Your model of the brain as multiple neural nets and a voter is a good and useful simplification. I think we still know relatively little about how accurate it is. You would expect evolution to have optimized the brain to avoid blind spots that threatened survival, and redundancy makes sense as a way to do this.
However, I wouldn't classify blind spots as 'no problem whatsoever'. If the simple model of multiple neural nets and a voter is a good one, then there will be cases where several nets give errors and the conclusion is wrong. Knowing what kinds of errors are produced after what kind of training is critical to understanding when a redundant system will fail. In the end though, I suspect that the brain is quite a bit more complicated that a collection of the neural nets like those this research is working with.
Your last comment is about political control. You seem to mean that our phenomenology will improve as big data mines human behavior and we learn to use that data and those ideas to control others. We do need to be careful who controls that. Although I suspect that the degree of control achievable by such means is easy to overestimate.
Many of us technical types would love for this line of inquiry to be fruitful. But to have a 'physics of people' you have to know the values of all the parameters needed to specify the current state of a person and you need to know all interactions of that person with the rest of the universe. Phrased like that you can see how ludicrous it is to dream of using the methods of physics for social science. Physics works because the fundamental constituents of the universe happen to be only a small number of particles whose interactions are amazingly simple. For example all electrons are exactly identical and interact via only 3 forces (with some uncertainties about effects on scales larger than galaxies and energies higher than trillions of electron volts). The hope for a theory of sociology is a false hope. The hope for a useful phenomenology might be more reasonable and big data can help.
It might have been better if I had phrased (2) a little more strongly like Carroll did: something like 'the physics of everyday life is completely understood' or 'there will be no practical applications of physics beyond our current theories (the standard model and general relativity)'.
Yes they might be compatible. But if (2) is true, then the discoveries of (1) quickly become irrelevant. This is what Horgan is getting at. If new theoretical ideas don't have any practical implications for our corner of the milky way galaxy, it is going to become very very hard to keep up a never ending sequence of experimentally confirmed discoveries.
Note that Carroll was not claiming that Newtonian mechanics explained all of everyday life. He was claiming that the standard model plus general relativity contained all of everyday life. The analogy to Newtonian mechanics is to help people see how little impact on everyday engineering practice even the discovery of quantum mechanics had (and QM is a hugely practical theory that explains materials, semi-conductors, etc as you note). My guess is that the quantitative discrepancies between current predictions and measurements leave little hope that discoveries beyond the standard model and GR are going to have any practical applications except maybe in the distant future if we need 15 digits in the magnetic dipole moment of the muon or are trying to travel outside our galaxy or are considering the heat death of the universe. Like Horgan, I would love to be wrong. But I just haven't heard any good empirical arguments to support claim (1) that are deeper than 'past performance predicts future results' mixed up with wishful thinking.
The mercury/Vulcan reference is an interesting one. Note that the precession of Mercury is a pretty small effect and the GR effects that produce it have only recently begun to have practical applications in the global positioning system clocks. Do you have modern candidates to propose where unobserved entities are hypothesized to patch up measurements with the standard model or GR? Dark matter and Dark energy are the obvious candidates. The Higgs particle also could have been such a thing. If it had not been observed we might have needed a paradigm shift, but even that probably would not have made much practical difference. You don't need to be able to predict the masses of the particles in the standard model for practical purposes. It works quite well to measure their properties and use that to predict how they behave.
A 19th century analogy is to chemistry. Chemists in the late 19th century were able to put in place much of modern chemistry without having correct ideas about how a chemical bond actually worked. Quantum physics did not replace chemistry. It mostly explained things that were already known. But it did something else...it suggested where to look for new discoveries: the addition of Hafnium to the periodic table for example. It also opened up precision calculations of bond energies, bond lengths, etc which have been quite useful. That is what happened when we finally figured out how ordinary matter works...the stuff we are made of. (1) is a guess that figuring out what dark matter is or the properties of possible non standard model particles formed at energies above 100GeV will guide us to new areas of inquiry that lead to breakthroughs and practical applications. But it seems an unsupported guess. We already have discoveries like the top quarks and tau neutrino that didn't lead to any significant breakthroughs or new technology. Does anyone have any dark matter hypotheses for which they have a potential application? It is just very hard to imagine what one could do with a source of gravitational force that interacts so weakly with ordinary matter than it is currently only detectable on length scales of an entire galaxy. And dark energy is much much farther from engineering use.
So let's keep trying to find new breakthroughs. But it is also good to be honest with ourselves about what we have reasons to expect and what is wishful thinking.
The subject of the likelihood of future breakthroughs in basic science is very important. But Horgan is not very good at articulating the main issues. Much better is Sean Carroll's blog: http://www.preposterousunivers...
To simplify the situation to make it comprehensible, consider two hypotheses about the future of science. (1) Science will have an eternal sequence of groundbreaking discoveries/paradigm shifts. (2) The highly successful models we currently use are so accurate that they will continue to be used forever.
The first hypothesis is beloved by scientists in search of funding and by sociologists of science who wish to treat science as merely a social construct. It really is a strange alliance, but a common cause can make strange bedfellows. The second hypothesis is much less widely defended. Partly because it is clearly false in a fundamental sense...we know that current models don't describe dark matter for example, and so they have to be wrong and are likely to be replaced. But the weight of the second hypothesis is on the 'accuracy' of our current models of fundamental physics. As Carroll clearly argues, there is nothing of practical importance in everyday life that we can show to be in violation of the current laws of physics. Of course there will be major breakthroughs in applied physics...major things like figuring out how atoms and cells form brains and intelligence or discovering how to compute solutions of quantum many body systems. But if we are forced to choose between the two hypotheses, I think the evidence leans toward Carroll's side: the fundamental physics of everyday phenomena does not deviate in any significant ways from known physics.
Many people can't seem to see the vast gulf that exists between the discoveries of Maxwell's equations or quantum mechanics (which are necessary to describe matter and light, fundamental aspects of our lives) and current work on dark matter and primordial gravitational waves (which require precision detectors observing things from outside our galaxy). Also, before you dismiss (2) with references to late 19th century quotes about the end of physics, take a few minutes and look at the history beyond the quotes. Those quotes were mined for science funding publicity. Many scientists in the late 19th century knew that they couldn't explain atomic spectra...Kelvin even worked on vortex models of atoms. And if you focus your attention on 'practical' physics, the claims that late 19th century physics was nearly complete turn out not to be too far off...engineers spend almost no time studying quantum mechanics or relativity. It might be 1 or 2 courses out of 20 that cover physics that was discovered since the end of the 19th century. In mechanical engineering there are typically zero courses on quantum mechanics or relativity.
Yes, a more realistic vision of managing science would be an important improvement. Currently you make your way to a permanent position by producing a lot of results that impress established scientists, which in practice often means you extend and confirm their work. This expands the community in which the senior established scientists run the show. But they are expected to manage and do science. Many of them are not skilled in managing, and when they do manage well, they are no longer able to engage in the actual science in a very substantial way. But the managing and doing can't be fully separated. How are management decisions to be made without an awareness of the subtle questions about where barriers to progress will pop up?
If I had one simple way to improve the situation, it would not be to encourage more maverick science. (It is just too difficult to separate true genius mavericks who will make major contributions from the much larger number of delusional smart people who are dreaming up new ways of being totally useless. If there are breakthroughs to be made by genius mavericks, they probably are going to need to make them while serving as patent clerks in the time honored model of Einstein.) I would replace the system of giving credit for large number of papers cited many times. This system reinforces a kind of 'follow the crowd' style of science that creates huge numbers of papers, none of which are particularly clearly written or provides a major advance. More credit needs to be given to people who write fewer clearer papers which waste less of their colleagues valuable time trying to review and decipher. The emphasis should be on the number of significant new ideas contributed and not on the number of highly cited papers which is more a measure of scientific fads than of substance.