[PhDwannabe]

So, reading another thread last evening I came upon a post by our forum member needimprovement which contained the following. Go ahead and read it carefully, if you will:

Quote

I'll preface my comments by a general remark: I'm a snob as far as science goes; science is formulating a theory, predicting results expected for that theory, doing the experiment to confirm the theory, so science is basically quantitative, physics and those disciplines which involve physics (including most of chemistry). For doing an experiment I'm willing to substitute making a measurement (as in astronomy, or geophysics, or molecular biology/genetics, physiology or neuro-physiology), so a bunch of other disciplines that people call science, I would call quasi-science: biology(other than the exceptions noted above), geology (other than geophysics). I wouldn't call psychology a science, because it lacks predictive theory, although it does do descriptions, and psychoanalysis is about on a par with witchcraft.

I wanted to talk about it, but I didn't want to take the topic of that thread too far afield. I messaged needimprovement about it, and he graciously agreed to discuss it in another post I'd here create, starting with the quote itself. The idea is that needimprovement will post below to expand on and clarify his thoughts--specifically, regarding psychology not being a science (though he may need to speak more broadly to make his points). Let me say a couple of things first--all or some of which may already be obvious to the screaming fans who already follow my every post:

1) As a doctoral student in clinical psychology, I find myself firmly on the other side. Psychological science is what I love, and what I would say I do all day.
2) As a cognitive-behaviorist, I also tend to agree with at least the spirit of the contention: "psychoanalysis is about on par with witchcraft." I don't tend to think of them as part of psychological science, so let's keep them out of it.
3) As a supporter of the movement within clinical psychology towards empirically supported treatments, I am by no means eager to defend clinical psychology's less scientific quarters (see point #2). I will be the first to admit that there are academics and practitioners in my field who don't think scientifically, don't do what I'd call quality science, or just aren't interested in it as a science. We have a lot of growing up to do as a professional guild--there are plenty of people who need to be thrown off the boat. Nonetheless, I don't think this embarrassing segment of the field makes psychology "not a science" any more than quack doctors call into question the idea of scientific medicine.
4) As somebody who has invested thousands of hours into something he calls a science, derives a lot of value and esteem from, and wishes to be his career, bias is unavoidable. I'm going to make a strong statement about it, directed towards needimprovement specifically: it is nearly impossible that I'll change my mind. This is bias at its highest level; I simply won't be open to a thundering and inviolable proof that psychology isn't a science as a layperson might be. This is called cognitive dissonance, and it's the same reason you won't generally convince a frat initiate that fraternities are a dumb idea--he just spent weeks getting paddled mercilessly and running naked around campus to join one. Needimprovement therefore is taking on the honorable task of debating someone who almost certainly will not concede. He is therefore free to regard any of my claimed victories as merely bias or close-mindedness on my part.

For people who are reading this, on either "side" of the "debate," or not on a side at all: if you would, I'd ask you to let myself and needimprovement to go at it for a couple of rounds before you begin posting on the thread. Let us have, say, five back-and-forths. Let the two of us get the ground of the debate established, and then have the peanut gallery join in if it wishes. Until then, make like a tennis audience, yeah?


So, needimprovement, my brave and worthy opponent, en garde! If you'd like to expand on and explain your quote above, sir, then I happily say: the floor is yours. What is a science?

[needimprovement]

I'll admit I was a bit harsh in that classification, but here's my definition of science: a general theory that can support sub-theories, that can be used to make predictions, which in turn can be verified by experiments or measurements--to either confirm or falsify the theory/model. So theory, quantitation, measurement are all necessary parts of this. These elements are present in physics and allied disciplines, and to lesser extent in other disciplines. Biology makes quantitative predictions in genetics and molecular biology, enzyme kinetics and some other parts, so to that extent it can be called scientific. Taxonomy and classical evolutionary biology I don't regard as all that scientific, since the quantitative element is missing. Geology is descriptive for the most part, putting aside geophysics, and to that extent is missing what I'd call an essential element. Psychology is missing a general theory (there seem to be lots of sub-theories, but nothing equivalent to what quantum mechanics or general relativity tells us about the world) and as far as I can see the quantitation is mainly in neuro-physiology. So I am probably closer to agreeing with you to consider all the IQ tests (and I'll include Personality Tests) to be total nonsense.

[PhDwannabe]

Thanks for your response, NI. Let me see if I understand some things right, and hopefully get some things clarified if I don't.

Quote

I'll admit I was a bit harsh in that classification

Why, that's quite alright. Let's agree to never let a reasonable discussion impinge upon our sense of drama, eh?

Quote

but here's my definition of science: a general theory that can support sub-theories, that can be used to make predictions, which in turn can be verified by experiments or measurements--to either confirm or falsify the theory/model.

It seems like you've defined science as pretty much exclusively consisting of what we might call a top-down process: a theory creates hypotheses or predictions, which you then design tests for, which provide data to either confirm or disconfirm the hypothesis, thus either lending weight to the theory or calling it into question. That seems to me to be a perfectly conventional way to test a theory, but not exhaustively descriptive of the scientific method or the scientific process. What about findings that are very much bottom-up?

When Hubble first realized that everything was looking a bit reddish, and all of those galaxies were flying away from one another, was he attempting to find support for his theory that the universe began with an enormous expansion event 13.8 billion years ago? I don't think that he was. When John Snow realized that all of those poor cholera victims were getting their water from the same pump on Broad Street, was he trying to find empirical support for the germ theory of illness? No--he was a skeptic of miasma theory and had the good sense to draw a map. While certain parts of the model of hypothetico-deductive reasoning typically occur in order--you form a hypothesis, you design a test, you gather the data and see if they support it--the placement of the theory has, I think, much less of a proscribed placement. I think that all sciences reason in both directions. Sometimes, amazing results come from studies which are trying to confirm a theory. Sometimes, theories are born and adjusted based on results that don't make sense, or on basic science which may be only loosely informed by theories.

Quote

So theory, quantitation, measurement are all necessary parts of this.

For the purposes of the discussion, I'll take this as settled. I think plenty of forum members may disagree--particularly with the way that you have decided whether or not certain sciences employ methods which meet your definitions of these three. I'll refrain from talking about, for instance, classical evolutionary biology, and focus on psychology, because I think what I call the science of psychology can easily meet these three criteria. With regard to your view of how psychology meets or doesn't meet these three:

Quote

Psychology is missing a general theory (there seem to be lots of sub-theories, but nothing equivalent to what quantum mechanics or general relativity tells us about the world) and as far as I can see the quantitation is mainly in neuro-physiology.

So, from what I understand, in your tripartite system, you believe that psychology: 2) generates hypotheses/predictions, as well as 3) makes measurements. You don't believe that it 1) has a "general theory." This is what makes it not a science. I hope I'm clear on that.

Just above, I said I'd take your tripartite criteria as settled, but what I can't take as settled is what you have decided "counts" and "doesn't count" for the first criterion. You say that it psychology has a lot of "sub-theories," but needs a "general theory." Since when must a science be so unified? What exactly makes a theory "sub," and what makes a theory "general?" Do a certain percentage of scientists in a field have to agree upon it? Do a certain number of papers informed by it have to be published every year? The closest you come to answering this question is when you say that psychology lacks a theory that is "equivalent to what quantum mechanics or general relativity tells us about the world." Should I take this to mean that a theory can only be "general"--big enough to count, as it were--if it tells us a large amount of stuff about the world? Well, if that is so, how much is enough? Isn't it difficult to compare the apple of what general relatively tells us about the world to the orange of what evolution by natural selection tells us, to the banana of what I think cognitive-behavioral theory tells us about the world? Who, exactly, gets to decide?

Let me give you an example of a theory that I think tells us a lot. As I said in my first post, I'm a cognitive-behaviorist--I largely subscribe to a theory relevant to the clinical practice of psychology called cognitive-behaviorism. In (very) short, the theory proffers that emotions--both negative and positive--are produced by cognitive evaluations and interpretations of events. It also describes, on several different levels, the ways that this occurs. This is a very large theory of human behavior, cognition, mental illness, and treatment. If the theory is correct, then altering those cognitions should alter the emotional experience of a person. To test something like that, we'd have to set up a study, of course. Suppose we exposed a bunch of individuals who say they are depressed or anxious to very specific interventions designed to adjust cognitions that the theory suggests might be relevant for depression or anxiety. Then, we measure outcomes--not only by just asking them, but by physiological measures of anxiety, by physiological indices known to be associated with depressed mood, and by yet more objective outcome indicators like suicide rates, occupational success, physical health, academic achievement, admissions to hospitals. And what if those things generally turn out better after the treatment? Not perfect, mind you, but a lot better?

Of course, that doesn't confirm our theory yet--we are skeptical, and constantly fret that a million other things might be going on. What if it's just the placebo effect of therapy rather than our fancy interventions--the mere fact that they're sitting and talking to somebody an hour or two a week? Good question. We'll create a placebo group that essentially gets supportive counseling (e.g., "hang in there, buddy.") What if the people who seek this out are weird, somehow--predisposed to do better, because those destined to recover well are the sort who're also interested in this sort of thing? Then we'll not assign them randomly to treatment or control groups, we'll do comprehensive checks of their demographic characteristics to see if they're different from the normal population. What if it only works for certain disorders? That'd certainly be a hole in a theory that seeks to explain the relationship between cognitions and emotions in general. Then I guess we'd better start checking them all.

And when we do, what do we find? Well, it still works. Not only does it work, it works as well or better than medication, a finding seen over and over again for the last twenty years, thirty years or more.

What I'm describing is the process of designing gold standard measures of therapeutic outcome, called randomized controlled trials. The hypotheses of these trials are informed by a massive theory which seeks to explain a large chunk of human experience. We think that it does indeed explain that chunk because of what our data seem to tell us. If you're curious about it, just Google randomized controlled trials for cognitive behavioral therapy, and check it out.

I understand that a lot of people may not realize that this is what many academic and practicing psychologists do with their time. This is the way that good practice is developed. What about this process doesn't sound like it's informed by theory? What about this doesn't sound like science?

I await your response.

[needimprovement]

PhDwannabe, on 20 August 2010 - 04:27 AM, said:

It seems like you've defined science as pretty much exclusively consisting of what we might call a top-down process: a theory creates hypotheses or predictions, which you then design tests for, which provide data to either confirm or disconfirm the hypothesis, thus either lending weight to the theory or calling it into question. That seems to me to be a perfectly conventional way to test a theory, but not exhaustively descriptive of the scientific method or the scientific process. What about findings that are very much bottom-up?

I'll be willing to agree that the idealized model I have is not descriptive always (indeed, not most of the time) about how scientists work...but that's a different issue from what science is.

PhDwannabe, on 20 August 2010 - 04:27 AM, said:

When Hubble first realized that everything was looking a bit reddish, and all of those galaxies were flying away from one another, was he attempting to find support for his theory that the universe began with an enormous expansion event 13.8 billion years ago? I don't think that he was. When John Snow realized that all of those poor cholera victims were getting their water from the same pump on Broad Street, was he trying to find empirical support for the germ theory of illness? No--he was a skeptic of miasma theory and had the good sense to draw a map. While certain parts of the model of hypothetico-deductive reasoning typically occur in order--you form a hypothesis, you design a test, you gather the data and see if they support it--the placement of the theory has, I think, much less of a proscribed placement. I think that all sciences reason in both directions. Sometimes, amazing results come from studies which are trying to confirm a theory. Sometimes, theories are born and adjusted based on results that don't make sense, or on basic science which may be only loosely informed by theories.

I'll agree to that, but I'm not talking about process. How one achieves faith has varied accounts, but faith itself is one thing.

PhDwannabe, on 20 August 2010 - 04:27 AM, said:

For the purposes of the discussion, I'll take this as settled. I think plenty of forum members may disagree--particularly with the way that you have decided whether or not certain sciences employ methods which meet your definitions of these three. I'll refrain from talking about, for instance, classical evolutionary biology, and focus on psychology, because I think what I call the science of psychology can easily meet these three criteria. With regard to your view of how psychology meets or doesn't meet these three:

It isn't black and white, but a matter of degree...

PhDwannabe, on 20 August 2010 - 04:27 AM, said:

So, from what I understand, in your tripartite system, you believe that psychology: 2) generates hypotheses/predictions, as well as 3) makes measurements. You don't believe that it 1) has a "general theory." This is what makes it not a science. I hope I'm clear on that.

pretty much, although again, the criteria are simplified for purposes of discussion.

PhDwannabe, on 20 August 2010 - 04:27 AM, said:

Just above, I said I'd take your tripartite criteria as settled, but what I can't take as settled is what you have decided "counts" and "doesn't count" for the first criterion. You say that it psychology has a lot of "sub-theories," but needs a "general theory." Since when must a science be so unified? What exactly makes a theory "sub," and what makes a theory "general?" Do a certain percentage of scientists in a field have to agree upon it? Do a certain number of papers informed by it have to be published every year? The closest you come to answering this question is when you say that psychology lacks a theory that is "equivalent to what quantum mechanics or general relativity tells us about the world." Should I take this to mean that a theory can only be "general"--big enough to count, as it were--if it tells us a large amount of stuff about the world? Well, if that is so, how much is enough? Isn't it difficult to compare the apple of what general relatively tells us about the world to the orange of what evolution by natural selection tells us, to the banana of what I think cognitive-behavioral theory tells us about the world? Who, exactly, gets to decide?

I guess I'm a reductionist. To be esthetically pleasing, scientific theories should proceed from the most general to the particular. This means chemistry comes from physics, biochemistry from chemistry, biology from biochemistry and (in principle) psychology from biochemistry.

PhDwannabe, on 20 August 2010 - 04:27 AM, said:

Let me give you an example of a theory that I think tells us a lot. As I said in my first post, I'm a cognitive-behaviorist--I largely subscribe to a theory relevant to the clinical practice of psychology called cognitive-behaviorism. In (very) short, the theory proffers that emotions--both negative and positive--are produced by cognitive evaluations and interpretations of events. It also describes, on several different levels, the ways that this occurs. This is a very large theory of human behavior, cognition, mental illness, and treatment. If the theory is correct, then altering those cognitions should alter the emotional experience of a person. To test something like that, we'd have to set up a study, of course. Suppose we exposed a bunch of individuals who say they are depressed or anxious to very specific interventions designed to adjust cognitions that the theory suggests might be relevant for depression or anxiety. Then, we measure outcomes--not only by just asking them, but by physiological measures of anxiety, by physiological indices known to be associated with depressed mood, and by yet more objective outcome indicators like suicide rates, occupational success, physical health, academic achievement, admissions to hospitals. And what if those things generally turn out better after the treatment? Not perfect, mind you, but a lot better?

When I say a discipline is not scientific, I'm not saying that what might be done with the discipline is not valuable both pragmatically and esthetically. I don't regard Oliver Saks as a scientist, but his insights and accounts are extremely interesting and valuable. Economics isn't a science, but it can provide insights and utility. I have a strong prejudice against psychology engendered by phony therapies applied to family members, phony assessments (the Briggs--whatever personality test applied in my 3rd year of Ecclesial Lay Ministry training as part of a "management" course, phony attempts to analyze mass behavior on the basis of meaningless terms). Were psychology a science, these phony parts could be swept into the dustbin as were discredited theories in physics and chemistry (e.g. cold fusion, polywater).

PhDwannabe, on 20 August 2010 - 04:27 AM, said:

Of course, that doesn't confirm our theory yet--we are skeptical, and constantly fret that a million other things might be going on. What if it's just the placebo effect of therapy rather than our fancy interventions--the mere fact that they're sitting and talking to somebody an hour or two a week? Good question. We'll create a placebo group that essentially gets supportive counseling (e.g., "hang in there, buddy.") What if the people who seek this out are weird, somehow--predisposed to do better, because those destined to recover well are the sort who're also interested in this sort of thing? Then we'll not assign them randomly to treatment or control groups, we'll do comprehensive checks of their demographic characteristics to see if they're different from the normal population. What if it only works for certain disorders? That'd certainly be a hole in a theory that seeks to explain the relationship between cognitions and emotions in general. Then I guess we'd better start checking them all.

And when we do, what do we find? Well, it still works. Not only does it work, it works as well or better than medication, a finding seen over and over again for the last twenty years, thirty years or more.

I don't contend that therapeutic practices don't (sometimes) work...I'm contending that there isn't a general law, quantitatively described,i.e with attributes that can be measured quantitatively, to tell whether a practice will work. Accordingly, psychology is not, in my view, a science. What you and others do is very probably valuable. But it doesn't negate that a phony outfit such as (to pick a name out of the hat) Sierra Tucson can take in millions from duped celebrities and not achieve therapeutic results.

PhDwannabe, on 20 August 2010 - 04:27 AM, said:

What I'm describing is the process of designing gold standard measures of therapeutic outcome, called randomized controlled trials. The hypotheses of these trials are informed by a massive theory which seeks to explain a large chunk of human experience. We think that it does indeed explain that chunk because of what our data seem to tell us. If you're curious about it, just Google randomized controlled trials for cognitive behavioral therapy, and check it out.

What is that massive theory? I will Google the site, but again; if this theory were scientifically valid, then it would be the only theory that psychologists/therapists followed, because it would be possible to falsify the other theories.

PhDwannabe, on 20 August 2010 - 04:27 AM, said:

I understand that a lot of people may not realize that this is what many academic and practicing psychologists do with their time. This is the way that good practice is developed. What about this process doesn't sound like it's informed by theory? What about this doesn't sound like science?

I think that is just my point. People don't realize this is what's done as good practice, because it is not possible to show other practice is false. Falsifiability is an essential element of science (see Karl Popper).

PhDwannabe, on 20 August 2010 - 04:27 AM, said:

I await your response.

Clarification - I am not trying to denigrate what you do in your profession. It sounds, from what you've written, to be a very worthwhile endeavor.. But the problem is that the other parts of psychology/psychiatry which are nonsense are still there and haven't been shown to be nonsense by scientific type measures. That's the basis of my judgment.
I've enjoyed this discussion, learned quite a bit (I'll follow up on your reference), and have modified my stark views (to a degree).

[PhDwannabe]

Quote

I guess I'm a reductionist. To be esthetically pleasing, scientific theories should proceed from the most general to the particular. This means chemistry comes from physics, biochemistry from chemistry, biology from biochemistry and (in principle) psychology from biochemistry.

Although many things that are true are beautiful, I'm not sure if we should decide what gets the "science" stamp and what doesn't based on what pleases us aesthetically.

Quote

When I say a discipline is not scientific, I'm not saying that what might be done with the discipline is not valuable both pragmatically and esthetically.

Well, I can honestly say that I don't believe that you're saying that. Believe me, I take no offense or affront at your views about my discipline.

Quote

I have a strong prejudice against psychology engendered by phony therapies applied to family members, phony assessments (the Briggs--whatever personality test applied in my 3rd year of Ecclesial Lay Ministry training as part of a "management" course

I have the same disgust with ridiculous therapies and assessments which I think lack empirical support--like the Meyers-Briggs Type Indicator that you mentioned, which to me has all of the clinical/scientific utility of the parlor game that it is. But, at least for myself, I wouldn't call that "prejudice"--my regard for these things is informed by the empirical findings relevant to them, or lack thereof.

Quote

Were psychology a science, these phony parts could be swept into the dustbin as were discredited theories in physics and chemistry (e.g. cold fusion, polywater).

Is this really a fair way to define a discipline as a science? Its success rate in sweeping the junk science that would claim that discipline as its support into some dustbin? But even taking as settled for the moment, look--they are swept into the dustbin. You'll find about as many academic psychologists who believe in, say, past-life regression therapy or "rebirthing" interventions as you do physicists who are working diligently on cold fusion. These things are discredited and laughed at in my discipline. Are we expected to police the thousands of hucksters who co-opt the language of our science in order to support those fantasies? There are plenty of people who call themselves physicists, use the language of physics--maybe even have degrees--and promote unsubstantiated, unscientific madness. Physics does not become "not a science" because "real" physicists don't bother to hunt them down and censor them.

Quote

I don't contend that therapeutic practices don't (sometimes) work...I'm contending that there isn't a general law, quantitatively described,i.e with attributes that can be measured quantitatively, to tell whether a practice will work. Accordingly, psychology is not, in my view, a science.

I appreciate that you're not saying that they don't work--it wouldn't matter if you were, since they, of course, do. But I just named a bunch of attributes that can be measured quantitatively in my discussion of randomized controlled trials. Are you here contending that "the proportion of individuals completing this course of therapy who go on to commit suicide" is not a quantitative measure? It may not be a "hard," physical property of the universe like mass, but that doesn't mean it is not measured quantitatively. The quantitative statistical methods we use to examine variables like that are not really often very different than what is used in, say, biochemistry.

Quote

What is that massive theory? I will Google the site, but again; if this theory were scientifically valid, then it would be the only theory that psychologists/therapists followed, because it would be possible to falsify the other theories.

I just told you what the massive theory was! Cognitive-behaviorism! And its validity is determined by its exclusive use by every single member of a discipline? Since when is that how we decide whether or not a theory is valid? I thought we used a logico-deductive reasoning process to do that?

Quote

I think that is just my point. People don't realize this is what's done as good practice, because it is not possible to show other practice is false. Falsifiability is an essential element of science (see Karl Popper).

Cognitive-behaviorism is a gigantic theory that informs tons of what psychologists study and do, and it is falsifiable. Much of the junk psychology that I don't believe in and almost no academic psychologists take seriously is often not falsifiable. That's why it's junk. Why does the presence of that junk bear on the scientific nature of the parts of psychology that are done seriously, rigorously, according to the scientific method? Look, it's not possible for an astrophysicist to show that the egg-laying space unicorn theory of stellar formation is false, either. That doesn't make astrophysics "not a science." The presence of a billion people calling themselves astrophysicists and preaching the egg-laying space unicorn theory of stellar formation wouldn't make astrophysics "not a science" either.

In my last post, I described, as an example, the process of creating randomized controlled trials to test theoretical assumptions of cognitive-behaviorism. I provided quite a few details of how that process worked and what it tends to find, and asked a big question:

Quote

What about this doesn't sound like science?

I'm still unclear on your response to that precise question. If the people creating these trials aren't doing science, what exactly are they doing?

[needimprovement]

PhDwannabe, on 21 August 2010 - 06:12 PM, said:

Although many things that are true are beautiful, I'm not sure if we should decide what gets the "science" stamp and what doesn't based on what pleases us aesthetically.

An important criterion for a physical theory is elegance, which is a shorthand term for universality, conciseness, and beauty. P.A.M. Dirac, the physicist who gave a unified picture for quantum mechanics and whose theoretical developments were the foundation for modern quantum mechanics, was all for beauty:"it is more important to have beauty in one's equations than to have them fit experiment (not sure I'd agree here)...it seems that if one is working from the point of view of getting beauty in one's equations and if one has a really sound instinct, one is on a sure line of progress."

PhDwannabe, on 21 August 2010 - 06:12 PM, said:

Well, I can honestly say that I don't believe that you're saying that. Believe me, I take no offense or affront at your views about my discipline.

I wasn't trying to be nice. My point-of-view has changed through these exchanges. Right now, rather than thinking in black-and-white terms, with physics (and physics derived disciplines) white and everything else black, I'd put it more in shades of gray (or maybe with a color palette to symbolize those three attributes I've mentioned for scientific endeavor).

PhDwannabe, on 21 August 2010 - 06:12 PM, said:

I have the same disgust with ridiculous therapies and assessments which I think lack empirical support--like the Meyers-Briggs Type Indicator that you mentioned, which to me has all of the clinical/scientific utility of the parlor game that it is. But, at least for myself, I wouldn't call that "prejudice"--my regard for these things is informed by the empirical findings relevant to them, or lack thereof.

I'm glad we agree on that.

PhDwannabe, on 21 August 2010 - 06:12 PM, said:

Is this really a fair way to define a discipline as a science? Its success rate in sweeping the junk science that would claim that discipline as its support into some dustbin? But even taking as settled for the moment, look--they are swept into the dustbin. You'll find about as many academic psychologists who believe in, say, past-life regression therapy or "rebirthing" interventions as you do physicists who are working diligently on cold fusion. These things are discredited and laughed at in my discipline. Are we expected to police the thousands of hucksters who co-opt the language of our science in order to support those fantasies? There are plenty of people who call themselves physicists, use the language of physics--maybe even have degrees--and promote unsubstantiated, unscientific madness. Physics does not become "not a science" because "real" physicists don't bother to hunt them down and censor them.

Well, as I said above, I'm now thinking in shades of gray. I don't think the situation is all that common that physicists in practice promulgate false theories that are not discredited within a very short time. On the other hand, when my professor in college uses the briggs--whatever test on hapless students in a management course (this was 11 years ago) some time after the theory had been in use, it means that people in subsidiary disciplines don't get the message from reputable psychologists that it's no good. Now maybe this is due to the "public's" lack of discernent about who to trust, but let me ask: if there are official organizations for psychology, do they, after some suitable period of time, call such theories discredited?

At any rate, I'm grateful that you've bothered to educate me and sandpaper away some of my prejudices.

PhDwannabe, on 21 August 2010 - 06:12 PM, said:

I appreciate that you're not saying that they don't work--it wouldn't matter if you were, since they, of course, do. But I just named a bunch of attributes that can be measured quantitatively in my discussion of randomized controlled trials. Are you here contending that "the proportion of individuals completing this course of therapy who go on to commit suicide" is not a quantitative measure? It may not be a "hard," physical property of the universe like mass, but that doesn't mean it is not measured quantitatively. The quantitative statistical methods we use to examine variables like that are not really often very different than what is used in, say, biochemistry.

PhDwannabe, on 21 August 2010 - 06:12 PM, said:

I just told you what the massive theory was! Cognitive-behaviorism! And its validity is determined by its exclusive use by every single member of a discipline? Since when is that how we decide whether or not a theory is valid? I thought we used a logico-deductive reasoning process to do that?

Cognitive-behaviorism is a gigantic theory that informs tons of what psychologists study and do, and it is falsifiable. Much of the junk psychology that I don't believe in and almost no academic psychologists take seriously is often not falsifiable. That's why it's junk. Why does the presence of that junk bear on the scientific nature of the parts of psychology that are done seriously, rigorously, according to the scientific method? Look, it's not possible for an astrophysicist to show that the egg-laying space unicorn theory of stellar formation is false, either. That doesn't make astrophysics "not a science." The presence of a billion people calling themselves astrophysicists and preaching the egg-laying space unicorn theory of stellar formation wouldn't make astrophysics "not a science" either.

In my last post, I described, as an example, the process of creating randomized controlled trials to test theoretical assumptions of cognitive-behaviorism. I provided quite a few details of how that process worked and what it tends to find, and asked a big question:

I'm still unclear on your response to that precise question. If the people creating these trials aren't doing science, what exactly are they doing?

You've made some good points, and I'll concede, without reading much else about it, that cognitive/behaviorism may be an excellent theory for people's behavior, but without further reading I'm not sure how much it goes into the general problem of consciousness and "mind". Ideally, chemistry can be developed (and is, partially) from the more general --quantum mechanics. What I'm thinking is that psychological theories of how people behave are sort of like chemistry without a quantum mechanical foundation, i.e. like chemistry was 150 years ago. But even then most people called chemistry a science, so to be fair I guess I'd have to bend and call psychology a science even though its foundations have yet to be discovered.

[PhDwannabe]

NI, I'll take your gracious concession, and I'll make a few gracious concessions of my own:

You bring up a good point--as you've brought up in some form several times, with this:

Quote

I don't think the situation is all that common that physicists in practice promulgate false theories that are not discredited within a very short time. On the other hand, when my professor in college uses the briggs--whatever test on hapless students in a management course (this was 11 years ago) some time after the theory had been in use, it means that people in subsidiary disciplines don't get the message from reputable psychologists that it's no good.

In my psychopathology class--a course that pretty much everybody in clinical psychology endures the first semester of graduate school--my professor was one day discussing various theories on the etiology of some particular mental illness I can't remember. Some theories we chuckled at far more than others. He noted with some vexation that, well, "that's the thing about these theories. They don't seem to ever just go away!" Even though the bulk of the field doesn't take them seriously, they somehow linger on longer than they should. As much as I'll make the case that psychology is scientific, that the majority of us are employing testable, falsifiable, rigorous theory to our work, it also seems that every bit of junk my discipline has ever believed is still believed by at least some small group in the profession. There are Jungians hanging around the halls of psychology in an amount--though very small--that dwarfs the number of chemists who subscribe to phlogiston theory. We don't seem to be able to trash the junk as completely as other sciences can, and do. There are a couple of things to be said about this--some kind to my discipline, some not...

1) Give us time! We're a young science. You can date psychology/psychiatry pretty much from Charcot's establishment of "hysteria" as a mental illness in the middle-to-late 19th century. Psychotherapy as we know it began with Freud, a little over a hundred years ago. I might say that other sciences take a while to figure out and solidify their big theories--it was almost a century between Kepler's laws of planetary motion described elliptical orbits and Newton's universal gravitation finally explained them. Wholesale acceptance of this new astronomical system didn't occur overnight. Even widespread acceptance of evolution by natural selection didn't just smack the natural sciences in the face immediately after Origin of Species, and why start there, anyway? Chuck's grandpa, Erasmus Darwin was talking about evolution in the late 18th century--there had been a rising scientific debate about the common descent of animals for a century by the time the big book we all know was published.

2) Give us a break! Human behavior is maddeningly complex, and human beings are superbly difficult to study in ways that rocks, heavenly bodies, finches, and nucelotide base pairs are not. These parts of nature can be interrogated directly--you stick a ruler next to them, put them on a scale, make them make a bunch of copies of themselves inside a PCR tube. The mind, on the other hand, is not measurable (though the brain is beginning to be.) We have to look at the mind's activity in this world: behavior. And even then, there only have two ways of doing that: either I have to observe it, or I have to ask people about it (this is called "self-report"). And wouldn't you know it, behavior changes when you look at it. People forget, and lie, and are biased and silly when you ask them about it. And when we circumvent all of that self-report nonsense and watch behavior, how do we make sense of it? "Does that guy look angry to you? He's tapping his foot. Is he nervous or pissed?" What I'm talking about is the need to code behavior--to operationalize a variable and then track it systematically. The complex research designs I've spent years learning about are, among other things, elaborate ways of getting around all of this insanity. These things aren't impossible to do, but they're really, really hard! Also, we're limited by technology, feasibility, and ethics in really studying psychological phenomena in the ways that would really tell us a lot. I can't forcibly separate twins at birth and assign them randomly to parents. If I could, I'd be able to tell you a hell of a lot more about this so-called "nature vs. nurture" thing. I can't follow someone around with a secret camera and watch all of their behavior 24/7. If I could, I wouldn't have to bother spending years developing reliable and valid self-report measures.

Look, I know that all scientists could sit around like the Four Yorkshiremen and complain about how bad they have it. The astrophysicist might tell me that I'm lucky to have a research subject closer than a million light-years away. The evolutionary biologist would say that I'm lucky my subject hasn't been dead for fifty million years, and the biochemist might tell me that he'd love to not need a multimillion-dollar grant for equipment to examine his subject at all. All quite true. But hell, allow me a little bit of sentimentality: humans are nuts! They're unpredictable, and they do all sorts of insane things--accidentally and on purpose--when you try to study even the most basic stuff about their minds, thoughts, emotions, and behavior. The complexities of trying to investigate them with the scientific method are legion. It's doable, but it's a really crazy task. It's therefore a lot less straightforward to cart a theory out of the dumpster. There are simply more questions in this science than in others about the integrity and generalizability of research. We've got all of the standard questions, plus issues of sampling, coder bias, integrity of measures, a billion confounding variables, the meaningfulness of control and comparison groups, fair and valid operationalization of variables, the fidelity of intervention delivery and reception... I could go on. I can't show a Freudian my numbers and expect him to scratch his head and say, "well, damn--you're right." There are a thousand other questions he'll ask me first, many of which haven't yet been addressed by my empirical literature. (Nonetheless, I continue to hold that it's still science, even if results aren't as clear-cut as I'd like them to be.)

3) We can do better. We are doing better, I think--the neuroscience revolution is prompting a lot of people in my field to take a hard look at the hard-scientific rigor of their study (often because they want a piece of the sexy new neuroscience action.) The attitude of the cognitive-behaviorists over the last several decades--"hell yes you can test therapy outcomes; you can and you should!"--has broken some of the sacred bonds of the therapy room that, for much of clinical psychology's history, didn't allow in a lot of empirical investigation. Even the psychodynamic folks--whom I regard as fairly goofy anyway--want their own outcome studies now. The difficulty seems to be that, since a lot of the earlier theories frequently lack scientific rigor and falsifiability, how are we supposed to definitively prove them wrong? I don't think I can conclusively show a Freudian that classical Freudian psychodynamic theory is wrong any more than I can really "disprove" Christianity to a priest. Because they're both religions.

Nonetheless, my discipline, and the professional guilds which oversee it--the APA in particular--may be lacking in a few things. As I said in my first post on this thread: there are plenty of people in psychology who need to be thrown off the boat. They don't do science, they aren't interested in science, and they don't deserve the title. I'm not alone in believing this. There is a parallel organization called the APS which tends to stick more rigorously to these values--although their own historical information wouldn't let you know this, the organization was largely founded by people who were disgusted at the APA's unwillingness to enforce more scientific standards, and jumped ship. I could sit around and debate all day about whether it'd be more useful to form another club or advocate from within, the bottom line is that standards need to get better. Some of that push can and should come from inside the discipline. It enrages most of us to no end that Dr. Phil is allowed to present himself as some sort of mental health professional, and calls himself a psychologist just like I will. But then again, hey, Deepak Chopra's a "doctor," right?


So, there are my defenses and my concessions. Again, I was happy to have had the debate. Hopefully it has informed some others, or piqued some others' interest as well.

At this point, I'll say: whoever you are out there, if you want to chime in now, chime in!

[pioneer]

Psychology is concerned with analyzing the internal data generated by the mind/brain. As an analogy, say a new strain of bacteria was generating new complex chemicals. The scientist would analyze the chemical output and then try to discover what enzymes and genes were responsible. The layman, who does know much about biology might say the gas part of the output stinks and that is it. But the scientist will see significance even in this unique stink and want to know what it is.

The mind/brain also generates output, such as dreams. This output starts inside the brain and creates data that is observed from within. This output is not as simple as a chemical output, since it has many layers that go down to chemistry-biology, all the way to complex and even organized patterns. The psychologist will try to trace the source. There are no brain reading machines to make this tracing easier, so they have rely on people who witness the output and report it.

Say our new bacteria was not only generating chemicals, but once these complex chemical formed, these chemicals arrangement themselves in geometric patterns, which morph as time goes on. This is not so unlikely with bacteria output. However, with the mind/brain, electro-chemical signals are not just being generated, but the final output of this complex generation can form complex sensory patterns which can take on almost sensory shape, we know, and even beyond. This internal data is coming from a natural source, and its output is just as natural as bacteria. The layman may think this the dream data stinks since they can't see any value. But to the skilled eye it tells us things. from which we can infer many things, some of which can used for medical treatment.

This post has been edited by pioneer: 24 August 2010 - 11:51 PM

[Mr Skeptic]

I think I would classify psychology as a "soft science", where the "soft" part is because of the difficulty in making rigorous predictions due to the complexity of the system being studied. Whereas in the "hard" sciences making a wrong prediction trashes your theory, in psychology a wrong prediction could easily be due to bad luck in your sample. While this sort of problem could theoretically be fixed by increasing the sample size, it turns out that humans are expensive to experiment with, so sample sizes must be small despite the high probability of getting poor results. And so, while many psychologists try their best to do science, it really isn't all that obvious if someone is instead just making up a bunch of BS.

This will eventually change as the science matures and as more technology becomes available (see work on mind/machine interface, neural net simulations such as the Blue Brain Project, etc) that will give us more direct access to the brain.

[DJBruce]

Is psychology a science? Absolutely. Is psychology a pure of science? No. Psychology by all means meets the definition of science. It seeks to explain the human mind and human behavior through a systematic and logical process of collecting evidence and observations. There are those areas of psychology, which are crackpoty, but all areas of science have those areas, and those who do dabble in these areas are outside of the mainstream community.

Although psychology is a science I do not believe it is a very pure science. The more fundamental the questions, which a field of science seeks to answer the more pure the field. For example, I would consider physics to be purer than biology as the overarching questions in physics are more fundamental than those in biology. So on the scale of purity I would say that ranking the major fields is:

1) Physics
2) Chemistry
3) Biology
4) Psychology

[pioneer]

The brain/mind generates internal data. Common sense says you need to observe internal data, internally, not externally.

One would not collect external data by looking inside their imagination. This makes as much sense as collecting internal data from the outside. From the outside there will always be a difference between what is actually occurring in the mind and what the evolving machines can pick up. This means external theory will always be slightly off from internal reality.

From the outside is like looking at an alien computer running a program but without a monitor. We might try to use a huge NMR to look inside the programming. From the inside is like adding a cheap monitor and a set of speakers so we can see what the computer is actually generating.

What you would need is someone who is trained in existing theory, so they are somewhat prepared for what they may encounter. This gives them enough background to be able to make as many objective observations as possible, based on what we know and expect. Then we lower them down into the deep. Their diving bell is analogous to their support theory. It may be designed to work better in shallow water and might start to leak under the higher pressure. This is the fear. The resourceful expert may be able to make some repairs in the field. If not, we pull our expert diver up, retool and try again. As the diving bell design improves, we eventually make it all the way to the bottom. Machine will helpful in the shallows, but without ever seeing the deep, even if you can see signals from the surface, we may not know what the signals mean, until we go down to take a look.

This post has been edited by pioneer: 25 August 2010 - 01:24 AM

[Ringer]

I would also like to say that despite what a lot of people seems to think psychology isn't just sitting in a room and talking to people. There are many different ways to experiment with psychology and how to test the results as reliable. An example is using an fMRI to watch the process the brain goes through when doing certain tasks and seeing if those processes happen for most people most of the time. There seems to be folk knowledge assuming that most, if not all, psychological studies are only case studies or surveys. This is not the case.

Also, I wouldn't say that psychology as a whole has a tough time throwing out strangely prevalent ideas. Clinical seems to have this problem more than others if only because of the large amount of people interested in keeping the practices they've been using for a long time and not have people believe they have been ripped off. There are major areas of psychology that don't have a lot of problems changing or reforming old ideas.

[ccdan]

PhDwannabe, on 19 August 2010 - 03:13 PM, said:

3) As a supporter of the movement within clinical psychology towards empirically supported treatments,

like almost all psychologists and psychiatrists, you seem to have the same terribly wrong urge: you're in a rush for solutions without being sure that there is any problem whatsoever!

psychiatry has two tools that deal with the supposed problems:

1. persuasion (also, called psychotherapy)
2. drugs

the success of the first one depends on a variety of factors, among which, the intelligence of both the "patient" and the "psychotherapist"

drugs on the other hand, tend work with "anyone" although it really depends what one means by "drugs work" ... you certainly won't find any drug that would specifically "treat" agoraphobia, although you can certainly can find some drugs that would give a person the courage to confront any animal they're afraid of, but that can prove deadly eventually... the problem with drugs and psychiatric treatments in general is the fact that they can be used to "treat" virtally any undesired behaviour/emotion by drugging people to the point that they no longer show the "symptopms" (actually the unwanted behaviors, no matter what they are and how "harmful" they are)
even the belief that psychiatry and/or psychology are scientific, can be "treated" with drugs...

so, before getting to the subject of "treatments", perhaps you should look into scientific nature of what some call "mental illnesses" and questions like the following should eb answered without the slightest hesitation: what is the concrete proof that a certain supposed "mental illnes" (like ADHD) is a genuine, verifiable "malfunction" of the brain and not just something "not liked" by some who like to call themselves "psychiatrists?" why is it that certain "malfunctions" of the brain are "treated" by neurology (like alzheimer, brain tumors) and "others" are "treated" by pychiatry?

This post has been edited by ccdan: 28 August 2010 - 08:43 PM

[DJBruce]

ccdan: Ignoring the fact that I believe most of what you have written is completely baseless,a nd wrong. How does any of that make psychology not a science?

This post has been edited by DJBruce: 28 August 2010 - 08:54 PM

[PhDwannabe]

Quote

like almost all psychologists and psychiatrists, you seem to have the same terribly wrong urge: you're in a rush for solutions without being sure that there is any problem whatsoever!

How have you decided that I'm not sure there's a problem? Could you point to one of my statements that would let you know that I am inappropriately unsure that there is a problem?

Also, I'd love if you could address DJBruce's comparably excellent question:

Quote

How does any of that make psychology not a science?

I'll answer a few of your questions as well. Consider that a down payment on the wonderful answers I'm dying to hear about.

Quote

what is the concrete proof that a certain supposed "mental illnes" (like ADHD) is a genuine, verifiable "malfunction" of the brain

There are two separate answers to this question, both of which I think are just about equally valid:
1) fMRI studies
2) the concept of mental illness need not be grounded in a "genuine, verifiable 'malfunction' of the brain." Social behavior, for instance, as a product and activity of the brain--a phenomenon which is certainly a product of that brain's activity, but not always reducible (nomologically or logically) to that activity.

Quote

and not just something "not liked" by some who like to call themselves "psychiatrists?"

Well, I don't call myself a psychiatrist, nor will I ever, so I can't be too sure who those who 'like to call themselves' psychiatrists think. But maybe you should, I don't know, shot in the dark: read some psychology research and see how these vicious beasts proffer to know what a mental illness is. Maybe then we can have a rational discussion about that process. Or maybe we can just say random opinions that come to our heads about things we don't know about. I think, for instance, on one of Saturn's moons there has to be an alligator named Phil. Just swimmin' around down there!

And, as a bonus, yet another question of my own:

Quote

drugs on the other hand, tend work with "anyone" although it really depends what one means by "drugs work" ... you certainly won't find any drug that would specifically "treat" agoraphobia, although you can certainly can find some drugs that would give a person the courage to confront any animal they're afraid of, but that can prove deadly eventually... the problem with drugs and psychiatric treatments in general is the fact that they can be used to "treat" virtally any undesired behaviour/emotion by drugging people to the point that they no longer show the "symptopms" (actually the unwanted behaviors, no matter what they are and how "harmful" they are)
even the belief that psychiatry and/or psychology are scientific, can be "treated" with drugs...

We're talking about psychology. Why are you talking about drugs?

This post has been edited by PhDwannabe: 28 August 2010 - 09:10 PM

[pioneer]

Humans are capable of generating language, body language and behavioral output, that does not reflect what is going on inside. This is harder for an animal. For example, the poker player can't allow his body language to reveal his hand to the other players. The other players, like observational scientists, are keenly observing all the external data from the other players, for tells and signs, so they can try to guess the relative value of each other's hand. The good bluffer, will alter the corresponding output data, relative to what is going on inside. The animal is easy to read, since when his tail wags he has a good hand. But humans can trick this cause and effect.

This would make science difficult, if all of science had this data collection problem. It would be like trying to study the stars, while aliens are messing with the matter; HeHe. One may draw what appears to be sound conclusions , but in the end it was a bluff.

The advertising agency will create an external data ambience and then try to get their product to overlap under the induced ambience mist, so the external data seems to indicate brand X is better at attracting mates. It works all the time. Picture if while science was running an experiment, I messed with the instrument calibration, so the instrument is now always biased high, so I can get the result I need for my company. Based on the external data we draw a certain conclusion. This is not something most of science has to deal with as a matter of course. With psychology, it can take a long time to get to the real output, since both conscious and even unconscious effects can create poker face as well as signs and tells. Humans are not a straight forward phenomena.

This post has been edited by pioneer: 28 August 2010 - 10:02 PM

[Ringer]

pioneer, on 28 August 2010 - 09:57 PM, said:

Humans are capable of generating language, body language and behavioral output, that does not reflect what is going on inside. This is harder for an animal. For example, the poker player can't allow his body language to reveal his hand to the other players. The other players, like observational scientists, are keenly observing all the external data from the other players, for tells and signs, so they can try to guess the relative value of each other's hand. The good bluffer, will alter the corresponding output data, relative to what is going on inside. The animal is easy to read, since when his tail wags he has a good hand. But humans can trick this cause and effect.

This would make science difficult, if all of science had this data collection problem. It would be like trying to study the stars, while aliens are messing with the matter; HeHe. One may draw what appears to be sound conclusions , but in the end it was a bluff.

What study has been solely based on observation of body language? How would someone bluff a brain scan? And subjects are very limited in their knowledge of what kinds of tests are going on so they don't mess up the data

Quote

The advertising agency will create an external data ambience and then try to get their product to overlap under the induced ambience mist, so the external data seems to indicate brand X is better at attracting mates. It works all the time. Picture if while science was running an experiment, I messed with the instrument calibration, so the instrument is now always biased high, so I can get the result I need for my company. Based on the external data we draw a certain conclusion. This is not something most of science has to deal with as a matter of course. With psychology, it can take a long time to get to the real output, since both conscious and even unconscious effects can create poker face as well as signs and tells. Humans are not a straight forward phenomena.

So are you that if you fix the experiment you get different data? Obviously.

[divagreen]

I was waiting for the five to five count between phwannabe and needimprovement to post, out of respect for original post, but I guess it is okay to post now...

Is psychology a science?

I dunno...does it have to do with reliable predictability based upon the specifics of quantifiable observation? Because I think that might be the difference...

I was going to say that it is considered to be a soft science, but someone beat me to it. So much for being considerate...

This post has been edited by divagreen: 30 August 2010 - 05:01 AM

[needimprovement]

I’m just curious..., how do psychologists measure mental states such as greed, love, anger, hate, desires, hopes, pain, etc?!? I guess the simplest way to quantify mental states is to simply survey people and ask them to rate how happy they consider themselves to be on a scale of 1-10, or something similar.

Can you not use Psychology to learn how the brain works? For example what part does what?

This post has been edited by needimprovement: 2 September 2010 - 10:32 PM

[kitkat]

divagreen, on 30 August 2010 - 04:11 AM, said:

I was waiting for the five to five count between phwannabe and needimprovement to post, out of respect for original post, but I guess it is okay to post now...

Is psychology a science?

I dunno...does it have to do with reliable predictability based upon the specifics of quantifiable observation? Because I think that might be the difference...

I was going to say that it is considered to be a soft science, but someone beat me to it. So much for being considerate...

Science has reduced us to being nothing more then chemicals reactions that keep the atoms that we are made of in a constant animated state of existence. Negative chemical reactions that produce mental illness that currently medication can only treat its symptoms. Most people that require the services of psychology is their inability to understand their behavior which is directly connected to the chemical in-balance in their brains and they need tools to gain some control over their behavior. We are chemical beings of life and behavior is the result of those chemical reactions. Until there is a cure for these mental issues they will still require behavior modification tools that psychology can provide. Our society that we interact with on a daily basis sets the standards of behavior so therefore this science is extremely valuable for human survival. Behavior modification is by no means limited to people suffering from mental issues but to the entire human race. Happier, well adjusted beings will interact more cooperatively with other beings and this is the foundation that is critical for human survival.

To say that other sciences are more valuable strictly because it can be tested and fits mathematical models proves nothing. Life is reduced to its physics and its chemical reactions so if that is the case then why is there no cures for diseases. Diseases would also be reduced to negative chemical reactions as a result of malfunctions of the atoms that make up its chemical reaction. Physics and biochemists that have mapped out all of the elements and its reactions should by now found cures for human diseases.

If life came into existence purely by accident and it evolves with no direction but by chaos and mutations that somehow end up producing new innovations simply by chance could never fit any mathematical model. Those models require purposeful direction to make the equation accurate. Physics deal with non living matter made of atoms so mathematical models works in that respect. It is the complete opposite in sciences dealing with living matter. It is not an exact science so believing that psychology is less of a science that other sciences that deal with life is wrong.