Hacker News Reader

Comment by brockf | Hacker News Reader

You should really talk to some cognitive scientists for some depth and perspective on the issues of studying the mind.

There are real things to consider about the study of the mind, predisposition and plasticity, dynamic systems versus symbolic computation, etc., and you're comparing p-values between particle physics and studies with human subjects? Armchair theorizing about autism? Calling all of psychology (which spans more sub-fields than one might imagine) a psuedoscience? Disappointing, to say the least.

p.s. The DSM is written by psychiatrists.

View on HN

> You should really talk to some cognitive scientists for some depth and perspective on the issues of studying the mind.

Oh, really? Would that be those same cognitive scientists whose therapeutic methods has been shown to be indistinguishable from a placebo effect?:

Link: http://www.brown.uk.com/depression/parker.pdf (warning: PDF)

Title: "Treating depression with the evidence-based psychotherapies: a critique of the evidence"

Conclusion: No reliable evidence supporting CBT over any other method or no method.

> There are real things to consider about the study of the mind ...

First, from a scientific standpoint, the mind doesn't exist. Second, psychologists only describe, they don't try to explain. Third, theories require efforts to explain, and finally, science requires theories.

No mind, no empirical observations, no efforts to explain, no theory, no science.

Prove me wrong -- send me a picture of the mind, or its weight, or its morphology, or its location, or its distance from the brain. I will accept a photograph of Post-Traumatic Stress Disorder as a substitute.

> ... and you're comparing p-values between particle physics and studies with human subjects?

Your objection seems to be that it's unrealistic to compare p-values between psychology and real science. Well, I agree -- that was the point of my example.

> Armchair theorizing about autism?

All theorizing about autism is armchair theorizing, because no one has offered an explanation, only differing descriptions -- the plague of psychology.

> The DSM is written by psychiatrists.

Nonsense. They're involved, as are clinical psychologists and others. In any case, a psychiatrist is a psychologist with an M.D. degree. You're comparing oranges and oranges.

> Calling all of psychology (which spans more sub-fields than one might imagine) a psuedoscience?

It is a pseudoscience until it offers testable, falsifiable theories that define the field itself, something that's not close to happening. BTW this is not only my claim, it is that of many critics of the field over decades.

Psychology, unlike legitimate scientific fields, has no theoretical core of testable claims that defines the field. This is a requirement for the status of a scientific field.

Physics has the standard model. Biology has evolution and genetics. Geology has physics and plate tectonics. Psychology has a number of poorly articulated, untestable anecdotes.

Medical researchers can and do control the activities of medical practitioners. The reason? A clear, tested theoretical core and solid scientific evidence.

Psychological researchers cannot control the activities of psychological practitioners. The reason? No theoretical core and no scientific evidence.

And until this changes, people like you need to stop claiming that psychology has a scientific basis -- it doesn't. The present state of clinical psychology puts the lie to that claim.

Richard Feynman made this general claim long before I or others did:

http://en.wikipedia.org/wiki/Cargo_cult_science (discussion)

http://neurotheory.columbia.edu/~ken/cargo_cult.html (article)

Quote: "I think the educational and psychological studies I mentioned are examples of what I would like to call cargo cult science. In the South Seas there is a cargo cult of people. During the war they saw airplanes with lots of good materials, and they want the same thing to happen now. So they've arranged to make things like runways, to put fires along the sides of the runways, to make a wooden hut for a man to sit in, with two wooden pieces on his head to headphones and bars of bamboo sticking out like antennas--he's the controller--and they wait for the airplanes to land. They're doing everything right. The form is perfect. It looks exactly the way it looked before. But it doesn't work. No airplanes land. So I call these things cargo cult science, because they follow all the apparent precepts and forms of scientific investigation, but they're missing something essential, because the planes don't land."

And, as proven by the Asperger's fiasco (the opportunistic introduction and hasty removal of Asperger's as a "disease" recognized by the DSM) among other things, nothing has changed.

This is wrong on so many levels. Psychology uses indirect measurements of cognitive/behavioural phenomena just like particle physicists. This idea of "real" versus "pseudo" (or "soft", as some people say) science is intuitive but does not hold up to real scrutiny. All scientific measurement is an approximation, an abstraction, etc., and good theories/measurements give us the opportunity to explain/model/predict phenomena without holding it in our hands. Has a string theorist held a vibrating string? Did Einstein travel the speed of light?

The problem with comparing p-values between psychology and particle physics is statistical. In psychology, there may be 2 groups of 30 participants being examined with a crucial manipulation. So there's ONE Fisher-type statistical comparison with a relatively few degrees of freedom (because there aren't millions of subjects). In particle physics, they make millions of comparisons (same with genetics studies, etc.) which, by random chance, will result in many statistically significant findings at the 5% level. So they use incredibly low p-values to restrict these Type 1 errors, and still have frequent unreplicable, false findings (see http://jama.jamanetwork.com/article.aspx?articleid=181647). Discussing p-values as if smaller p-values == better science is incorrect, and does not reflect either field or their statistical contexts accurately.

Read this article. It's a model that makes testable predictions about cognition. It's one of tens of thousands. http://groups.psych.northwestern.edu/gentner/papers/Gentner8...

> This is wrong on so many levels.

Fine. Tell me the testable, falsifiable theory or theories that define psychology. There are scientists in psychology, and they publish scientific papers, but this cannot by itself turn psychology into a science.

If I publish an astrology paper, one that meets all the requirements for a scientific paper, have I made astrology a science? No? Why not?

Since your claim is that psychology is a science, show me the evidence that prevents psychologists from practicing Recovered Memory Therapy. In fact there is no such evidence, and psychologists are free to practice Recovered Memory Therapy. The only difference from the 1990s is that courts of law have ruled that they will no longer hear cases in which recovered memories are the primary evidence. In other words, the social corrections made to prevent further injustices took place outside the field.

> Read this article. It's a model that makes testable predictions about cognition.

"Testable predictions"? False. There are no subjects, no measurements, no control group, and no quantitative statistics.

And you certainly chose a typical psychology paper. It's a description, not a testable explanation, and there is no basis for falsification. And it won't be replicated, for a number of reasons, including that it is not science, it is philosophy.

> It's one of tens of thousands.

Yes, I know. I read them all the time. I can't believe what psychologists thinks represents science.

> Discussing p-values as if smaller p-values == better science is incorrect ...

A typical psychology claim, and false in general. Meanwhile, in the world of science, physicists avoided making any kind of claim about the Higgs Boson until they achieved a p-value of 5 Sigma, or 3×10-7, or about 1 in 3.5 million.

http://blogs.scientificamerican.com/observations/2012/07/17/...

As to the p-values typical in psychological work (i.e. .01 to .05), this is what scientists have to say about it:

http://www.indiana.edu/~kruschke/articles/Kruschke2011PoPSco... (warning: PDF)

Quote: "Despite its routine use, NHST has many deep problems (as will be described later), and psychologists are transitioning away from NHST to Bayesian data analysis."

On that basis, before you try to lecture people on the role of p-values in psychology, perhaps you should become aware that the standard is being abandoned for cause.

Another reference:

http://ist-socrates.berkeley.edu/~maccoun/PP279_Cohen1.pdf (warning: PDF)

Abstract: "After 4 decades of severe criticism, the ritual of null hypothesis significance testing (mechanical dichotomous decisions around a sacred .05 criterion) still persists. This article reviews the problems with this practice, including near universal misinterpretation of p as the probability that H₀ is false, the misinterpretation that its complement is the probability of successful replication, and the mistaken assumption that if one rejects H₀ one thereby affirms the theory that led to the test. Exploratory data analysis and the use of graphic methods, a steady improvement in and a movement toward standardization in measurement, an emphasis on estimating effect sizes using confidence intervals, and the informed use of available statistical methods are suggested. For generalization, psychologists must finally rely, as has been done in all the older sciences, on replication."

Needless to say, in psychology, replication is still nearly unheard of.

Another reference:

http://www.nature.com/news/nobel-laureate-challenges-psychol...

Title: "Nobel laureate challenges psychologists to clean up their act"

Quote: "Nobel prize-winner Daniel Kahneman has issued a strongly worded call to one group of psychologists to restore the credibility of their field by creating a replication ring to check each others' results. ... 'For all these reasons, right or wrong, your field is now the poster child for doubts about the integrity of psychological research,' Kahneman writes. 'I believe that you should collectively do something about this mess.'"

It seems the psychologists to which Kahneman addressed this call to action, have not acted.

> The problem with comparing p-values between psychology and particle physics is statistical.

Now you're making my points for me, and it's not the first time this has happened. The experimental groups in psychology are too small for the prevailing statistical methods to produce anything but conjecture, as the above article shows, and as you point out:

> In psychology, there may be 2 groups of 30 participants being examined with a crucial manipulation. So there's ONE Fisher-type statistical comparison with a relatively few degrees of freedom (because there aren't millions of subjects)

Yep. It's not science, it's anecdote. And the plural of anecdote is not data. Replication would help, but that's an unrealistic expectation.

> Psychology uses indirect measurements of cognitive/behavioural phenomena just like particle physicists. [emphasis added]

Perfect nonsense. Physicists study nature. Psychologists study the mind. The mind is not an organ, it is a philosophical abstraction.

Prove me wrong. Provide a link to a photograph of Asperger Syndrome. Or to a sensitive scale in the act of weighing pathological narcissism (which should weigh more than the everyday kind).

> Did Einstein travel the speed of light?

You are obviously unaware of the copious, repeatable, falsifiable evidence that supports modern physics. And biology, and geology, and medicine.

In case you think I'm psychology's only critic, here is what the sitting director of the NIMH recently had to say about it in Scientific American:

Source: http://www.scientificamerican.com/article.cfm?id=faulty-circ...

Quote: "In most areas of medicine, doctors have historically tried to glean something about the underlying cause of a patient’s illness before figuring out a treatment that addresses the source of the problem. When it came to mental or behavioral disorders in the past, however, no physical cause was detectable so the problem was long assumed by doctors to be solely “mental,” and psychological therapies followed suit.

Today scientific approaches based on modern biology, neuroscience and genomics are replacing nearly a century of purely psychological theories, yielding new approaches to the treatment of mental illnesses." [emphasis added]

Couldn't have said it better myself.

You literally have no idea what you're talking about. I can't even argue with you because you clearly do not understand the definition of science, and so we're talking past each other.

"Replication is nearly unheard of?" Are you kidding? Read a journal. Seriously. You're citing popular news articles as if you understand anything about the field.

Oh, and I'm still waiting for my picture of gravity. But not here, because this thread is no longer worth my time to check on.

> You literally have no idea what you're talking about.

When you have a coherent argument, post again. It may interest you to know that the sitting director of the NIMH takes the same position.

> "Replication is nearly unheard of?" Are you kidding?

In psychology, replication is nearly unheard of. Richard Feynman commented on this at length many years ago, and much more recently, the same comments appear for the same reason:

Source: http://www.nature.com/news/nobel-laureate-challenges-psychol...

Did you think I just made it up? Psychology is in the midst of a credibility meltdown, and replication is one of the key issues.

> ... as if you understand anything about the field.

I just proved that I understand the field better than you do.

Also, by this logic, neither gravity nor string theory are scientific. You can't weigh them! Or define their location!

> ... by this logic, neither gravity nor string theory are scientific. You can't weigh them! Or define their location!

False analogy. As to gravity, its properties are open to experiment, which leads to theories that can be tested and falsified. Gravity has gone through many revisions over the years -- from being instantaneous action at a distance obeying a 1/r^2 rule, to resulting from spacetime curvature, and all tested, and always open to revision. And the book on gravity is by no means closed -- we need a quantum theory of gravity to be able to unite gravitational and quantum theories.

As to string theory, its practitioners agree that it needs to produce testable predictions in order to become scientific. Which is more than I can say about psychologists, who are in the same position but who insist that what they do is science.

At the moment, physics only understands 4% of mass-energy, while the rest (dark matter, dark energy) is unknown. The difference? Physicists freely acknowledge what they don't know.

Cognitive Psychology is a scientific field. It is different from "Psychology" though, an in fact it repudiates Behavioral Psychology, for instance. Cognitive Psychology is quite distinct from CBT, so criticisms of CBT do not apply to Cognitive Psychology.

The claim that the mind does not exist is a throwback to the dark ages of Behaviorism. They made this claim because they sought to make psychology only about things that can be very easily measured and very easily modeled. They claimed that this was "scientific", and that any talk of the mind was "unscientific". It was the Behaviorists, however, who were unscientific. In their goal to elevate Psychology past the unfalsifiable claims of Freudianism, etc., they made a model that was far too simple, and they ignored all the evidence that a more sophisticated model was necessary, and to make matters worse, they actively acted to suppress science done with more sophisticated models (i.e., Cognitive Psychology), holding the field back by decades.

As to claiming that you can't photograph or weigh minds, this is a non sequitur. You can't photograph or weigh a computer program either. A computer program is a mathematical entity that can be instantiated on hardware, but is distinct from the hardware and is as non-corporeal as is the number 2. The mind is the software of the brain, and Cognitive Psychology is the field that aims to reverse engineer this computer program. Neuroscience, no doubt, has compatible goals, which is why at MIT Neuroscience and Cognitive Psychology are in the same department.

In any case, you do the world no favors by promulgating Behaviorism's ill-conceived notion of science and their specious reasoning.

Reply on news.ycombinator.com