Hacker News new | past | comments | ask | show | jobs | submit login

I suspect that after 400 years of the scientific method, that we may be reaching the limits of single variable experiments in a number of fields. Statistical methods can find those patterns, and as we advance in those areas I expect us to advance in messy sciences like psychology. We’ll be able to more reliably look at people or other chaotic systems and see how three inputs work together to create a single effect.





The math for multivariate experiments has been well understood and applied since the early 20th century.

Modern industry would not exist without it.

The problem with psychology experiments is that the mind has many hidden variables which cannot be easily accounted for.


Moreover hard science is more or less based on at least three things: objectivity, quantifiability, and empiricism. We suppose that objective phenomena exist and can be independently verified and confirmed by other observers, using quantifiable measures to agree upon exactly what it is we are observing when conducting an experiment.

The issue is that the mind is by nature subjective and wholly private, inaccessible to outside observers. Further, whereas when conducting an experiment in physics where it becomes possible to separate the experimenter from the experiment (to an extent, observer effect and quantum mechanics notwithstanding), as regards matters of the mind the observer and the observed are one and the same of necessity; you are in fact part of the experiment yourself.

> The problem with psychology experiments is that the mind has many hidden variables which cannot be easily accounted for.

This is where the pillar of quantifiability breaks down, and barring advances in techniques for inspecting the brain with greater spatial and temporal resolution it's hard to see how one can quantify what cannot be directly observed.

Empiricism is the only aspect of the three aforementioned that can still hold up within this domain, and there is a large centuries' long tradition of studying the mind subjectively, qualitatively (as opposed to quantitatively), and yet in a still empirical fashion with falsifiable hypotheses; various mystical and meditative practices exist in this space, such as the fire kasina practice, especially as formulated by a trauma physician [0] complete with falsifiable hypotheses and steps for independent reproduction. What distinguishes practices in this space from harder sciences is that the phenomena observed are subjective, not "out in the physical world," and there does not exist any instrumentation for measuring them apart from that of your own faculties of perception.

As an aside I found the comparison to shamans and spiritual teachers in TFA interesting, since I have always considered psychology and spirituality to concern themselves with the same subject matter and problem domain of minds; one could say that they are both proto-sciences of mind, and psychology is the latest iteration of this tradition (though as TFA states somewhat misguided and out of touch with its roots in mysticism; on this latter point see Jung's Psychology and Alchemy for a discussion of where analytical psychology connects with the older esoteric traditions).

[0] https://firekasina.org/wp-content/uploads/2017/11/the-fire-k...


This is a misunderstanding. The the mind is not wholly private and subjective, its objectively quantifiable. Dualism ended a long time ago, and psychology is a materialistic science.

> The issue is that the mind is by nature subjective and wholly private, inaccessible to outside observers.

Not by nature, it's not. Unless you define it to have some immeasurable spiritual quality to it which is obviously not experimentally discoverable and so of little use discussing here.

> This is where the pillar of quantifiability breaks down, and barring advances in techniques for inspecting the brain with greater spatial and temporal resolution it's hard to see how one can quantify what cannot be directly observed.

Pretty much every experiment has a massive amount of relevant states which we cannot quantify.

There's a whole lot of quarks in 1 kg of steel but we don't need to know all of their states to measure macro quantities like its temperature and strength.

The mind has proven very reluctant to this sort of useful and measurable macro properties.

It's a typical case of a chaotic system. Perhaps innovations in the modeling of such complex systems (not too different from the advancements we're seeing in ML) will be the key to better insights into the mind.


> Not by nature, it's not. Unless you define it to have some immeasurable spiritual quality to it which is obviously not experimentally discoverable and so of little use discussing here.

There is no need to bring in "spirituality" (at least, as commonly and colloquially understood) or "souls" into the picture - see for instance qualia, whose existence is self-evident, and yet are also not amenable to examination by external observers. Irrespective of how much you can pick and probe at the brain and measure wavelengths of light, there is "something it is like" to actually experience the phenomenon of "red", and this experience itself is not readily accessible to objective/quantitative methods.

> It's a typical case of a chaotic system. Perhaps innovations in the modeling of such complex systems (not too different from the advancements we're seeing in ML) will be the key to better insights into the mind.

I could be convinced that minds are ultimately emergent phenomena of plain physical and mechanical processes which are too frighteningly complicated for us to analyze with contemporary methods.


> is not readily accessible to objective/quantitative methods.

Just because the mind is not readily quantifiable with current technology doesn't mean that it's subjective "by nature".

> for instance qualia, whose existence is self-evident,

No, it isn't and I see no way to test for "qualia" so, applying Newton's flaming laser sword, it's not worthy of debate.

It's a very dangerous thing to draw conclusions about empirical phenomenon from metaphysics. I suggest you stick to the scientific method when trying to understand the empirical world, it has been far more successful than philosophical rambling.


> No, it isn't and I see no way to test for "qualia"

You are verifying the existence of qualia every instant of your existence. It's the most immediately apparent empirical fact conceivable, since it is sensory experience itself, and you are testing its presence by the mere fact of being alive (philosophical zombies [0] notwithstanding).

> Before she left her room, she only knew the objective, physical basis of those subjective qualities, their causes and effects, and various relations of similarity and difference. She had no knowledge of the subjective qualities in themselves. [0]

I suggest you confirm the definitions and senses of terms you criticize before being so dismissive of them.

[0] https://plato.stanford.edu/entries/qualia/#Irreducible


The experience of qualia doesn't define any test. By what metrics would we even test it against? Are qualia orderable in any way? Do they have weight, a size, or substance? Do processes of material 'exist' physically in any meaningful sense?

We have not devised a test for others' qualia, and the only evidence we have of qualia is our own experience, which is a model of objective reality and not reality itself.

Bringing up philosophical zombies: we can't know if everyone inhabits their own universe, independent of each other, filled with zombie replicas. From anyone's perspective: that scenario would be 100% indistinguishable from what Occom's razor suggests. It therefore follows the qualia do not exert any sort of physical presence.

There are nonphysical things that we consider to exist. Numbers being the prime example (har de har har). Numbers do not physically exist, but are a property we impose on various groups (of which we distinguish in our own mind). Similar processes like experience do not 'physically exist' but is a property of physical existence (of which, we politely assume of other people to possess, and not call them zombies).


This may be true, but what does it have to with psychology as a science? How psychology deals with meta cognition (thinking about thinking) is correlate it with objective measures related to the phenomenon, (performance accuracy, changes in BOLD response in task fMRI. We don't care if its qualia, or just reports of experience. that is a different debate and field entirely.

There's a feedback loop between technology and science. Without progress in science there can't be progress in technology. But slso without progress in technology there can't be progress in science.

In applied science they call this the ladder. It’s the same reason some people are desperate to keep certain sorts of manufacturing in country. Once you lose that mutualism you can’t just spin up a new factory. You have to do what every second world country on the road to being a first world manufacturing superpower has done: you make cheap shit until you can make mediocre products until you can make good ones until you can make high end items.



Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: