raw=true" /></p>
<p><img src="https://github.com/jss367/antools/blob/gh-pages-2.3.4/assets/images/mirrors/jaguar2.gif?raw=true" /></p>
<h3 id="pumas">Pumas</h3>
<p><img src="https://github.com/jss367/antools/blob/gh-pages-2.3.4/assets/images/mirrors/puma.gif?raw=true" /></p>
<h3 id="asian-elephants">Asian Elephants</h3>
<p>One in 3 has passed based on strict criteria. However, the others seem to display other characteristics. It’s going to look different in different animals, and sometimes you just know it when you see it. An animal using a mirror to clean its teeth is definitely self-aware, although that certainly shouldn’t be a precondition for concluding an animal is self-aware.</p>
<h3 id="pigs">Pigs</h3>
<p>I don’t think they pass, although they recognize what a mirror is.</p>
<p>Animals <em>using</em> the mirror, as in gaining information from their reflection, <em>is</em> interesting. Basically, there was food visible in the mirror but hidden from direct view, and the question was could the pig use that information to get the food. You can <a href="https://www.youtube.com/watch?v=mqD3FkDgGYk">see the same setup in this video</a>. But it’s not the same as the mirror test, in which we hope to learn how or of the animal sees themselves in the world.
It certainly means something. In that article, the authors state that “To use information from a mirror and find a food bowl, each pig must have observed features of its surroundings, remembered these and its own actions, deduced relationships among observed and remembered features and acted accordingly.” These are all interesting things. They’re not the same as recognizing oneself, but we’re looking at a spectrum of ideas and this doubtlessly helps us put pigs on that spectrum.</p>
<p>In the article <a href="https://www.sciencedirect.com/science/article/abs/pii/S0003347209003571?via%3Dihub">Pigs learn what a mirror image represents and use it to obtain information</a> they certainly discuss the pig <em>using</em> the mirror, but not necessarily recognizing themselves.</p>
<h3 id="pandas">Pandas</h3>
<p>Out of 34 tested, none passed</p>
<h3 id="humans">Humans</h3>
<p>Humans begin passing at around 18 months. Not all humans pass though (many severely autistic people do not).</p>
<h3 id="bonobos">Bonobos</h3>
<p>Bonobos have passed the test.</p>
<h3 id="orangutans">Orangutans</h3>
<p>Bornean orangutans have passed. I’m not sure about other orangutan species, although I would guess they would.</p>
<h3 id="chimpanzees">Chimpanzees</h3>
<p>Chimps usually start out by thinking it’s a different chimp, then they figure it out.</p>
<p>Chimps pick food out of their teeth, like to look at their mouths.</p>
<p>Chimps are known to be able to see themselves in mirrors, and you can see that happening here</p>
<p><img src="https://github.com/jss367/antools/blob/gh-pages-2.3.4/assets/images/mirrors/chimps.gif?raw=true" /></p>
<p>The youngest chimp I’m aware of to show self-awareness was Megan, at 3 years and 3 months old (Self-Recognition in Chimpanzees (Pan troglodytes): Distribution,
Ontogeny, and Patterns of Emergence)</p>
<p>Here’s another video of chimpanzees:
https://www.youtube.com/watch?v=OsoNKlyFtpI</p>
<p>The official method is to place a dot on the animal and see if it moves to inspect the dot. That’s probably a good approach but it’s not the only option. Sometimes the effect is so obvious that we don’t need to rely on the dot. For example, this chimpanzee is clearly looking in the mirror and using it to look at its teeth.</p>
<h3 id="gorillas">Gorillas</h3>
<p>I think the results are inconclusive. Some results have suggested they do not pass, but I think others suggest they do. Many people would say they do not, but I think this video suggests that they do. It only takes one. Gorillas behave very aggressively so are often anesthetized during the experiment. It would be interesting to see what happens with daily exposure to a mirror over a long period of time.</p>
<p>Gorillas are generally considered not to have passed the mirror test. Sometimes they have become violent so are often anesthetized before the test.</p>
<p><img src="https://github.com/jss367/antools/blob/gh-pages-2.3.4/assets/images/mirrors/gorilla.gif?raw=true" /></p>
<p><img src="https://github.com/jss367/antools/blob/gh-pages-2.3.4/assets/images/mirrors/primates.gif?raw=true" /></p>
<p>I think this one is recognizing himself: https://www.youtube.com/watch?v=-cMyBS-87Lk</p>
<p>He’s certainly not behaving like it’s an intruder. Here’s an example of that:</p>
<p>https://www.youtube.com/watch?v=_xMd6pyyMbc</p>
<h3 id="macaques">Macaques</h3>
<p>Three species have been tested, none have passed</p>
<h3 id="baboons">Baboons</h3>
<p>Hamadryas baboons have been tested and not passed.</p>
<p>Based on what I’ve seen with chimpanzees, I believe the negative result to be significant for baboons. They seem roughly equally visual as chimpanzees (the bright hindquarters suggest this), but do not display any of the same behavior. Maybe these findings <em>do</em> point to a lack of self-recognition in baboons.</p>
<p>Here’s a baboon:</p>
<p><img src="https://github.com/jss367/antools/blob/gh-pages-2.3.4/assets/images/mirrors/baboon_mirror.gif?raw=true" /></p>
<p>Baboons are monkeys, not apes. It seems like, in general, most monkeys I have seen do not pass.</p>
<h3 id="other-monkeys">Other Monkeys</h3>
<p>Some monkeys are believed to pass after an extended time in front of the mirror.</p>
<h3 id="dolphins">Dolphins</h3>
<p>Bottlenose dolphins are believed to pass the test. I think this shows that the test works for quite a broad range of animals.</p>
<h3 id="orcas">Orcas</h3>
<p>Orcas (killer whales) also pass the test, again showing the test’s applicability.</p>
<h3 id="cleaner-wrasse">Cleaner wrasse</h3>
<p>The cleaner wrasse is the only fish that researchers have claimed to have passed the test. I haven’t been able to get a video of this yet. If true, would this make them the animal that can pass the test at the youngest age?</p>
<h3 id="manta-rays">Manta rays</h3>
<p>I’ve watched videos of manta rays in front of mirrors and it’s very unclear to me what to learn from it. Some (who spend a lot more time with manta rays than I do) claim they have passed.</p>
<h3 id="eurasian-magpie">Eurasian magpie</h3>
<p>Eurasian magpies are known to be very smart birds, so it’s no surprise that they were the first bird to pass the test (in 2008). Interestingly, many other birds thought to be very intelligent have not passed.</p>
<h3 id="grey-parrots">Grey parrots</h3>
<p>Have not passed</p>
<h3 id="new-caledonian-crow">New Caledonian crow</h3>
<p>Has not passed</p>
<p>Have been tested in front of mirrors though. They were able to use mirrors to find food, but it’s not conclusive. The result was statistically significant, but not absolute. It’s maybe that they were seeing food in the mirror and searching for food near there, thereby finding the food reflected in the mirror more often, without realizing they were seeing a reflection.
Notably, New Caledonian crows always react aggressively.</p>
<h3 id="pigeons">Pigeons</h3>
<p>Apparently, pigeons can pass with training. Very interesting.</p>
<h3 id="octopuses">Octopuses</h3>
<p>No octopus has passed the mirror test, although it has been attempted. This does seem to me to be a possible situation where we might not recognize their “self-recognition” behavior.</p>
<h3 id="weasel">Weasel</h3>
<p><img src="https://github.com/jss367/antools/blob/gh-pages-2.3.4/assets/images/mirrors/weasel.gif?raw=true" /></p>
<h3 id="unknown-small-mammal">Unknown Small Mammal</h3>
<p><img src="https://github.com/jss367/antools/blob/gh-pages-2.3.4/assets/images/mirrors/little_mammal.gif?raw=true" /></p>
<h3 id="ants">Ants</h3>
<p>Whether ants have ever passed the mirror test remains disputable in my opinion. In the paper <a href="https://www.scinapse.io/papers/2180773430">Are ants (Hymenoptera, Formicidae) capable of self recognition?</a> Two researchers from Brussels claim they passed, although they didn’t provide any video evidence. I can’t find any other research that has attempted to reproduce it.</p>
<p>You can read the whole thing <a href="https://www.udocz.com/read/22623/are-ants-capable-of-self-recognition-pdf">here</a>.</p>
<p>The author is still an active researcher: https://ebe.ulb.ac.be/ebe/People.html</p>
<p>Some commentary on it here: http://www.animalcognition.org/2015/04/15/list-of-animals-that-have-passed-the-mirror-test/</p>
Mirrors was originally published by at Animals Using Tools.