7 ms·

> 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 evidenti

> 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. 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.

lutusp 12y ago
> There are plenty of theories. They are typically considered theories on motivation and behavior rather than behavior itself. 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-diagnosis.shtml 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.
- HelloMcFly 12y ago
  My hesitance to continue this debate is that I find places like HN and reddit to be pretty hostile audiences, and engaging in the conversation occupies more of my thoughts than I find it should. It's a form a self-regulation to know when I should stop, particularly given the audience. 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.
otakucode 12y ago
> 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. 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?
- lutusp 12y ago
  You and I are speaking about two very different things. The incompleteness theorems and the halting problem (which are deeply connected, by the way) are positive proofs. I am speaking of a negative proof. 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?

lutusp 12y ago

> There are plenty of theories. They are typically considered theories on motivation and behavior rather than behavior itself. 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-diagnosis.shtml 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.

HelloMcFly 12y ago
My hesitance to continue this debate is that I find places like HN and reddit to be pretty hostile audiences, and engaging in the conversation occupies more of my thoughts than I find it should. It's a form a self-regulation to know when I should stop, particularly given the audience. 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.

otakucode 12y ago

> 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. 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?

lutusp 12y ago
You and I are speaking about two very different things. The incompleteness theorems and the halting problem (which are deeply connected, by the way) are positive proofs. I am speaking of a negative proof. 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?