Hacker News Reader

Comment by ars | Hacker News Reader

The article does not support the HN title.

I think it does. The study authors present evidence that fish, as measured by a stress hyperthermia response, respond in a manner consistent with an emotional response.

As a recanted neurobiologist, I find that surprising and I'm willing to go along with the headline. It would be even more surprising (to me, at least) to find that something as fundamental as emotional response were not to be found in more primitive branches of the phylogenetic tree.

I don't buy it. If I attach a switch to an incandescent light bulb, then the bulb will get hot if I press the switch. If I press the switch enough, the light bulb will burn out. But it would not be accurate to say that the light bulb "died of stress" or that this was evidence that the light bulb could "feel" anything. It certainly doesn't prove the light bulb is conscious.

Humans try so hard to believe consciousnesse is boolean, like "humans have it" and "every other living thing does not" is almost funny. I have news for nerds,, conciseness is a float, imprecition and all.

So cannibalism and eating dolphins is way more closer than the average human is comfortable with, so people like to err on the lest uncomfortable side.

While true that doesn't progress the discussion. The question is whether this evidence suggests the floating point number assigned to fish should be higher. That a lightbulb at ~0.0 consciousness could perform similarly is a good reason to give little weight to this study. What's the counterargument?

Is there no C line in the animal kingdom? If so we're saying spiders and cockroaches are conscious? Surely there's a distinction, and nobody really doubts elephants, dolphins and great apes, but we can clearly see the difference in quality of C. So if there is some kind of line or distinction within the animal kingdom, where would it be? Are insects the end of mechanical behavior? Or could that line extend to small fish and crustaceans? I think the article is using the word "feeling" a little too freely here. By the same measure, There could be an experiment suggesting cockroaches get stressed, and if so, would we be so quick to say they have feelings?

The problem is that consciousness is such a loaded concept. It can mean higher-order thought, self-awareness, ability to feel emotions, the phenomenon of subjective experience itself, and more. I find that most conversations about it involve lots of talking past each other.

They have not demonstrated that stress hyperthermia = feelings.

Just that stress = hyperthermia in a fish.

It's a result, but it's not the result in the HN title. Which is not in the article BTW.

But according to the article stress hyperthemia is one trait used to identify consciousness. Is consciousness the same as "having feelings"?

>One particular trait used to identify consciousness, which was previously thought to be absent in fish, is the capacity for stress-induced hyperthermia or “emotional fever”. This is a physical reaction similar to a fever caused by infection, but in this case the trigger is a stressful situation. Basically, the body gets warmer in response to stress.

I googled stress-induced hyperthermia and got very little.

That would imply that if it's a marker for consciousness it's not a well accepted one. Maybe it's one marker among many, but certainly not enough to support the HN title "Fish have feeling".

And simply on a logical level I'm not seeing the connection between stress causing heat, and consciousness.

Animals obviously can get stressed, and animals obviously can have a fever. I'm not seeing why consciousness is necessary to link the two.

Do it using scholar.google.com and you will get quite a bit of primary research regarding it.

I did. I found lots of research, very little of it having anything to do with consciousness (or the lack thereof).

Actually, it does. Unless you mean that it isn't surprising that there is evidence that fish have feelings.

There is no evidence that fish have feelings. There is evidence that fish increases its temperature in response to stress.

Which is, in the literature about animals and feelings, considered evidence that an animal may have feelings. Did we read the same article? It says quite plainly that this response is considered evidence of feelings in animals, and the supposed lack of this response in fish was evidence that fish did not have feelings.

So, if the lack of response (in previous measurements) was considered evidence against fish having feelings, why would this better experiment, which did observe the response in fish, not be considered evidence of fish having feelings?

Because "feelings" is undefined. If the only evidence of having/not having "feelings" in fish is the increase of temperature, why not simply talk about the increase of temperature, without inventing "feelings"? Not inventing non-material objects unless something is impossible to explain without them is the basis of science, it is called Occam's razor.

My server also increases temperature under high load. Should I conjecture that it has feelings?

It is one of many aspects of the fish mind mentioned in the article; things like developing memories, comparing oneself to other fish, exhibiting preferences, etc. Each of them alone may not be enough to argue "fish have feelings" or "fish can think" or "fish are sentient beings", but they are evidence of those things. They don't make it so, but they hint at something that it is hard to figure out without a number of data points.

Occam's razor can just as easily be applied here to come to a conclusion that is opposite your conclusion.

Fish do things that mammals (humans) do. We know that mammals do those things (stress fever) because they are experiencing things in a way at least somewhat comparable to the way we experience them (they are experiencing discomfort and fear and their body is responding to it, just as ours do). What is the simpler explanation here?

1. Your conjecture that fish are utterly and completely different from other living things that exhibit this response and that they are exhibiting this response for reasons unlike our reasons (i.e. they aren't experiencing discomfort and fear, but are still heating up like we do).

2. They're more like other sentient beings than we'd previously assumed, and are experiencing things in some ways similarly.

In short: If you're coming at this with the strongly held belief that fish are impossibly different from humans and can't possibly have feelings, you would interpret your way. But, if you're coming at it with no strongly held beliefs on the subject, and you see a fish behave in ways that indicate they are sentient and experience suffering like other animals...well, you assume the responses and behaviors are because they are kinda like those other animals.

Your assumption (that this response has no relation to the similar response in mammals) requires a new explanation for why the response happens. Thus, it seems to not be the simplest explanation (to me).

> Occam's razor can just as easily be applied here to come to a conclusion that is opposite your conclusion.

I don't conclude that fish don't have feelings and human have. What I say is that "feelings" and "experiencing" are undefined terms. So, both statements "fish have feelings" and "fish don't have feelings" are meaningless.

> Your assumption (that this response has no relation to the similar response in mammals)requires a new explanation for why the response happens.

I didn't assume that.

"Experiencing fear" is not an explanation for heating. The real explanation requires less abstraction, modelling organism as close to physics as possible, explaining biological processes that happen inside, not just comparing fish to mammals in a way that is more abstract than biology itself.

> Thus, it seems to not be the simplest explanation (to me).

Occam's razor is not about the simplest explanation. The simplest explanation for everything is "god will". Just as inventing "god" is simpler, inventing "fear", "feelings", "mind" and other non-material things is simpler, but unnecessary and does not explain anything really.

So, humans don't have feelings either, because it's simpler to explain it as a mechanical process?

See, I never said that fish does not have feelings or that fish have feelings. I think it is meaningless.

"Feelings" is an abstraction that makes it easier to explain human behaviour. With fish, asking questions like "does it have feelings", "does it have free will", "is it conscious" etc. is just the same as asking the same questions about machines: it does not help you to make any decisions about how to interact with it, so it does not matter.

So, to extend the question I asked before: Does that mean because we cannot ever really know what goes on in another human's mind, does that mean human "feelings", "free will", or "consciousness", are meaningless and does not help us make decisions about how to interact with other human beings?

The HN title is the one the BBC had used in linking to the article from their homepage; it made sense to me and so I went with it for the HN post (I didn't actually notice until afterwards that the title in the article itself was different). Hope that helps clear up the reasoning behind it.

Reply on news.ycombinator.com