Why Many People Perceive the Study of Human Behavior as Unscientific (2011) [pdf]
web.missouri.edu
web.missouri.edu
It goes as far as to compare the "highly probabilistic predictions" of psychology to those of quantum physics, and it even cites Richard Feynman. It's just laughable.
In one instance it uses the word "stochastological".
I'd also suggest that one of the reasons people perceive the study of human behaviour as unscientific is that it so frequently is.
Studies of human behaviour are inextricably bound up in politics. Many studies are designed, whether consciously or unconsciously on the part of the authors, to advance the authors' own political or social agenda.
Even if you avoid that problem, many studies are just plain statistically lousy, and employ deeply non-random samples.
And finally you've got the annoying problem that people know they're being studied and may modify their behaviour accordingly.
> “Psychology Does Not Yield Repeatable Results” How replicable (repeatable) are the results of psychology compared with those of the hard sciences? Larry Hedges (1987) decided to find out. He compared the replicability of findings in particle physics, ostensibly one of the most rigorous domains of physics, with those of several areas in psychology, including the effect of teacher expectations on students’ IQ scores, gender differences in verbal and spatial ability, the effects of desegregation on educational achieve- ment, and the validity of student course evaluations. Using various statistical metrics of consistency, Hedges found that the results of particle physics studies aimed at estimating the mass or lifetime of stable subatomic particles (e.g., the muon) were in general no more consistent that those of psychology. Hedges’s findings suggest that the claim that psychology’s results are far less dependable than those of physics are not supported by data.
> Still, we should not overstate the implications of Hedges’s (1987) findings. As Hedges acknowledged, he did not sample randomly within either physics or psychol- ogy, so his results may be unrepresentative of the domains within these broad fields. Nevertheless, Hedges observed that the results of studies in several other domains of physics, including the estimation of chemical and thermo- dynamic constants, appear to be about equally consistent (or inconsistent, depending on whether one chooses to view the glass as half full or half empty) as those within many domains of psychology.
http://scienceblogs.com/principles/2011/05/05/the-most-preci...
Psychologists have many non-repeatable results but don't shape reliable theories about their subject. Physicists have many non-repeatable results but do shape reliable theories about their subject. The computer I'm sitting at is proof of that.
From: The Trouble with Psychology http://www.arachnoid.com/trouble_with_psychology/index.html
Mr. Lutus has many articles on psychology. Asperger's By Proxy (http://www.arachnoid.com/psychology/aspergers.php) is extremely relevant to the software industry.
HTML: http://arachnoid.com/science_of_mind/index.html
PDF: http://arachnoid.com/science_of_mind/resources/science_of_mi...
You seem to be overlooking inflammation. If only for fun, run your web backwards - starting from dementia (e.g. Alzheimer's), then sleep disorders, then anxiety.
Appropriate diet, moderate exercise, and a good night's sleep are vital as a foundation for good health.
Acceptance that normal behavior is pathologized leads to the understanding that "mental illness" is legislated.
Thank you for making your research publicly available, and open to review and criticism.
---
1) For over five decades I had been 'locked in'. Comedy at it's finest - I cannot contribute, as being misdiagnosed (word needed for profit motivated diagnosis - prodiagnosed?) prevented the ability to communicate that I could not communicate.
2) Tens of millions suffer needlessly. A few months away from improper 'medicine' and glowing rectangles would provide much needed data points.
Tom Lehrer's recent(-ish) song "Sociology" seems very relevant here.
You are correct in that much research, especially research older than 10 years ago, tended to ignore culture. I know my field of study (I/O Psychology) references cultural frequently, either through overt study when possible (desirable given the publication potential) or at least remarking on the limitations of any findings. I don't read much outside I/O research though, so maybe it hasn't progressed much on the whole and the global nature of I/O work for businesses rewards more recognition of culture. It's hard though - accessing globally-relevant datasets for studies of human behavior is often impossible. That doesn't make all studies meaningless, or unscientific.
And yes, you're correct about value-laden framing of results. That doesn't invalidate the science of the results, it just makes it more necessary to think critically when interpreting them rather than relying on abstracts and headlines alone.
Very true. A chaotic system has specific properties that we would be very pressed to identify in human behavior.
> ... and I'm interested in the mathematical proof that behavior can't be predicted.
Think abut what you're saying. Proof that behavior cannot be predicted would require proof of a negative. Proof of a negative, or (the same thing) an evidentiary burden requiring evidence of absence, is a logical error. So this is a non-starter.
Let's say I believe in Bigfoot, and I think others are obliged to dissuade me from by belief. The problem is that this also requires proof of a negative, and Bigfoot could be hiding under a rock somewhere on a distant planet. Bigfoot cannot be proven not to exist, so I'm secure in my evidence-free belief.
This is why scientists adopt the null hypothesis precept, the idea that an idea is assumed to be false until evidence supports it -- it's a simple way to avoid logical pitfalls like believing in Bigfoot.
> We can make many confident statements about the likelihood of certain behaviors/reactions/response, in the aggregate, under given real-world circumstances.
Yes, true, but until those observations lead to a falsifiable theory of human behavior, it's not science. Until then, we can only describe, we can't explain. Testable explanations are the threshold of science.
> And yes, you're correct about value-laden framing of results. That doesn't invalidate the science of the results ...
No, of course not -- thee reason is that such results aren't science. An observation without an explanation isn't science. Here's an example -- if I sit under a clear nighttime sky, then report that I saw many tiny points of light, I've made an observation of very little scientific usefulness, and one that can't meaningfully be falsified by repeated observation. But if I claim those points of light are actually distant thermonuclear furnaces like our sun, my assertion can be tested and possibly falsified. By moving beyond observation, by offering a tentative explanation, I've crossed the threshold of science.
Yes, I know. That was the point of the comment in response to the parent's statement: "[human behavior] is mathematically proven to be unpredictable." But thank you for the lecture.
> Yes, true, but until those observations lead to a falsifiable theory of human behavior, it's not science.
There are plenty of theories. They are typically considered theories on motivation and behavior rather than behavior itself. It's a notable difference, but a sensible distinction. Many of these theories are falsifiable, and in fact have been falsified (e.g., expectations-based models, for example). But they are typically piecemeal, as the complexity of human motivation and behavior currently requires. If the key requirement to be considered "science" is for The One Great Unified Theory, then I agree that it will never be considered a science in my lifetime.
You seem to think psychology is all observation with no attempt at explanation, much less construction of logical theories supporting those explanations. That is not the case in my little fiefdom (though I have no doubt it is true in some disciplines, or acceptable in some journals). Much work has gone into proposing, testing, and refining many theories on motivation and behavior (e.g., reinforcement theory, goal-setting theory, social-cognitive theory or workplace behavior, job design theory, motivated action theory, etc.).
It's not perfect, and it's got a long way to go, but it's also come a long way. Yet it's hardly just observing the stars and making claims, and it's certainly shown utility in helping design systems to influence behavior.
But there, I've said my piece. I've usually got the discipline to stay out this pet topic for the HN crowd.
So you also believe that we have not proven incompleteness of mathematics? Have we also not proven that it is impossible to solve the Halting Problem?
Imagine that I believe in Bigfoot. How shall I be dissuaded from my belief by evidence -- it would have to be negative proof. If I started out assuming Bigfoot wasn't real, that would be much, much easier.
In the same way, people assume that logical incompleteness isn't real until there's positive evidence. They assume there's no halting problem until there's positive evidence. See the difference?
Those aren't theories as science defines the term. Among other things, they aren't falsifiable and they don't make predictions about unobserved phenomena, both hallmarks of scientific theories.
> You seem to think psychology is all observation with no attempt at explanation ...
Wait -- that's not my idea. That's the reason the NIMH gave for abandoning the DSM -- it's all symptoms (descriptions) and no search for causes (explanations/theories):
http://www.nimh.nih.gov/about/director/2013/transforming-dia...
Quote: "While DSM has been described as a “Bible” for the field, it is, at best, a dictionary, creating a set of labels and defining each. The strength of each of the editions of DSM has been “reliability” – each edition has ensured that clinicians use the same terms in the same ways. The weakness is its lack of validity."
"Unlike our definitions of ischemic heart disease, lymphoma, or AIDS, the DSM diagnoses are based on a consensus about clusters of clinical symptoms, not any objective laboratory measure. In the rest of medicine, this would be equivalent to creating diagnostic systems based on the nature of chest pain or the quality of fever. Indeed, symptom-based diagnosis, once common in other areas of medicine, has been largely replaced in the past half century as we have understood that symptoms alone rarely indicate the best choice of treatment. Patients with mental disorders deserve better."
> Much work has gone into proposing, testing, and refining many theories on motivation and behavior ...
You aren't using the scientific meaning of theory -- a formalization of observations using empirical evidence, that generalizes specific observations, that predicts things not yet observed, and that is falsifiable.
> Yet it's hardly just observing the stars and making claims, and it's certainly shown utility in helping design systems to influence behavior.
There are many ideas that meet this description, but they aren't scientific theories as that term is defined.
Prove me wrong. Point to a theory in the class you describe that has been used to conclusively falsify someone else's theory in the same field, a case in which the defender of the falsified theory acknowledges that his theory can been proven false and has no further utility. I say it this way because it happens all the time in science, and it never happens in psychology.
Consider the example of Asperger Syndrome. It was thrown out over a year ago by the editors of DSM-5, described as an unscientific and meaningless term, but I am still getting emails from parents whose children are being assigned the diagnosis.
Consider recovered memory therapy. Many people have been accused of imaginary crimes based on it, and many studies prove the unreliability of this therapy, but it's still offered to anyone foolish enough to ask for it.
Falsifiability is a big issue with respect to psychological theories, for the reason that it doesn't exist.
> But there, I've said my piece. I've usually got the discipline to stay out this pet topic for the HN crowd.
This isn't a contentless, rambling argument in which people say their piece -- at least, I hope not. It should be a discussion of ground truth, with the possibility of a conclusive outcome, like in science, in which someone says, "My position has been falsified". The fact that this never happens within psychology doesn't mean it can't happen here -- in principle.
I hope you see the degree of self-reference that's present here.
And I want to make one thing clear about psychology which just never seems to be clear: psychology is way, way bigger than clinical psychology, yet clinical psychology (and stories from it) seem to be the only thing that is ever brought up in these conversations. I can't speak to clinical psychology. There are 50 divisions of the APA, some significantly better than others. My division is #14.
> Among other things, they aren't falsifiable and they don't make predictions about unobserved phenomena, both hallmarks of scientific theories.
Almost all of those theories are falsifiable, elements of them have been falsified, and they make predictions for unobserved circumstances.
> That's the reason the NIMH gave for abandoning the DSM
The DSM is a diagnostic tool from a branch of psychology I know little about. It is not the singular representative of the social sciences, or even halfway related to much work being done in behavioral prediction. It's not theory-based, as I understand it it is not meant to be, and I've no interest or ability to defend it.
> Point to a theory in the class you describe that has been used to conclusively falsify someone else's theory in the same field, a case in which the defender of the falsified theory acknowledges that his theory can been proven false and has no further utility.
The last part of your request is unreasonable - proving that there is no utility whatsoever in old theories is unnecessary. Elements of older theories can have utility. And as for the original theory's proponent abandoning it? I don't see that often, nor see it as a requirement. Journals I read don't publish mea culpas from those who were ultimately proven wrong, they just publish the evidence proving them wrong.
I pointed out one specifically already in my previous comment (edit: my mistake, I took that out just referenced expectancy theories more generally). Victor Vroom introduced the VIE expectancy model for explaining and predicting behavior in 1964. Over 20 years the research indicated that the theory as proposed did not jive with the observed evidence, and the model transition from one based on "psychological force" to "outcome attractiveness" as chronicled by Pinder in 1984. Van Erde and Thiery, in 1996, that the preponderance of evidence on the model, revised or otherwise, demonstrated that even the revised theory was inadequate. The individual elements had predictive power, but as a model it just didn't work. Did Victor Vroom abandon it? Who knows? The field certainly did. But did the field note that nothing in theory had any utility? No, certainly not. Locke and Latham had integrated some of the elements into explaining goal commitment, a critical component of goal-setting theory.
There was also the feedback theory of workplace behavior popularized by Komaki in the 80s and early 90s, and was effectively dismantled as a stand-alone theory both logically and empirically Locke and Latham in the mid-to-late 90s. I have no idea whether Komaki acknowledged they were wrong publicly, but the field moved on. Even something so basic as goal-setting theory has wholesale change in just the past 5 years with the measurement and observation of subconscious goals, and how they don't work with traditional goal-setting theory.
Deci and colleagues outlined and empirically supported Cognitive Evaluation Theory in the 90s and early 00s. Among others, Rynes dismantled the bulk of it in the mid-to-late 00s, and as far as I know it's essentially dead.
Another one is job characteristics theory, which I don't recall the whole story of off the top of my head. It came, it didn't prove to be reliable, it was abandoned. Again elements have found their way to better homes (i.e., links of autonomy to workplace behavior patterns) but on the whole it didn't cut it.
If that is too old, a more modern one is the debate between Social Cognitive Theory in the workplace (from Wood and Bandura, at some point). It's a theory, lots of testable hypothesis, and great empirical support. A pillar of explaining aspects of workplace behavior! A professor named Vancouver out of Michigan State has demonstrated empirically multiple times that the model's predicted outcomes are often at odds with reality. There's been some back-and-forth on this one, and I'm not up to date on it. But it became clear the theory as proposed was not adequate.
> Consider the example of Asperger Syndrome... recovered memory therapy.
They are not my domain of psychology. Psychology, and social science in general, are very broad terms. There are bad examples. I cannot and will not defend those things, but you should not accept those as archetypal representations of all psychological and social science research.
EDIT: One more thing to add. These things tend to have longer timescales than in other harder sciences because of the nature of the subjects. One or two empirical studies may be enough to definitively disprove ideas elsewhere, but because getting large numbers of people across adequate contexts is so onerous, particularly with non-student, non-US samples, the lag-time for debate can be large.
Yes, and there's a reason. The ground truth about the medical field is what happens in clinics. If clinicians are shaking dried gourds over patients, that tells us about medicine itself, from top to bottom -- the only way something unscientific could happen in a clinical setting is if there's no reliable science anywhere in medicine.
In the same way, if someone builds a bridge by guess and by golly, without any consideration of loading and stresses, this calls into question the scientific basis of civil engineering.
So when I hear a psychologist complain that clinical psychology isn't representative of the entire field, I have to say that kind of compartmentalizing doesn't work anywhere else -- it suggests that psychology isn't unified by a respect for science, or by reliable, falsifiable theory.
>> Consider the example of Asperger Syndrome... recovered memory therapy.
> They are not my domain of psychology.
This sounds increasingly like one of those joke conversations about which kind of Southern Baptist you are. Religion works that way, science doesn't. In science, evidence and tested theories forge a consensus that unifies all legitimate fields, at least until someone produces a better theory, then the process repeats.
Do you suppose that a doctor could say, "I don't care what biologists are doing, it doesn't affect me of my practice"? Do you suppose that a geologist could say, "I don't care what the physicists are doing in Earth Science, that has nothing to do with my work"?
This is not realistic, because the example fields are united by a common respect for the primary role of evidence and the scientific method. I say this to remind you how you sound then you try to describe psychology as though it's a bunch of disconnected pieces. Not that that isn't so, but it provides us with a measure of the significance of science in psychology.
> ... but you should not accept those as archetypal representations of all psychological and social science research.
If that's true, then psychologists lose the right to describe themselves as scientists, or to receive funds earmarked for scientific activities. Taxpayers are willing to contribute to physics because (a) it's a science with an enviable track record, and (b) physicists are on the same page -- they don't say things like, "Don't listen to the particle physicists, they're not the real thing." Only psychologists say things like that.
> The last part of your request is unreasonable - proving that there is no utility whatsoever in old theories is unnecessary.
Then you do not understand science. When the existence of the ether was disproven, to save limited resources the falsification was accepted by everyone -- opinion played no part. When the Phlogiston theory was disproven, it was disproven for everyone, everywhere. This kind of certainty and consensus is what makes science science.
True, but psychology is not the medical field. You've picked the flashiest "branch" of it to make your archetype, but it's probably the least representative of the whole (though admittedly perhaps the best funded). Clinical psychology is more focused not on what is normal, but what is abnormal, how to diagnose it, and how to respond. Almost all of the rest of psychology is focused on what is normal, typically within a given context (childhood, workplace, developmentally, etc.)
> psychology isn't unified by a respect for science, or by reliable, falsifiable theory.
Psychology is not unified by a single reliable theory. I have no problem admitting as much. Maybe someday, but not while I'm alive. Again, if that's your baseline for what it takes to be considered scientific, I'm fine saying "then by that definition it's not scientific" and feeling no shame.
> This sounds increasingly like one of those joke conversations about which kind of Southern Baptist you are.
Then you're not thinking critically. Your doctor-biologist example doesn't fly because one's work (biology) directly and materially affects the work of the other (medicine). What clinical psychologists do, more than almost any other branch, has no impact on me presently, nor I on them. Our populations are different, and the goals of our research with respect to our populations are different. Ridicule the division if you wish, but what do you want me to do? Do you expect my field to reconfigure itself entirely to become experts on abnormal behavior and diagnoses? Should I, in my effort to increase workplace productivity and satisfaction, educate myself on the ins-and-outs of autism or psychological disorders? Human behavior is so large and complex to support these various divisions.
Many, many branches are connected in some way. It's a web. I have no ability to speak beyond the basics of clinical psychology. I can speak to psychometrics, social personality, experimental, educational and consumer psychology, and behavioral analysis (all of which are branches of the APA). Outside of a general interest in human behavior with no regard for context, my goals are as different from a clinical psychologists as a biologists are from neurologists. It's hard enough to do good work in specific contexts, unfortunately. Psychology right now is a bottoms-up field because we're just too ignorant as a species (despite efforts) to do it top-down right now.
> I say this to remind you how you sound then you try to describe psychology as though it's a bunch of disconnected pieces. Not that that isn't so, but it provides us with a measure of the significance of science in psychology.
I don't have to try to describe it that way, because it is. It's a pragmatic response to the complications of understanding people on the whole. The whole is too disparate, so look at the pieces. That bothers you, but most don't seem as perturbed. I'm sure you see this as concrete evidence for your point, but you won't be surprised that I think that's poor criteria. "Psychology" as a term is almost certainly overly broad, which is why so many divisions exist in the first place.
> If that's true, then psychologists lose the right to describe themselves as scientists
It is true, and I disagree. Bad science is found everywhere. Look at what bad science has done for vaccines. Do I hold it as an archetype of medical science? Your argument seems to be "we're not perfect at explaining literally all cognition and behavior in a Single Unified Theory now, mistakes have been made in the past, so sod the whole thing."
Physicists are on the same page because their work is infinitely more provable. Ultimately human cognition and behavior is a much more practically difficult, nebulous topic at the present time that's going to take longer to get right on the whole. It's all probably just math itself in some respect, and maybe we'll know the equation someday à la Hari Seldon in The Foundation. We're not there yet.
>> The last part of your request is unreasonable - proving that there is no utility whatsoever in old theories is unnecessary.
> Then you do not understand science.
Then you're not thinking critically. It's this kind of thing that turns me off these conversations completely. For better-or-worse, many theories in psychology are not all-or-nothing. While the theory on the whole may prove false, certain elements of the theory or the direction may prove useful in new theories. Old, falsified theories are invaluable for paving the way towards new theories, and not all elements of all falsified theories need be disposed of whole-hog. You can completely reject and recognize the positive contributions it has made, and the utility elements of it still possess.
I think we know where each other stand on this now. At the very least, you got your examples.
EDIT: Passovers for grammar.
> True, but psychology is not the medical field.
This is becoming circular. I say psychology isn't scientific, you say I have to say which part of psychology I'm referring to, I say the very question indicts the scientific standing of the field, because scientific fields are united by consensus on the meaning of evidence. I use medicine as an obvious example.
Then you reject my comparison with medicine by saying that psychology isn't medicine. One, psychiatrists would vigorously disagree (their livelihood depends on an association with medicine), and two, it's a red herring. The issue is whether a field can call itself scientific if it's not guided by evidence, and if there's no consensus about the meaning of the theories that define the field. In the case of psychology, there are no such field-defining theories, consequently the field is splintered into warring factions like sects of a religious order.
> For better-or-worse, many theories in psychology are not all-or-nothing.
You're missing the point that there are no scientific psychological theories. There are no such theories that are open to falsification by empirical evidence. And there has never been a scientific falsification of any psychological theory, ever, anywhere. Your use of the word "theory" is misguided, because it resembles how scientists talk.
>> If that's true, then psychologists lose the right to describe themselves as scientists
> It is true, and I disagree.
This is not a literature debate and it is not a matter of opinion -- it must be resolved by evidence. No scientist, when given a chance to present evidence, instead says, "I disagree".
> Bad science is found everywhere.
Psychology is not bad science, it is non-science. In bad science, theories are weak and short-lived, but they are falsified, as with astrology, phrenology, the Phlogiston theory, the N-ray theory, and so forth. In non-science, theories aren't open to a comparison with empirical evidence, consequently they cannot be falsified.
No falsifiability, no science.
No, that is not what I said. I said I personally cannot speak to certain branches of psychology, nor defend them, nor think they should be considered archetypes. I won't speak to something I know little about. I see it doesn't seem to stop you.
I didn't reject your comparison because psychology isn't medicine, I reject it because it is a bad comparison. I'll use a more apt one. Different branches of psychology don't interact much because they don't share populations or, often, goals. You want to use medicine as a comparator.
I know for a fact that most pediatricians (my wife among them) don't stay up to date, or much care, what's happening with geriatricians. I know that radiologists aren't terribly interested in the work of physical therapists. Ophthalmologists and cardiologists? Probably not reading the same stuff. And yet they are linked by care and interest in the general human physiology. But does one's work impact the other? Not really, not much. But biologists can certainly impact both. For psychology, the cognitive psychologists probably is the most "upstream." I care what they do like dermatologists care what biologists do.
> You're missing the point that there are no scientific psychological theories. There are no such theories that are open to falsification by empirical evidence. And there has never been a scientific falsification of any psychological theory, ever, anywhere. Your use of the word "theory" is misguided, because it resembles how scientists talk.
I'm only "missing the point" because your insisting that's the critical issue, and I'm saying you're wrong not in premise but in reviewing the state of things. If your use of the word "theory" is for a Single Unified Theory of all Human Behavior and Cognition, then I agree. I don't think that's the baseline for a scientific psychological theory, and there have been plenty of theories within psychology that have been falsifiable, have been falsified, and no longer exist. You even asked for some examples, which I provided. But what you do then is twist the criteria for what counts to meet only the standard you set. You'll say "it isn't my criteria" I'm sure, but I'm not seeing anything that says otherwise.
>>>>> Consider the example of Asperger Syndrome... recovered memory therapy.
>>>> ... but you should not accept those as archetypal representations of all psychological and social science research.
>>> If that's true, then psychologists lose the right to describe themselves as scientists
>> It is true, and I disagree.
> This is not a literature debate and it is not a matter of opinion -- it must be resolved by evidence.
I'm not entirely sure of your point here. What scientific evidence are you asking for that some examples of bad practice / science / outcomes should not stand as the primary representation of all psychological research?
> Psychology is not bad science, it is non-science... No falsifiability, no science.
See, this is the issue. You start with it as a non-science and work your way back. I've already listed several falsified theories in my own line of work. Outside of that: Skinner's theories? Falsified. Piaget's core theories of cognitive development? Falsified. McGregor's Theory X and Y? Falsified. Unlike other branches they don't vanish entirely, but elements remain through new theories and contributions.
But yet, the DSM exists, so it's all hogwash. We are indeed talking in circles.
> No, that is not what I said. I said I personally cannot speak to certain branches of psychology, nor defend them, nor think they should be considered archetypes.
Thanks for that clarification, which changed nothing about the exchange or its meaning.
> See, this is the issue. You start with it as a non-science and work your way back.
No, I start with how science is defined, note that psychology cannot meet the definition, and draw the sole possible conclusion. As did the chairman of the NIMH as he ruled that the DSM may no longer be used as the basis for scientific research proposals, on the ground that it has no scientific content:
http://www.newyorker.com/tech/elements/the-rats-of-n-i-m-h
> there have been plenty of theories within psychology that have been falsifiable, have been falsified, and no longer exist.
This has never happened in a scientific sense. Ideas that become unpopular or that fall out of clinical use aren't falsified by those circumstances.
> I've already listed several falsified theories in my own line of work.
Nonsense. None of the were falsified as that term is defined in science, by way of empirical evidence, the kind of evidence not accessible to those who study the mind.
> Skinner's theories? Falsified. Piaget's core theories of cognitive development? Falsified.
Your claim is absolutely false. These ideas weren't stated clearly enough to be open to scientific falsification, and they were certainly not falsified -- they both have active practitioners who would be surprised to hear your claim.
http://en.wikipedia.org/wiki/B._F._Skinner
No mention of the falsification you claim exists.
http://en.wikipedia.org/wiki/Jean_Piaget
No mention of the falsification you claim exists.
http://en.wikipedia.org/wiki/Phlogiston_theory
Falsified, with a clearly explained reason, based on empirical evidence.
> Different branches of psychology don't interact much because they don't share populations or, often, goals.
Like science? The same might be said about different branches of physics, except that physics is a science, consequently particle physicists, who study nature at the smallest scales, and cosmologists, who study science at the largest scales, productively read each other's literature and attend each other's conferences. They have science in common.
> Thanks for that clarification, which changed nothing about the exchange or its meaning.
It boggles the mind that you'd even say that. The difference between me saying "you must choose a branch of psychology to call unscientific" and "I do not personally possess the knowledge or capability to stand up for branches of psychology I do not know enough about it" seems pretty substantive.
> No, I start with how science is defined, note that psychology cannot meet the definition
If you're relying on wikipedia, allow me to do so as well: science is the systematic enterprise that builds and organizes knowledge in the form of testable explanations and predictions about the universe. You take issue with the latter, but many psychological theories do them, no matter how much cotton you stick in your ears.
I do not argue that the DSM lacks scientific credibility.
> None of the were falsified as that term is defined in science, by way of empirical evidence, the kind of evidence not accessible to those who study the mind.
No matter how loud you say it, it doesn't make it more true. You demand that it is true, but it is not. You sit at the armchair, reading no research or studies whatsoever, and judge me for 1) being unwilling to comment on issues which I'm ignorant on and 2) not being an expert at all things psychology.
Re: Skinner - you are correct, though he had few "theories" to be wrong about. I meant E.L. Thorndike, Skinner's predecessor (and the other one with a famous "box"). His theory of learning was his greatest achievement, but he eventually repudiated it.
Re: Piaget - that is the person wiki, this is the relevant one: http://en.wikipedia.org/wiki/Piaget%27s_theory_of_cognitive_... Nobody is a classic Piaget psychologist anymore. Did they throw it all out and start over? No, they did not, the iterated and keep testing. If that's the red flag to you, then there you go.
> Like science?
Sure, like science. Whatever you say. I don't disagree that physics is a more mature science, where connections between different branches are more clear.
Have a nice weekend.
> If you're relying on wikipedia ...
No, Wikipedia is not my source. I rely on a much better source, the thousands of sources that define science in the same way, unambiguously, including many court rulings, for example those that keep Creationism out of public school classrooms by defining science as something that requires empirical evidence and objective falsifiability.
> Alright, this is the end of the road for me.
Of course it is. Before we had this conversation you had no idea what constituted science and falsification, and now, you still have no idea.
I want to thank you for taking part in this conversation -- I often think it's not possible for someone not to understand science basics -- it's really rather simple -- but then I have a conversation like this, and I get it all over again.
> I don't disagree that physics is a more mature science, where connections between different branches are more clear.
What you think, your opinions, do not matter. Only evidence matters. And it occurs to me that you will never get this.
Scientist A says, "There is a largest prime. I don't know which prime it is, but there is a largest one. Infinity is not a number, it's a concept, therefore there cannot be an infinity of primes. Therefore there is a largest prime. Call it p."
Scientist B says, "I will falsify your claim. I agree that we shall call the largest prime p. Now let q = p factorial, i.e. q = p * (p-1) * (p-2), etc. At the end of the process, q is either a prime larger than p, which falsifies the claim, or q is composite, composed of primes, at least one of which is larger than p, which falsifies the claim. There are no other possibilities. I have falsified the claim."
Scientist A says, "Yes, you have falsified the claim, my opinion doesn't matter, how I feel about it doesn't matter, the falsification relies entirely on incontrovertible, objective evidence, evidence on which all educated observers must agree."
That is science. That is manifestly not psychology.
Another voice on this topic:
https://www.cfa.harvard.edu/~ejchaisson/cosmic_evolution/doc...
Quote: "Today, all natural scientists throughout the world employ the scientific method. Normally it works like this: First, gather some data by observing an object or event, then propose an idea to explain the data, and finally test the idea by experimenting with Nature. Those ideas that pass the tests are selected, accumulated, and conveyed, while those that don’t are discarded — a little like the evolutionary events described on this Web site. In that way, by means of a selective editing or pruning of ideas, scientists discriminate between sense and nonsense. We gain an ever-better approximation of reality. Not that science claims to reveal the truth—whatever that is — just to gain an increasingly accurate model of Nature."
"Despite an emphasis on objectivity, some subjectivity does affect the modern scientific enterprise, for this is work done by human beings having strong emotions and personal values. Yet, with the test of time and repeated observations, objectivity eventually emerges and then dominates, enabling us to reach conclusions free of the biased viewpoint of any one scientist, institution, or culture. As a rational investigative approach used to formulate descriptions of natural phenomena, the scientific method is designed to yield a reasonably objective consensus on the nature, contents, and workings of the Universe." [emphasis added]
Couldn't have said it better myself.
Have a nice day.
> Wait -- that's not my idea.
Yes, it is.
> That's the reason the NIMH gave for abandoning the DSM -- it's all symptoms (descriptions) and no search for causes (explanations/theories)
"The DSM" is not "psychology". NIMH began a project to create a new framework for research funding (the RDoC) rather than using the DSM categories as a basis for research because the DSM is a clinical tool, not a summary of research. While causes are sometimes addressed specifically in the DSM, "search for causes" -- or search for anything -- is not the focus of the DSM, which aims to summarize for clinicians (and for insurers) the general judgement of the clinical community on appropriate categorization of diagnosis. It ideally ought to be guided by the output of research, but its not meant to be either a general purpose summary of research or a basis for research.
That's certainly recognition of the defects of the DSM when applied to a role for which it is not intended, but its not an indictment of psychology.
> Consider the example of Asperger Syndrome. It was thrown out over a year ago by the editors of DSM-5, described as an unscientific and meaningless term
[citation needed] As far as I understand, the reclassification of developmental disorders was not based on the fact that the DSM-IV diagnoses that were eliminated were "unscientific and meaningless terms", but that the DSM-IV definitions were not clear enough for consistent clinical use, and a single named diagnosis with a well-defined severity scale was more useful and would be more consistently applied.
> but I am still getting emails from parents whose children are being assigned the diagnosis.
The DSM 5 is not a mandate. And, indeed, there are many government (state level, certainly) programs that are mandates, and whose rules are written around DSM-IV diagnostic categories, and which haven't been revised based on DSM 5. Consequently, completely independently of the scientific validity of DSM 5 categories and the scientific acumen of the psychological community, there are reasons why DSM-IV categories may persist in use.
> Consider recovered memory therapy. Many people have been accused of imaginary crimes based on it, and many studies prove the unreliability of this therapy, but it's still offered to anyone foolish enough to ask for it.
Yes, and people willing to by snake oil for physical ailments will find people willing to sell it to them. That doesn't mean that medicine is unscientific.
> Falsifiability is a big issue with respect to psychological theories, for the reason that it doesn't exist.
Falsifiable psychological theories exist. In fact, in your own post you point to a psychological theory being empirically falsified, and then note (as if it invalidated psychology somehow) the fact that people commercially exploit the general public's ignorance of that falsification.
> Yes, it is.
I provided the source for the fact that the DSM is being abandoned because of its unscientific basis, you snipped out my evidence in bad faith.
> That's certainly recognition of the defects of the DSM when applied to a role for which it is not intended, but its not an indictment of psychology.
If I took the pre-eminent physics textbook, one used in every physics classroom in the country, and if I proved that it was based on nonsense, would that indict physics itself? Yes, of course it would. The reason? Physics is a science, and all its branches are validated or falsified by the same corpus of central, defining theory, theories described in the field's textbooks.
If psychology were a science, the same rules would apply -- there would be a defining corpus of falsifiable theory, vulnerable to scientific disproof, and explained in the texts that students read and practitioners (i.e. civil engineers) rely on for guidance.
But this isn't true, there is no such theory, in fact psychologists don't even shape falsifiable theories. This places psychology beyond refutation or even definition.
>> Consider the example of Asperger Syndrome. It was thrown out over a year ago by the editors of DSM-5, described as an unscientific and meaningless term
> [citation needed]
DSM-IV has a listing for Asperger Syndrome (AS). DSM-5 does not. The reason is that the "scientists" responsible for DSM-5 held a secret vote and voted against including AS. If psychology were a science, those who voted would have done so in public and they would have provided their reasons in a scientific paper.
But AS was not included in DSM-IV through a scientific process, and AS was not removed from DSM-5 by way of a scientific process. Both decision were based on popularity votes.
Here are some comments about AS during the process that led to its removal, by one of those who voted:
http://www.nytimes.com/2009/11/03/health/03asperger.html?pag...
Quote: "'Nobody has been able to show consistent differences between what clinicians diagnose as Asperger’s syndrome and what they diagnose as mild autistic disorder,' said Catherine Lord, director of the Autism and Communication Disorders Centers at the University of Michigan, one of 13 members of a group evaluating autism and other neurodevelopmental disorders for the manual ... 'Asperger’s means a lot of different things to different people,' Dr. Lord said. 'It’s confusing and not terribly useful.' ... 'It’s not an evidence-based term.'"
The above is as close as one can get to science in psychology -- people's opinions.
> Falsifiable psychological theories exist.
Name one. But first, find out what "falsifiable" means. In science, "falsifiable" doesn't mean "no longer popular with a voting committee", it means proven false with incontrovertible evidence, evidence the meaning of which all similarly equipped observers accept.
> In fact, in your own post you point to a psychological theory being empirically falsified ...
Asperger Syndrome was not falsified, it was abandoned without ever being evaluated scientifically. In the same way, it was earlier accepted into DSM-IV without ever being evaluated scientifically. We still have no idea whether it is, or is not, a distinct condition with a distinct biological cause. The reason is that we have no idea what causes either autism or AS.
Asperger Syndrome wasn't evaluated using scientific evidence -- there is none. It was accepted into DSM-IV by means of a vote. It was removed from DSM-5 by means of a (now-secret) vote.
This is why the NIMH has decided to abandon the DSM, and why the chairman of the NIMH describes psychiatry (psychology's clinical practice) as a pseudoscience:
http://www.newyorker.com/tech/elements/the-rats-of-n-i-m-h
I can anticipate your objection to someone describing psychiatry as a pseudoscience (something along the lines of "psychiatry isn't psychology") -- if someone described particle physics as a pseudoscience, that would indict all of physics, because physics is defined by a single system of interdependent theories. Falsify the theory, falsify the subfield. Falsify the subfield, falsify the field.
> The DSM 5 is not a mandate.
1. Your objection above is missing something -- there is no falsifiable definition of psychology, anywhere. The DSM happens to be a collection of symptoms and associated diagnoses, but it is as close to a definition of psychology as exists.
2. Nice -- abandon the vessel piecemeal, while holding onto the remaining debris. If psychology's defining text is found wanting, abandon it. If someone's pet theory is ridiculed, abandon it. But science doesn't work that way. In science, fields are defined by a central corpus of falsifiable theory. Physics has the Standard Model -- it if's falsified, so is physics.
http://en.wikipedia.org/wiki/Standard_Model
Biology has evolution by natural selection. If it's falsified, so is biology.
http://en.wikipedia.org/wiki/Evolution
Psychology is not defined by a scientific theory, and it is therefore not falsifiable. It is therefore not a scientific field.
> "The DSM" is not "psychology".
Let's avoid a few intermediate steps and take this to its logical conclusion -- nothing is psychology. By contrast, the Standard Model, and relativity theory, are certainly physics, defining it and guiding its activities. Evolution by natural selection is certainly biology, defining it and guiding its activities.
When the existence of a luminiferous ether was falsified by the Michelson-Morley experiment in 1887 --
http://en.wikipedia.org/wiki/Michelson%E2%80%93Morley_experi...
-- it instantly falsified much of the physics of the day, and every physicist knew it. This threw physics into a turmoil until Einstein crafted a replacement (1905) --
http://en.wikipedia.org/wiki/Special_relativity
-- a testable, falsifiable theory that was in many ways better than its predecessor theory (able to define more and predict more). Had this not happened, had Einstein not produced a replacement theory, physics would have been abandoned. The reason? Physicists are thinking people, people who don't offer absurd rationalizations when their theories are falsified. People who wouldn't dream of calling an opinion a theory.
You think that it would be difficult to show that human behavior has sensitive dependence upon initial conditions? Given that behavior is the consequence of electrochemical activity in the brain, and the brain is an adaptive organ which is molded by every experience it is exposed to... how could you possibly avoid an extreme dependence on initial conditions?
> You think that it would be difficult to show that human behavior has sensitive dependence upon initial conditions?
No, I think it would be difficult to show that human behavior meets the formal definition of chaotic.
> ... how could you possibly avoid an extreme dependence on initial conditions?
To see the problem, imagine proving that human behavior is not sensitively dependent on initial conditions.
It's not about whether it's so, it's only about demonstrating it to the satisfaction of a scientist.
If science required double-blind trials, then you'd be throwing out a lot of physics, too (for example, the study of planetary motion). Science is 'robust reporting and observation', not 'double-blind trials'. The more robust your methods, the stronger your case - double-blind trials add a lot of robustness. Predictive value is also helpful, but again not required to be considered science (eg dark matter, which we still have no idea about).
On the contrary, that is exactly what it means. Science requires empirically testable, falsifiable theories backed up by empirical observations. If we only have observations, it's not science, because observations are neither particularly useful nor falsifiable.
> Predictive value is also helpful, but again not required to be considered science ...
Many commentators on this issue argue that an empirical, falsifiable theory that is able to make predictions about things not yet observed is a requirement for science.
http://en.wikipedia.org/wiki/Scientific_theory
Quote: "A scientific theory is a well-substantiated explanation of some aspect of the natural world that is acquired through the scientific method, and repeatedly confirmed through observation and experimentation.[1][2] As with most (if not all) forms of scientific knowledge, scientific theories are inductive in nature and aim for predictive power and explanatory force."
"The strength of a scientific theory is related to the diversity of phenomena it can explain."
Also, very important, observations (descriptions) can only be a preliminary to shaping an explanation, a testable theory (an explanation). Observations without a theoretical dimension are indistinguishable from stamp collecting and are a step toward science, but are not science themselves.
As far as I understand, it only limits professional diagnostics, shouldn't it be OK if it's not done as a practicing psychatrist (i.e. call it an analysis or interpretation and not a diagnostic)?
Imagine if computer security experts couldn't talk, in general, about the security of a line of products offered by a company, or about the probable cause of a specific issue that had been observed in a particular product.
The parallels ARE actually there; it's not as strange an analogy as it looks at first glance.
I understand some of the reasons the rule exists. I think this is probably not the best solution, in that it creates a public who have very strange ideas about psychiatry, and that is actually a great harm.
I'm neutral on most of this article, but like, specifically, if public perception is an issue, finding a better way to resolve the ethical dilemma that spawned the Goldwater Rule sounds like a pretty good idea to me.