Hacker News Reader

Comment by exo-pla-net | Hacker News Reader

It's not an assumption. Self awareness has been scientifically tested in many species, most prominently by using the mirror test: testing if an animal can recognize itself in a mirror by observing if its behavior changes after surreptitiously placing onto its body a mark, which the animal can only see through the mirror. One can argue about the merits of the mirror test, but it's inaccurate to say that scientists are merely making assumptions. They're doing their best to rigorously test self awareness.

View on HN

The mirror-test is not even close to exhaustive with respect to conclusions about self-awareness. It is certainly conceivable that a self-aware creature could simply be incapable of understanding a mirror or that they might be otherwise indifferent to what they see in it. For example, dogs fail the mirror test, but they are obviously self-aware creatures.

> but they are obviously self-aware creatures.

Is it that obvious? Mine gets scared by her own farts sometimes.

I think it's axiomatically true. If we don't consider dogs to be "self-aware" then I think the discussion becomes an issue of semantics, because dogs clearly possess an emotional intelligence that necessarily implies a kind of "self-awareness" that is common to humans.

I would think that self-aware would imply not just emotions, but having awareness of oneself, not just "of one's body", but like, being able to think about how one would react to something, or at least some things somewhat like that.

Wikipedia says "While consciousness is being aware of one's environment and body and lifestyle, self-awareness is the recognition of that awareness."

Wouldn't that be evidence of NOT being self aware? If the animal is self aware, it knows a fart is coming and where it's coming from, so it should not be surprised.

Yes, I was providing that as a (rather weak) counterexample.

Cats supposedly fail the mirror test, but I think they're just disinterested. There are a lot of videos of cats being shown live video of themselves and their owners, with the owner's face modified to look like a cat. In many of these the startled cat immediately turns around to look at their owner, seemingly indicating an awareness that the video they were being shown contained themselves and that the weird cat-like human creature seen in the video was behind them.

Compilation of such reactions that was posted on HN a fee days ago: https://youtube.com/watch?v=Jto2peSOLac

Here is a cat discovering its own ears in a mirror: https://www.youtube.com/watch?v=akE2Sgg8hI8

Cats grow old too. They have time to learn.

Very late comment, but I find this truly amazing. Yes, some cats are too dumb to fully grasp what's happening, but they do give the human a look of concern. Their reactions could be manipulated, but I find interesting how dogs look forth and back, while most cats remain fixated in the real human. Not very rigorous, but very cool anyway.

I've always thought a good mirror test shouldn't be just a mark, but something that looks like a venomous insect, to stimulate the subject to look for it.

We had a mirror that our dog walked by every day, one day because of a surgery she was wearing a t-shirt to help prevent her from bothering the surgery site. First time she walked by the mirror she startled, realized she was wearing a t-shirt, and walked off.

Similarly when she walked by and we wave she immediately snaps her hear around to look at us.

Certainly seems like she understands mirrors and never treated a dog in the mirror as anything but herself.

They’re doing it within their frame of definitions which is largely evidence of constrained models leaving options unknown outside the realm of reason thereby excluding outliers of surprising potential.

This sounds like critiquing science for not having all answers. Do you have a better approach?

My critique of is shaded by an irritation of within the both the "non-scientific" community that has been indoctrinated to take what "scientists" say as expert factual irrefutable statements and the "scientific" community that has become not what I'd call scientific and more close to dogma, orthodox, ideological, and or protectionist to break novel grounds. The scientific community is explicitly sanctioned by a centralized order of groups that endorse individuals within a process enforced upon anyone born. You either accept the game and fight to get into "science" or you're not a "scientist."

I know plenty of individuals who are more-expert, more scientific than any scientist publishing papers funded by poorly designed incentive structures between education, research, monetary incentive, and survival.

There's a lot of "clique politics" within the academia of science and it deeply troubles me. It's largely based on the university system being in control of it and the processes that contribute to that structure.

Luckily it does seem to be getting more attention so that is good. Additionally the trend towards self-organization with the internet and wherever that leads us will most likely lead to new structures that allow more freedom of information within science to propagate without as much dependence on centralized entities being corrupted and tainting the scientific process.

Outside of that, my critique here is largely based on various modes within Functionalism (theory of mind)

> In an analogous manner, the role of mental states, such as belief and desire, is determined by the functional or causal role that is designated for them within our best scientific psychological theory. If some mental state which is postulated by folk psychology (e.g. hysteria) is determined not to have any fundamental role in cognitive psychological explanation, then that particular state may be considered not to exist . On the other hand, if it turns out that there are states which theoretical cognitive psychology posits as necessary for explanation of human behavior but which are not foreseen by ordinary folk psychological language, then these entities or states exist.

I think the solve I'm looking for is required on both ends to communicate truth and "fuzziness" of truth. I'm not sure if English and current languages will ever trend towards this because every statement has a value of assigned truth based on existence/to-be. So short-term would be communicating the meme of this through the zeitgeist/noosphere and embed it within the culture - then potentially evolve towards a language that is fuzzy based.

This doesn't seem particularly helpful for the conversation here. Claiming something is self-aware because I say so or because I don't trust the state of research will hardly convince anyone. Random social media videos of animals exhibiting interesting behavior is not sufficient either, it is interesting but can be misattributed due to anthropomorphisation.

And if you want to preach about better communicating uncertainty and statistics then I am afraid you're preaching to the choir in a particularly ranty sermon.

Yes there are plenty of uncertainties, but those uncertainties cut both ways.

Well I appreciate the audience for whatever its worth. The exchange added value. Thanks!

If you have any better ideas, the scientific community would legitimately love to hear them.

It’s not like nobody realizes that we don’t have flawless definitions for “self-aware”, “intelligent”, or even “life”. Or that the experiments we devise to test for these things come from our own experiences and bias, and might exclude creatures with novel presentations of these traits.

But they’re the best we have for now. And smart people have spent countless lifetimes trying to improve upon what we have, attempting to broaden these categories and their testing apparatuses without expanding the definitions past the point of usefulness.

So, yes, we get it. Everyone knows our tests and definitions aren’t perfect. We know they’ll exclude life forms with presentations of these traits that we haven’t even conceived of. Not only doesn’t science have all the answers yet, but it hasn’t even agreed on the wording of the questions yet. Pointing this out for the thirteen trillionth time just isn’t useful or insightful.

If it keeps hurting, it should keep being mentioned. When things that hurt are turned into silence, then things don't change that can stop hurt.

I understand your point though, and I'm sorry if it came off as an attack. It's a criticism rooted in a few things. I typed up more here: https://news.ycombinator.com/item?id=24601665

I mean, would you go around reminding your friends of their shortcomings and then reminding them how helpful your feedback is?

Sometimes people are well aware of the problems and are actively trying to solve them.

I won’t remind them of my feedback but I’d never give up on telling them where they should improve.

Mentioning it would be helpful if this was something the relevant scientific community wasn’t already keenly aware of. “We don’t have all-encompassing definitions nor tests of self-awareness” isn’t something people have forgotten— it’s virtually the most well-known open question in the entire field!

It’s like popping into a discussion of general relativity and “helpfully” chiming in with the criticism that “But we know GR must be wrong, since it’s incompatible with quantum mechanics.”

Fair enough - however I haven’t heard the scientific community talk much of this problem. The unfortunate side of gated communities and information.

I will be more mindful of my ranting. Thanks for sharing.

I think it's likely that social animals are almost all self aware. How can an individual animal manage social relations within its pack without a knowledge of self and of other selves?

I disagree. Ants are highly social, but few would ascribe them self awareness. They pursue social objectives using instinct.

But pack animals? Maybe. We don't have an answer to how our own self awareness works, so we're largely working in the dark.

We're definitely working in the dark, but i don't see any compelling reason to suggest that ants aren't self-aware; why wouldn't they be? What about their behavior makes you feel like they aren't?

Extreme self-abnegation.

What are your thoughts on kamikaze pilots in this regard?

Humans are capable of self sacrifice for a cause. Duty, honor, love of one's homeland, spiritual convictions, etc.

These are psychological motives and forces and their factual impact on the subject person is observable.

I just recently saw Eien no Zero (aka The Eternal Zero). If that movie reflects the actual events, then the pilots were subject to serious psychological stress coming to terms with their duty to Japan.

If we can ever measure 'psychological stress' in ants, then we can observe if they experience a shift from a normal psychological state to a stressful one before "deciding" to sacrifice themselves on whatever heroic ant task awaits them.

Are monks, philosophers, and soldiers not self-aware?

A society composed entirely of selfless monks?

(Philosophers are definitely not self-abnegating. They are on a career track. Soldier do exhibit self-sacrifice, but quite a lot are conscripted against their will, and require a bit of re-education in boot-camp.)

The compelling reason for me is the mirror test, and what you can infer from it. The tiniest bit of self-awareness lets you pass the mirror test, as toddlers demonstrate. Apes pass, elephants pass, and magpies pass. Monkeys fail. What's different about monkeys? They have smaller, simpler brains. If a monkey by all appearances has no concept of self, an ant with 0.008% as many neurons assuredly doesn't.

Why does it matter if few (or many, for that matter) ascribe them self awareness? Popularity is no measure of truth.

And who's to say what actions are "free will" (whatever that means) over instincts?

wait a minute, instinct? So hard wired behaviors are different than acquired behaviors when it comes to self awareness? Come to think of it, no one has defined self awareness so how can we begin to say who has it and who does not? Is a three year old self aware? A three year old clearly responds to other people but many would say he/she sees the world as part of him/herself.

The mirror test also gives widely different results in humans. Canadian children pass it at 2 years old, while Kenyans fail at 5: https://www2.psych.ubc.ca/~henrich/pdfs/Journal%20of%20Cross...

Would you conclude from it that Kenyan children are less (or not at all) "self-aware" than Canadian ones? Or that the leap from "recognize reflection" to "self-awareness" is unjustified?

What a minefield of a question

Skimming the paper, the researchers already propose their own interpretation of possible causes, but one thing I couldn't find is whether they controlled for familiarity with mirrors. A full-height mirror (within reach of small children) might simply be more common in western households.

Edit, nevermind, there's a whole paragraph on that towards the end:

> It may be important to note the general use of mirrors in each of the cultures. Although we did not include an independent measure of the extent and nature of mirror experience in these cul-tures, the experimenters visited many homes in these various communities. Although mirrors are present in most non-Western and all Western homes, their use varies considerably. It is more conspicuously and frequently used in the West, compared to any of the other cultures considered here. Western children are more likely to have shared mirror exposure with others or to have seen themselves in a mirror in the presence of others, a situation presumably unfamiliar to the non-Western children we tested. The unfamiliarity with public mirror exposure may be linked to the enhanced “freezing” behavior by non-Western children, particularly when discovering that their face is marked. This inhibition may correspond to the fact that children do not know how to behave, one way or another, in this context.

They also cite a different study:

In contrast, research with humans suggests that children’s relative familiarity with mirrors, which greatly varies across contexts, does not correlate with the age at which the mark test is passed (Priel & deSchonen, 1986). Priel and deShonen (1986) tested Bedouin nomadic children with no previous mirror experience and compared them to same-age Israeli children familiar with mirrors. They found no significant difference in the developmental onset of MSR between mirror familiar and unfamiliar children.

Thinking about it, the idea that familiarity with mirrors would affect the results really throws into question that what the test measures is "self-awareness". It would imply that self-awareness can be increased by exposure to mirrors...

It is a proxy-measure, not what we're actually trying to measure. So having some factors to control for wouldn't be particularly unexpected.

It does not necessarily mean that familiarity with mirrors would cause/improve self-awareness, it could just mean that the test subject would have to learn how to use it or associate the visual stimulus with the concept of self. On the other hand it doesn't exclude the possibility that a mirror can be helpful in learning something related to the self.

Reply on news.ycombinator.com