[wyager]

Our perception of reality absolutely agrees with QFT -- in the medium-energy medium-scale environments that humans operate in. Everyday physics is a limiting case of QFT.

Our mental heuristics don't work for very high/low energies or small/large scales because we don't operate on those scales.

We had no reason to evolve other heuristics.

[Noseshine]

I think I understand your point - but I had a very different angle. What you write actually is fully in support with my point.

You simply declare as "reality" that what you perceive, the model we (our brains) have come up with. But that's exactly it - that's your version of "reality" created by how you look at it - with "macro" sensors. Different sensors would have a different result. In addition we don't even have raw sensor data in our mental world model, instead the data is extremely processed before it is assembled to our mental model of the world. Take a neuroscience course on perception, I recommend. Somebody said we live in a VR world created by our brains, and that is a good fit. Of course the input that goes into creating that "VR world" is from sensors getting data from interactions with the actual world.

If you don't understand what my point is don't say "you are wrong".

[pixl97]

>But that's exactly it - that's your version of "reality" created by how you look at it

https://en.wikipedia.org/wiki/Reality_tunnel

>In addition we don't even have raw sensor data

Of course that can be turned around to point out that our senors (for machines) only provide us with very filtered information. It just seems to be 'raw' compared to the information available to our conscious minds.

Intelligence isn't the ability to gather massive sums of information, but instead to filter it.

There is simply too much information around you. As a biological creature with a limited energy budget and a limited lifespan this is a big problem. As we know from computing, two things can save massively on your energy requirements. Good algorithms and early pruning of unneeded information. For example interpretation of UV and IR wavelengths has been unimportant for the continued survival of our species, and for most of the human timeline it has been abstracted away. We knew something existed because we were affected by it, but without science we created things like magic and gods to explain it. How right or wrong our mental model (subjective reality) is really doesn't matter much until we reach a critical junction of events were the fitness algorithm kicks in, survival of the fittest as Darwin coined.

We have come to the point of evolution where we realize 'reality' is much larger than our set of filters and learning algorithms. We also realize our algorithms are not well adapted for the massive increase of knowledge and rate of change we are experiencing.

[visarga]

> Of course the input that goes into creating that "VR world" is from sensors getting data from interactions with the actual world

This process might not be so arbitrary, for example in an experiment at Google, an unsupervised machine learning system discover the concept of cat from raw images all by itself. The concept of cat might be present in the world objectively and we just gave it a name.

[amatic]

It discovered a statistical measure of features that happen to occur on cat images, on the pixels of those images. It might be something like brains do, it might not.

[Karnickel]

    > The concept of cat might be present in the world objectively

It exists in a world with sensors similar to that of humans, and hardware and software made by humans mirroring their world experience. You are still within the exact same (world) box.

[wyager]

The reality constructed in our brains is not substantially different from actual reality. Our mental model of reality is just much lower fidelity. At no point has physics demonstrated "wow, humans are totally wrong about the world".

>Different sensors would have a different result

All of our experience with robotics suggests otherwise.

[Noseshine]

    > The reality constructed in our brains is not substantially different from actual reality.

Now you're being facetious. Good joke though.

    > > Different sensors would have a different result
    > All of our experience with robotics suggests otherwise.

I meant DIFFERENT sensors, not the same version of the same sensor. You are telling me a sensor for microwaves gets the same result as one for different wavelengths? Good god. And that's actually related sensors, then there's touch, smell etc.