Is scientific research hitting a wall?

[lpetrich]

Is Science Hitting a Wall?, Part 1 - Scientific American Blog Network Is Science Hitting a Wall?, Part 2 - Scientific American Blog Network

From the first one,

The Session was inspired in part by research suggesting that scientific progress is stagnating. In “Are Ideas Getting Harder to Find?”, four economists claim that “a wide range of evidence from various industries, products, and firms show that research effort is rising substantially while research productivity is declining sharply.” The economists are Nicholas Bloom, Charles Jones and Michael Webb of Stanford and John Van Reenen of MIT.

...
These findings corroborate analyses presented by economists Robert Gordon in The Rise and Fall of American Growth and Tyler Cowen in The Great Stagnation. Bloom, Jones, Webb and Van Reenen also cite “The Burden of Knowledge and the ‘Death of the Renaissance Man’: Is Innovation Getting Harder?”, a 2009 paper by Benjamin Jones. He presents evidence that would-be innovators require more training and specialization to reach the frontier of a given field. Research teams are also getting bigger, and the number of patents per researcher has declined.

The economists are concerned primarily with what I would call applied science, ... But their findings resonate with my claim in The End of Science that “pure” science—the effort simply to understand rather than manipulate nature--is bumping into limits.

It may depend on what counts as a breakthrough. Could many breakthroughs be evident as breakthroughs well in retrospect?

Also, as author John Horgan notes, it could simply be that the easier stuff has already been done. Improved technology likely mitigates this problem, but apparently not enough.

Some scientists at The Session scoffed at the idea of a scientific slowdown. Biologists, pointing to CRISPR, optogenetics and other advances, were adamant that the pace of discovery is, if anything, accelerating. My response: Yes, fields like genetics and neuroscience are indeed churning out findings, but to what end? Gene therapy has been an enormous disappointment, and treatments for mental illness remain appallingly primitive.

I think that those biologists are right. There is a big biological mystery that has yet to be solved: how organism development works. We have such tantalizing hints as Hox genes, but not much more.


The second one discusses how drug development has become more difficult:

Eroom’s Law. The paper notes that “the number of new drugs approved per billion U.S. dollars spent on R&D has halved roughly every 9 years since 1950.”

The better than the Beatles problem.
The cautious regulator problem.
The throw money at it tendency.
The basic-research-brute force bias.
The devil is in the details problem.

John Horgan mentions a solution: "Appoint Dead Drugs Officers." To see what went wrong in a drug-development failure.

In theoretical science, one could appoint a "Dead Ideas Officer", one that could come after the likes of Freudian psychoanalysis and string theory.

 

[Politesse]

Are we sure it is a dearth of research hypotheses which is at fault, rather than inefficiencies in the actual production and publication of scientific research? As an economic fact, scientific careers are much less supported than they used to be when whimsical ideas frequently turned into brilliant new technologies. Success requires more than just effort - much effort in an inefficient system can be the same in outcome to lesser effort in a productive one.

 

[skepticalbip]

It is rather humorous that this "analysis" is done by economists rather than scientists or even science historians. What time length are they considering developments over, the last fiscal year?

 

[barbos]

It's a side effect of of explosive growth of knowledge during 20 century. If you compare what Einstein knew after finishing his university and what todays students supposed to know there is no comparison. His education which included pretty much all the physics known so far is roughly equal to first 1-2 years of modern physics major which is petty much nothing.
He did most of his famous stuff before his 30. Nowadays you really start produce anything interesting after you are 30.

 

[barbos]

It is rather humorous that this "analysis" is done by economists rather than scientists or even science historians. What time length are they considering developments over, the last fiscal year?

Well, they confined themselves to medical research which is driven by economy laws, certainly more so than more fundamental stuff.

 

[Keith&Co.]

four economists claim that “a wide range of evidence from various industries, products, and firms show that research effort is rising substantially while research productivity is declining sharply

Well, my first thought is that industrial research is funded on a pretty explicit expectation of results. They do not fund advancing the boundaries of knowledge just to know.
That's how you get those news articles mocking someone who spent $x bazillion on finding out if butterflies like jazz.

So I wonder if research within certain settings is hampered, not by a limitation on science, but by the researchers only being rewarded for advancing their company's industry, and choosing research accordingly. So instead of finding out the next new mystery, they spend time finding a new way to arrange things they already know. Shrinking a product by another 2% to fit ten more units on the shelf space at Best Buy.

 

[Jimmy Higgins]

My understanding is quite the opposite, that there is just too damn much coming out. It makes it extremely hard to absorb even a tiny fraction of it, within a field.

 

[barbos]

My understanding is quite the opposite, that there is just too damn much coming out. It makes it extremely hard to absorb even a tiny fraction of it, within a field.

Quantity went up, but quality overall went down it seems.

 

[steve_bank]

If there is nothing below particles, fields, and electromagnetics, yes.

Our instruments are limited by the quantization of the electron, in general.

Cosmology is limited by our ability to detect photons for example.

 

[untermensche]

So-called "neurosciences" in general are a joke.

They have no working model for the phenomena of consciousness.

When they look at brain activity they have no idea what parts of it are creating consciousness.

They know that many areas are involved yet have no idea how the phenomena we have intimate contact with on a daily basis arises.

The idea that it is some effect of electricity or chemistry has yielded nothing.

No model = no understanding

 

[steve_bank]

So-called "neurosciences" in general are a joke.

They have no working model for the phenomena of consciousness.

When they look at brain activity they have no idea what parts of it are creating consciousness.

They know that many areas are involved yet have no idea how the phenomena we have intimate contact with on a daily basis arises.

The idea that it is some effect of electricity or chemistry has yielded nothing.

No model = no understanding

If you are looking for a satisfying philosophical mystical emotional explanation of ourselves then science will never satisfy you.

Software neural nets that can be taught or learn have been around for decades. There are commercial products based on neural ners.


The obvious comparison is a computer with software that becomes self aware or 'conscious' realizing it is comprised of electronic circuits. And then rationalizes alternate explanations of its existence.

I must be more than a bunch of transistors....and so on.

 

[untermensche]

So-called "neurosciences" in general are a joke.

They have no working model for the phenomena of consciousness.

When they look at brain activity they have no idea what parts of it are creating consciousness.

They know that many areas are involved yet have no idea how the phenomena we have intimate contact with on a daily basis arises.

The idea that it is some effect of electricity or chemistry has yielded nothing.

No model = no understanding

If you are looking for a satisfying philosophical mystical emotional explanation of ourselves then science will never satisfy you.

Software neural nets that can be taught or learn have been around for decades. There are commercial products based on neural ners.


The obvious comparison is a computer with software that becomes self aware or 'conscious' realizing it is comprised of electronic circuits. And then rationalizes alternate explanations of its existence.

I must be more than a bunch of transistors....and so on.

Teaching some man-made thing to "learn" does not demonstrate anything about the human mind except it can construct things that can learn. And calling something a "neural net" is just a bunch of nonsense. They have no connection to neurons at all.

I want scientific models for phenomena, that is all.

Not a bunch of claims to knowledge.

In science: No model to test = No understanding

 

[steve_bank]

Ethics limits what neuroscince can do in some ways. I am not starting yet another debate on mind. From my experince in electronics and systems it is not hard to see how the brain may work, working out a detailed model that couuld be simulated is anoter matter. I am sure people are working at it. Goedel said he thought an articial analog to the human brain might be constructed and grown as a human would grow.

IMO fundamental scince is at a deadend unless something entirely new is found. Particle physics, fields, electromagnetics is about it .

For us to detect something it has to interact with our reality in such a way trhat we can sense it.

Sensing boils down to a force or voltage or current or detecting photons. If a phenomena can not be reduced to SI fundamental units we can not detect it.

 

[untermensche]

From my experince in electronics and systems it is not hard to see how the brain may work...

Better than you have been working on it for decades.

The 1990's was "The decade of the brain". Massive funding of research took place and has ever since. worldwide.

And we still have no objective model for the phenomena of consciousness.

And certainly have no electrical model for it.

A computer does not experience what it is doing. No matter what it does it is never experiencing anything.

No computer made experiences. They merely respond to programming.

They are not something that will ever be conscious. Despite the stupid efforts to make them behave as if they were conscious.

 

[steve_bank]

From my experince in electronics and systems it is not hard to see how the brain may work...

Better than you have been working on it for decades.

The 1990's was "The decade of the brain". Massive funding of research took place and has ever since. worldwide.

And we still have no objective model for the phenomena of consciousness.

And certainly have no electrical model for it.

A computer does not experience what it is doing. No matter what it does it is never experiencing anything.

No computer made experiences. They merely respond to programming.

They are not something that will ever be conscious. Despite the stupid efforts to make them behave as if they were conscious.

HeeHee. What is experince, pain, and pleasure? And away we go.

Genetic codes are sequncers of events, AKA a program.

The brain 'nodes' are essentially logic gates. A Mealy or Moore state machine is a good first order model to talk to. From Theory Of Computation logic trees and graphs can not solve all problems. Working memory is required, hence the Turing Machine. Of course generalizations applied to the brain, but something to talk to. That is how complex problem solving and science begins. A simple model that evolves.

The concept of AI in the 80s was all the rage. It was going to be the be all end all of engineering. What resulted was forms of AI embedded in CAD that did made big gains in engineering design, but the big predictions never materialized.

I have never heard anyone in scince claim scince will see anf know all. Are you one of those anti-science ranters?

It's been what, 50 or 60 years since the double helix. Genetics has come a long way.

AFAIK medical science does not have a complete working simulation of the biochemistry of the bidy. So what?

 

[untermensche]

You are waving your arms.

Not making arguments.

You have no objective understanding of consciousness or how it is produced.

There is no evidence. None. That consciousness is an electrical phenomena.

Cells respond to electrical current. That is all.

There is no electrical model that can produce consciousness.

 

[Malintent]

So-called "neurosciences" in general are a joke.

They have no working model for the phenomena of consciousness.

When they look at brain activity they have no idea what parts of it are creating consciousness.

They know that many areas are involved yet have no idea how the phenomena we have intimate contact with on a daily basis arises.

The idea that it is some effect of electricity or chemistry has yielded nothing.

No model = no understanding

If that were true, then explain the demonstrated effectiveness of Trans-cranial Magnetic Stimulation (TMS) in Patients that did not respond well to antidepressant medication. It is a non-invasive, non-chemical, FDA-passed, covered by insurance, and there is a growing database of over 6,000 cases as of present.

 

[untermensche]

If that were true, then explain the demonstrated effectiveness of Trans-cranial Magnetic Stimulation (TMS) in Patients that did not respond well to antidepressant medication. It is a non-invasive, non-chemical, FDA-passed, covered by insurance, and there is a growing database of over 6,000 cases as of present.

We use TENS to mediate pain as well. It is just electrical current that stimulates vibratory nerves. The theory is the vibratory nerves move quickly and dampen the slower pain signals.

None of this tells us how the brain creates the sensation of pain.

Pain is more a psychological event than a physiological.

The mind either attenuates the physiology or amplifies it.

This of course only happens in people with minds.

Being able to effect depression with artificial stimulation of the brain is good but it tells us nothing about depression or the consciousness.

Exercise is very good for depression too. That doesn't tell us anything about depression and it's cause either.

 

[ronburgundy]

I think it is largely an issue of the "low hanging fruit" having already been picked. There is no shortage of questions, but every time a question is answered, the new questions it gives rise to are typically harder to answer. In part this is because the new questions are typically about more and more specified mechanisms and mediating and moderating factors. Finding out that X causes Y is relatively easy compared to then having to answer how does X cause Y and under what circumstances, and if the answer is via variable Z, then how does X exactly cause Z and how does Z then cause Y.

Just think about the theory of evolution. The experiments you need to test the rather crudely broad notions of Darwins day are much easier to construct than testing modern questions like what is the specific mechanism responsible for the transfer of information from the gene to phenotypic trait, or what causes some genomes to evolve much more rapidly than others?

 

[Kharakov]

That working model of quale filtration is the one that impressed me the most.

Seriously- when that was published, I was like, finally, I know how pink is separated from the superposition of all possible states, and that's pretty damn cool.