neurologicablog Feed

Yesterday I started a response to this article, which seems to me fits cleanly into a science-denial format. The author is making a lawyers case against the notion of climate change, using classic denialist strategies. Yesterday I focused on his denial that scientists can ever form a meaningful consensus about the evidence, conflating it with the straw man that a consensus somehow is mere opinion, rather than being based on the totality of the evidence. Today I am going to focus on the notion of “post-normal” science. Macrae gives this summary of what post-normal science is:

“The conclusions of post-normal science aren’t ultimately based, then, on empirical data, with theories that can be rigorously tested and falsified, but on “quality as assessed by internal and extended peer communities,” i.e., “consensus,” i.e., informed guesses.”

This is another straw man. He is creating a false dichotomy here, based on his misunderstanding of science (he is a journalist, not a scientist). Yesterday I gave this summary of how science works:

“Science is not a simple matter of proof. There are many different kinds of evidence – observational, experimental, theoretical, and modeling (computer modeling, animal models, etc.). Scientific evidence can use deduction, induction, can start with observation or start with a hypothesis, can use theoretical constructs, can make observations about the past and make predictions about the future. All of these various activities are part of the regular operation of science. No one type of evidence is supreme or perfect – they all represent different tradeoffs. Scientific conclusions are always a matter of inference – scientists make the best inference they can to the most probable explanation given all of the available evidence. This always involves judgement, and some opinion. How are different kinds of evidence weighted when they appear to conflict?”

He seems to believe that the only “real” science is one based on pure evidence, requiring no opinion or judgement – but this does not exist. There is no proof in science, only inference based on the evidence, which is always partial and imperfect. But this is the strategy of science denial – create an artificially narrow definition of science (which may sound reasonable to a non-scientist) then try to exclude the science you want to deny from “real” science. So, evolution deniers claim that no exploration of the past can be “real” science because you cannot do repeated experiments on the past. No one was there to observe it. Now Macrae is saying we cannot do science about the future, because you can’t experiment on the future, only make “guesses”.

Macrae is also repeating another common evolution-denial tactic of saying that climate change cannot be falsified. He has to go there because his notion that you cannot do science about the future is obviously false when you consider that science often functions by predicting what will happen in the future, and that such prediction can potentially be falsified. He claims climate models are not real science (they are one piece of doing climate science) because even if they are wrong, climate scientists don’t change them. But in order to make this point, he has to misrepresent how well the climate models over the past 50 years have matched actual warming.

To do this he again employs a common denial tactic – reference outliers that agree with your position. There are three prominent climate change denying scientists that always seem to be quoted – Lindzen, Spencer, and Christy. A thorough exploration of their claims is beyond this post, but suffice to say, they are a minority opinion, far from the  mainstream of their field. Every field has such outliers. Again – this is why we look to see if there is a consensus in a discipline, to see where the weight of opinion is. Otherwise you can play – choose your own expert – to find whatever opinion suits you. In this case, Macrae cites Christy to claim that climate models have over-called warming. He states this as a fact, without disclosing that Christy’s analyses are controversial at best, and clearly in the minority.

Here is a good review of climate models by an academic source. They conclude:

“Climate models published since 1973 have generally been quite skillful in projecting future warming. While some were too low and some too high, they all show outcomes reasonably close to what has actually occurred, especially when discrepancies between predicted and actual CO2 concentrations and other climate forcings are taken into account.”

You have likely seen these projections, with a line surrounding by a zone of uncertainty, projecting temperature into the future. Actual warming has been within this zone (within 2 standard deviations from the average predicted warming). What they mean by taking discrepancies into account – if you ran a model in 1980 and plugged in a predicted amount of CO2 release, but the actual CO2 releases was more or less, the model will be off not because it doesn’t work, but because the wrong amount of CO2 was entered. We can then run the model again with the correct CO2 and see how its predicts warming. But even without this, the models have done generally very well. They are not perfect, but accurate, and are being tweaked all the time to get more sophisticated and more accurate.

Much of what Macrae says after this is based on the false premise that climate models don’t work but scientists ignored this – hence climate science is not falsifiable. But this is nonsense – most analyses find that the climate models work just fine.

Macrae also, even within his false premise, is committing another denialist trope – saying that because the models were allegedly off (they weren’t) they are therefore wrong. Evolution deniers do this a lot as well – because scientists were wrong about the branching pattern of evolutionary relationship among certain species, perhaps evolution did not happen at all. Even though the cherry-picked outlier he chose shows the models were off, they still predicted warming and the globe is warming. They were correct about the direction and persistence of warming, just off in terms of magnitude (again, according to Christy, but not the majority of climate scientists).

Taken together these strategies that Macrae is using are common among many campaigns to deny accepted science (accepted because the totality of evidence favors those conclusions). But of course, he denies the denial, even while blatantly engaging in it.

The post Is Climate Science “Post Normal” Science – Part II first appeared on NeuroLogica Blog.

This article is from a year ago, but it was just sent to me as it is making the rounds in climate change denying circles. It is by Paul Macrae, who is an ex-journalist who now seems to be primarily engaged in climate change denial. The article (a chapter from his book on the subject) is full of the standard climate denial tropes – for the sake of space, I would like to focus on three specific points. The first is the claim that climate science is “settled”, the second is the notion of “post-normal science”, and the third is a factual claim about the accuracy of prior climate models.

Of course, if there is a consensus among climate scientists that global warming (I will get into more details on what this means) is “settled”, that makes it difficult, especially for  a non-scientists, to question the conclusion. So order number one – deny that there is a consensus, deny that consensus is even a thing in science, and deny that science can ever be settled.  I don’t suspect that I will ever be able to slay this dragon, it is simply too useful rhetorically, but for those who are open to argument, here is my analysis.

First – consensus is absolutely a thing in the regular operations of science. A consensus can be built in a number of ways, but often panels of recognized world experts are assembled to review all existing scientific data and make a consensus statement about what the data shows. This is often done when there is a policy or practice question. For example, in medicine, practitioners need to know how to practice, and these consensus statements are used as practice guidelines. They also set the standard of care, so as a practitioner you should definitely be aware of them and not violate them unless you have a good reason. Obviously, the question of global warming is a serious policy question, and so providing scientific guidance to policy makers is the point, such as with the IPCC. Consensus is also used to set research and funding priorities, to establish terminology, and resolve controversies. But to be clear – these mechanisms of consensus do not determine what the science says. That is determined by the actual science. The point is to provide clarity regarding complex scientific evidence, especially when a practice or policy is at issue.

The reason we need such expert review is because scientific evidence, as regular readers here know, is complicated. Science is not a simple matter of proof. There are many different kinds of evidence – observational, experimental, theoretical, and modeling (computer modeling, animal models, etc.). Scientific evidence can use deduction, induction, can start with observation or start with a hypothesis, can use theoretical constructs, can make observations about the past and make predictions about the future. All of these various activities are part of the regular operation of science. No one type of evidence is supreme or perfect – they all represent different tradeoffs. Scientific conclusions are always a matter of inference – scientists make the best inference they can to the most probable explanation given all of the available evidence. This always involves judgement, and some opinion. How are different kinds of evidence weighted when they appear to conflict?

So it is a meaningful question to ask – to different scientists looking at the same question from different perspectives come to roughly the same conclusion? For example, do those doing ice-core analysis large agree with those looking at tree rings? We have data from different kinds of temperature measurement, from physicists looking at the activity and influence of the sun, and the role of CO2 as a greenhouse gas. We may even have data from planetary astronomers. Then we have various computer models, which take input from many sources of information. No one source of information is definitive. So scientists triangulate from different perspectives and see if they align on the same answer. The only way to know is to see if there is a consensus among the various scientists, each with different pieces to this enormous puzzle.

But of course, in order to be precise, you have to break down the question of global warming into it’s specific pieces – does CO2 drive warming, are there other sources of climate forcing, what is the climate sensitivity to CO2, is the planet actually warming and how much, and what will be the consequences of different levels of warming? We can’t just say – climate change or global warming, we have to address each component separately.

Is science ever “settled”. Macrae conflates this notion with “certain” (there are such straw men throughout his article). Science is never 100% certain, and it is never done. But there is a place for the notion that some claims in science are so well-established that they are functionally settled, meaning we no longer have to specifically establish them over and over again. We can take them as a given and move on to more detail and other sub-questions. For example, it is settled that the Earth is roughly a sphere, while planetary scientists continue to revise greater and greater detail. It is established that life on Earth is the result of organic evolution from a common ancestor, that the brain is the organ of the mind, that DNA is the molecule of inheritance, that plate tectonics is real, and that multiple sclerosis is an auto-immune inflammatory disease. Research in all these areas is ongoing, but there is very strong agreement (a consensus) that these basic fundamental questions are settled.

They could still, theoretically, be overturned, but the probability is so close to zero we can treat it functionally as zero. It is simple not a serious scientific possibility that the Earth is flat, that life was created 10,000 years ago, that consciousness lives in the heart, that proteins carry inheritance, that the Earth is completely static, or that MS is caused by an imbalance of the humors. Would Macrae agree that any of these questions are scientifically “settled”? Should we give serious consideration to flat-Earthers?

With regard to climate change, it is well-established (use whatever phrase you like) that CO2 is a greenhouse gas, and that the planet is warming. It is also well-established that industrial release of previously sequestered CO2 into the atmosphere and therefore the carbon cycle is forcing the climate to become warmer on average. There is a range of possible climate sensitivity, which is open to revision, but that range has statistical confidence intervals and the range is narrowing as our confidence increases through further research. Exactly how much warming will occur is open to further study and revision, but again there is a range with confidence intervals. What will the consequences be? This is difficult to predict, but there are some very reasonable statements that we can make, informed by climate models and what has already happened over the last 30 years. But there is a lot of uncertainty – and of course, that uncertainty cuts both ways. It could be better than the average prediction, or it could be worse.

What is not reasonable is to assume everything will be fine. This is like facing the possibility of cancer, with the same degree of uncertainty. Just hoping that it’s all fine and doing nothing is likely not a rational course of action. This doesn’t mean you have to opt for the most radical surgery either. There is often a range of options, which can be determined by the level of evidence and the resulting risks vs benefit. Doing nothing about climate change until we have a high degree of certainty is also not a rational course, because climate change gets harder and harder to deal with the longer it goes and the worse it gets. Solutions also take decades to unfold.

Typically, those in the denier camp use the most unreasonable or extreme version of climate mitigation strategies as if they are the only option. This is like alternative medicine proponents characterizing all cancer treatment as “cut, burn, and poison.” Macrae similarly writes:

“And shouldn’t we be especially wary when this science, with its attack on fossil fuels, threatens the very foundations of Western-style civilization?”

Now who’s the alarmist? Sure, there are extremists on the fringe of every movement. In terms of actual proposed policy, however, and the center of gravity of climate discussion, we are mostly talking about investing in R&D, investing in infrastructure, and jiggering the markets away from fossil fuels and towards green energy. There is no serious policy discussion about banning fossil fuels and collapsing western civilization. There is nothing like that in the Paris Accords, or in any UN recommendation. Whatever you think about the effectiveness of Biden’s policies, they were entirely carrots for industry to increase investment in green energy. About the most radical actual policy proposal is a carbon tax, which most economists agree would likely be effective. This would hardly collapse civilization.

These are all inevitable technologies because they are superior to burning fossil fuels on many levels – cleaner air, reduced health care costs, more energy independence. We just want to make them happen faster.

Tomorrow I will write part II, covering post-normal science and the accuracy of climate models.

The post Is Climate Science “Post Normal” Science? first appeared on NeuroLogica Blog.

I am currently in Dubai at the Future Forum conference, and later today I am on a panel about the future of the mind with two other neuroscientists. I expect the conversation to be dynamic, but here is the core of what I want to say.

As I have been covering here over the years in bits and pieces, there seems to be several technologies converging on at least one critical component of research into consciousness and sentience. The first is the ability to image the functioning of the brain, in addition to the anatomy, in real time. We have functional MRI scanning, PET, and EEG mapping which enable us to see cerebral blood flow, metabolism and electrical activity. This allows researchers to ask questions such as: what parts of the brain light up when a subject is experiencing something or performing a specific task. The data is relatively low resolution (compared to the neuronal level of activity) and noisy, but we can pull meaningful patterns from this data to build our models of how the brain works.

The second technology which is having a significant impact on neuroscience research is computer technology, including but not limited to AI. All the technologies I listed above are dependent on computing, and as the software improves, so does the resulting imaging. AI is now also helping us make sense of the noisy data. But the computing technology flows in the other direction as well – we can use our knowledge of the brain to help us design computer circuits, whether in neural networks or even just virtually in software. This creates a feedback loop whereby we use computers to understand the brain, and the resulting neuroscience to build better computers.

The third technology is the brain machine interface (BMI). This allows biological brains to talk to computer software, and through that to robotic prosthetics and any other application that can be run digitally. So far it seems like our brains are happy to accept input from software and can learn to control robotic limbs. A robotic hand, for example, can have sensory feedback in addition to motor control, and this closes the loop in the brain so that the user feels as if they own and operate the robotic limb, more like their original biological limb.

All these technologies together, but especially the first two, are building toward a final goal (among many) of creating a human connectome – a map of all the circuits in the human brain at a functional level of resolution. Along the way there have been some interesting milestones. Back in 2011 researchers built the first computer model of a mouse cortical column, a complete circuit in the brain. Since then this kind of research has taken off.

There are several things happening at once: Researchers are modeling brains in some combination of hardware and software, and sometimes in “wetware” that mimics how neurons function. They can also create circuits that combine silicon and living neurons, which can function and learn. Further they can map brain circuits virtually to see how they behave, learn and function. They are also using our knowledge of how the brain and neurons work to design computers and AI that is perhaps more efficient and powerful.

I see no reason why all of these technologies, working together, will not eventually achieve, through nothing but incremental advances, a complete model of a human brain, either virtually in software, or in hardware, or in some combination. This will enable us to confirm (sort of) the ultimate question about the human mind – is the mind an emergent property of brain circuits functioning in real time? If so, then a virtual or silicon brain should be conscious.

Of course, we will not know if the virtual brain experiences its own existence, only that it acts as if it does. We may create an AI p-zombie – a philosophical zombie that acts sentient but does not experience its own existence (no qualia, as the philosophers say). But at this point I think we will be obligated to treat a virtual brain as if it is a sentient being, since it will be indistinguishable from one.

But even if we set aside this question aside, we will be able to model human brain function and measure its behavior and output. This would give us an amazing research tool. We can endlessly alter the circuits to see what happens. We can model psychiatric conditions, like schizophrenia. We can find out how different circuits behave and interact to create the human mind in all its aspects.

Back to BMI – we will also be able to network all of this with biological human brains. We will be able to merge with our AI, to extend our brain capacity with silicon. Will this work? Every indication so far says that it will. I imagine it can function like a third hemisphere. Each hemisphere of our brain is capable of generating independent consciousness – each hemisphere is you. They also contribute their unique function. But they are so robustly connected and networked together that they function like one mind (to our subjective experience).

So – a third silicon hemisphere, robustly connected to the two biological ones, should also function as part of a single mind, just one with expanded capacity and function. Imagine living much of your life with such a computer extension. It would become part of you – it would become you. If it were powerful enough, you may not even notice when the biological hemispheres are damaged or die – unless they are still needed to interface with your body. But if that could be duplicated as well, to create redundant connection from your silicon brain to all the brain’s inputs and outputs, then you would not notice.

But even if we cannot do that last part, your consciousness would continue, perhaps with little change. It could theoretically be placed in a virtual environment, or in an android, or (as in the series Altered Carbon) in another biological body.

These last applications are for the far future – but creating an entire human brain in some combination of hardware, wetware, and software will likely happen sometime this century. Perhaps it will run on a quantum computer, or some advanced neural network that models human neurons as much as possible. Either way, it is the ultimate extension of the current paradigm of neuroscience – the mind is what the brain does, it is an emergent property of brain function. Silicon or virtual brains should demonstrate the same emergent behavior.

The post The Future of the Mind first appeared on NeuroLogica Blog.

The popularity of the acai berry as a so-called “superfood” is a triumph of marketing over reality. This is a berry from the Amazon that was eaten by local people – because it was available – but was then marketed by a company called Sambazon and became an international sensation.

There are lots of berries around the world that are a fairly good source of vitamins, but none of them deserve the moniker “super”. That is pure marketing hype. Acai is bitter (does anyone actually like acai?) and has to be imported from the Amazon, while if you live in the US there are lots of better choices, like blueberries. Why would people bother? Because they were lied to by celebrities.

The idea is that a superfood packs so much nutrition, including things like antioxidants, that they have specific health benefits. This is not a credible claim, and there is no solid scientific data to back up such claims. Generally the companies trying to market these foods will finance some studies designed to generate marketing copy, but nothing am0unting to serious research.

The very concept of a “superfood” is flawed, and likely to be counterproductive. What matters is one’s entire diet, not one component of the diet. The best approach is a simple one – eat a varied diet containing plenty of fruits and vegetables. That’s it – there, I just saved you from having to buy any healthy eating books, spending time or money on fancy diets, or obsessing over minutiae regarding your diet. Just eat your fruits and veg.

If you want to make sure your diet is varied enough, then read up on some basic nutrition facts (like we all had to in middle school). If you are healthy then it’s not complicated – again, variety is your friend. If you have a medical condition or a family history affected by your diet then you will want to consult a health expert for special dietary issues.

You can get pretty close to an optimal diet by simply following that basic guide, but people often obsess, trying to squeeze out an extra percent or two of benefit (of course, they have been false lead to believe the benefits are much more). You end up spending lots of money and time, and likely do more harm than good. Relying on an alleged superfood is likely to make your diet less varied and may also give you a false sense of security.

This also relates to another inherent flaw in the superfood narrative – it is largely based on the notion that if some is good, then more must be better. But nutrition (and biology in general) does not work that way. Once you are getting enough vitamins, more is of no magical benefit.

Since almost every item marketed as a superfood makes the antioxidant claim, let me quickly review why this is also bunk. Your body makes its own antioxidants, and they are 1000 times (literally) more potent than anything you can eat. We evolved a dynamic homeostasis in which our mitochondria produce oxygen free radicals as part of generating energy from food. The oxygen radicals can cause cell damage, which is why we produce the antioxidant to scavenge these oxygen radicals before they can do damage. But (this is critical) our immune systems also use oxygen radicals as part of how they kill invading cells. Also, oxygen radicals are a good measure of metabolism, so we also evolved to use them as a signal to also perform other important measures to keep cells healthy.

If it were better to shift this balance in the direction of anti-oxidants, we would have evolved that balance already. You are not going to hack millions of years of evolution and make is better simply by eating some anti-oxidants. This is not something you need to worry about – unless you have a genetic illness which impairs your normal anti-oxidants. And of course, the clinical research does not show any benefit to routine supplementation with anti-oxidants (it’s actually possible with high enough doses to cause harm, but this is not clear in the research).

Researchers had this figured out two decades ago, but the marketing ploy of “superfoods” and “anti-oxidants” is still effective, so companies are still exploiting this bit of fiction. They are also always on the look out for the next superfood – it’s a great money-making formula. Find some obscure food that no one wants, create false hype about it being a superfood, as a bonus you can pay for some bogus research you can spin to support this narrative, get a celebrity to endorse it, and then watch people flock to trendy cafes to give you their money.

The post Superfoods Are Bunk first appeared on NeuroLogica Blog.