The Last Psychiatrist and Narcissism

[lymphocyte]

I was an avid reader of The Last Psychiatrist, and I'm sad that he no longer posts.

He can certainly be polarizing, but some of his articles were genuinely insightful.

In particular, his conceptualisation of narcissism was utterly fascinating. He used it to explain murder-suicide, honour killings, advertising, relationships, politics etc. in very compelling ways. It seems to have tremendous explanatory power. But where does this conceptualisation come from? What's the intellectual heritage? It doesn't seem to comport with the DSM, and the way he explains it is kinda... opaque. So, is it idiosyncratic? Is there one "must read book" for experts in the field? A collection of essays? Do you have to start at Kernberg and Kohut, or is there a good historical survey? Some of the related general literature, like George Trow's article "Within the Context of of No-Context," gives me goosebumps because it's so damned prescient, and I'd love to read more.

---

Members don't see this ad.

 

[PsyDr]

Not idiosyncratic at all. It's typical psychoanalytic with an emphasis of ego psychology, kohut and lacan. With some of Menninger and resnick thrown in for some of the murder suicide stuff. Anna Freud for the advertising. Basic ego psychology for politics and media.

Typically, the difficulty in understanding this line of thought is: it's not about the content of your thoughts, but the process of thought.

Beyond Freud is a decent and concise summary of this line of thought.

Gabbard, McWilliams, and the pdm are more clinically useful.

 

[lymphocyte]

Not idiosyncratic at all. It's typical psychoanalytic with an emphasis of ego psychology, kohut and lacan. With some of Menninger and resnick thrown in for some of the murder suicide stuff. Anna Freud for the advertising. Basic ego psychology for politics and media.

Typically, the difficulty in understanding this line of thought is: it's not about the content of your thoughts, but the process of thought.

Beyond Freud is a decent and concise summary of this line of thought.

Gabbard, McWilliams, and the pdm are more clinically useful.

The ones below? I'm not at the point of clinical anything... still trying to wrap my mind around the ideas, which don't seem to get taught or even mentioned to medical students...

Freud and Beyond: A History of Modern Psychoanalytic Thought
Psychodynamic Psychiatry in Clinical Practice (4th Edition)
Psychoanalytic Psychotherapy: A Practitioner's Guide

 

[romanticscience]

I was in your position once, bewildered by the explanatory power of his ideas. Although I still respect his genius, I've come to understand that much is rooted in psychoanalytic theory.

Elements of his work can be found in the Cornell/Kernberg model of severe personality disorders (borderlinedisorders.com), especially malignant narcissism and its related aggression. They have developed a coherent model and treatment approach for these individuals, codified in a manualized-based psychoanalytic psychotherapy (Transference-Focused Psychotherapy).

Check out this video of a member of the group speak about the clinical process in TFP oriented to narcissistic personality disorder.

I've taken the long-road by reading through the development of this model, starting with the results of the Menninger Psychotherapy Research Study to the recent TFP manual. PM me for resources.\






Sent from my iPad using SDN mobile

 

[PsyDr]

Those are the authors. The Mitchell book would absolutely be the place to start. Several case vignettes to illustrate. The latter two are the right authors, but there's several choices for each and I'm working off of visual memory so the colors are different.

Basic psychoanalysis 101: person says A, then goes to B, then C. There is a mental procedure there. The content of abc are much less meaningful than the process of how abc were selected, how they are related in the persons mind, how other items were rejected, why they were rejected or relayed, in what manner, what the overall desire is, etc.

I wouldn't worry too much about this stuff. Psychoanalytic training is akin to a fellowship, happens after residency.

 

[lymphocyte]

.

 

[romanticscience]

I would like to differentiate psychodynamic principles, as part of a broader conceptual model of "dynamic psychiatry" (as opposed to purely descriptive or "biological" psychiatry), from psychodynamic psychotherapy. The principles are the unconscious, transference/counter- transference, defenses, etc. These very principles may be utilized in all patient encounters (eg, medication management). They are essential for assessing personality structure and making a diagnosis of personality disorder (eg, BPD). Some would say that personality formulation is irrelevant for treatment planning; though, I disagree. There is research to suggest that untreated BPD limits recovery from other "Axis I" conditions and it should be prioritized if remission is the goal.

Most residencies (I attend a very biologically focused program) will provide training in descriptive psychiatry (structured interviews, DSM diagnoses, evidence-based pharmacologic treatments). This is based on an assumed, pure "medical-model" of psychiatry. A more broad, "psychodynamic psychiatry" model, which takes into account development, identity, and the meaning of symptoms, builds on biological psychiatry. Sadly, there were some who saw psychoanalysis as artificially separate from psychiatry, and this contributed to the schism.

Nowadays, contemporary psychoanalytic ideas incorporate biological theories of mental illness. Some big names in education include Glen Gabbard, Otto Kernberg, and Deborah Cabaniss.

I believe the economic and managed care pressures affecting psychiatry has reinforced the separation of psychodynamic and biological models of psychiatry. Now psychiatrists, primarily prescribe medications where they used to practice the art of combining Med management with talk therapy. Less psychiatrists practice integrated care because it requires a solo private practice (no company will want to pay you to do psychotherapy).

It would be great if all people presented symptoms (anxiety, mood, etc) within a normal or neurotic personality. This level of personality functioning is quite well-adapted and not in need of intensive therapy. However, many people function at a borderline level, and treatment which does not take this into consideration leads to apparent "treatment resistance," as primarily characteralogic issues are reified as exclusively biological conditions.


Sent from my iPad using SDN mobile

 

[splik]

I would like to differentiate psychodynamic principles, as part of a broader conceptual model of "dynamic psychiatry" (as opposed to purely descriptive or "biological" psychiatry), from psychodynamic psychotherapy. The principles are the unconscious, transference/counter- transference, defenses, etc. These very principles may be utilized in all patient encounters (eg, medication management). They are essential for assessing personality structure and making a diagnosis of personality disorder (eg, BPD). Some would say that personality formulation is irrelevant for treatment planning; though, I disagree. There is research to suggest that untreated BPD limits recovery from other "Axis I" conditions and it should be prioritized if remission is the goal.

I have a soft spot for psychoanalytic principles, but it's nonsense to say that these principles are essential for assessing personality structure or diagnosing personality disorders. There are several models one can use that are much more empirically grounded and not psychodynamically based. Also in European psychiatry (which is truly descriptive - I dont agree that American psychiatrists at most programs learn descriptive psychopathology adequately), an assessment of personality and premorbid personality is essential and done without recourse to psychodynamics. There was never the false distinction between axis I and axis II - personality was also on axis I because it was believed mental disorder, including the psychoses arose in those with morbid personalities. Other models of personality include the five-factor model, and cloninger's psychobiological model. While personality is an artificial construct and all models have their flaws and limitations there are many useful models that are more empirically based than voodoo. From a forensic perspective, a good assessment of personality is essential, but you refer to psychodynamic principles at your peril....

Edit: Here are my 3 recommended reads on some other approaches to personality:

McAdams DP. What do we know when we know a person? J Personality 1995; 63:365-396

McCrae RR, John OP. An introduction to the five-factor model and its applications. J Pers 1992; 60:175-215

Mischel W. Toward an integrative science of the person. Annu Rev Psychol 2004; 55:1-22

 

[lymphocyte]

I have a soft spot for psychoanalytic principles, but it's nonsense to say that these principles are essential for assessing personality structure or diagnosing personality disorders. There are several models one can use that are much more empirically grounded and not psychodynamically based. Also in European psychiatry (which is truly descriptive - I dont agree that American psychiatrists at most programs learn descriptive psychopathology adequately), an assessment of personality and premorbid personality is essential and done without recourse to psychodynamics. There was never the false distinction between axis I and axis II - personality was also on axis I because it was believed mental disorder, including the psychoses arose in those with morbid personalities. Other models of personality include the five-factor model, and cloninger's psychobiological model. While personality is an artificial construct and all models have their flaws and limitations there are many useful models that are more empirically based than voodoo. From a forensic perspective, a good assessment of personality is essential, but you refer to psychodynamic principles at your peril....

Edit: Here are my 3 recommended reads on some other approaches to personality:

McAdams DP. What do we know when we know a person? J Personality 1995; 63:365-396

McCrae RR, John OP. An introduction to the five-factor model and its applications. J Pers 1992; 60:175-215

Mischel W. Toward an integrative science of the person. Annu Rev Psychol 2004; 55:1-22

These are very helpful. Thank you. But you seem to place a lot of emphasis on empiricism, and, in particular, the kind of empiricism most valued by modern scientific paradigm: reproducibility (that's essentially what a p-value means).

But there are other reasons to support a theory. Bohr's model of the atom may be empirically false, but it's simple, it produces many accurate predictions, it provides a unifying framework, it has explanatory force, etc.

To pick on a great British example, we may all be neo-Keynesians now, but Keynes's original arguments are still widely taught, despite (or because of) the lack of empirical obfuscation.

I have similar feelings about Kant and Hume (who in many ways were the first cognitive scientists, making powerful observations about human experience through careful reasoning alone) or even the more recent philosopher Alva Noe, who uses philosophy of art (particularly embodied cognition) to make strong and verifiable predictions about AI, as published in Nature and Science.

I feel like psychiatry could benefit from having slightly less "science-envy" and embracing the irreducible aspects of human behaviour, mentation, and experience.

 

[clausewitz2]

These are very helpful. Thank you. But you seem to place a lot of emphasis on empiricism, and, in particular, the kind of empiricism most valued by modern scientific paradigm: reproducibility (that's essentially what a p-value means).

But there are other reasons to support a theory. Bohr's model of the atom may be empirically false, but it's simple, it producez many accurate predictions, it provides a unifying framework, it has explanatory force, etc.

To pick on a great British example, we may all be neo-Keynesians now, but Keynes original thoughts are still widely taught, despite (or because of) the lack of empirical obfuscation.

I have similar feelings about Kant and Hume (who in many ways were the first cognitive scientists, making powerful observations about human experience through careful reasoning alone) or even more recent philosophers Alva Noe, who has used philosophy of art (particularly embodied cognition) to make strong and verifiable preredictions about AI published in Nature and Science.

I feel like psychiatry could benefit having slightly less "science-envy" and embracing the irreducible aspects of human behaviour, mentation, and experience.

And Bohr's model of the atom was relegated to high school chemistry classrooms because it was falsified through experimentation and subsequent theories made better, more accurate predictions about the world and explained phenomena that Bohr's atom could not account for. It is not currently in use by any actual scientists.

The problem with much psychoanalytical is not that it was strictly superceded by a more accurate iteration of the same model, but that it was not falsifiable in an obvious way to begin with. You like history of science references, so I will contend that it was like Wolfgang Pauli put it - "not even wrong".

 

[lymphocyte]

And Bohr's model of the atom was relegated to high school chemistry classrooms because it was falsified through experimentation and subsequent theories made better, more accurate predictions about the world and explained phenomena that Bohr's atom could not account for. It is not currently in use by any actual scientists.

I know. I'm sorry. Even I was a little embarrassed to trot out Bohr (I did study graduate physics), but it's the paradigmatic example in philosophy of science. Perhaps a better example would have been Newtonian physics or special relativity.

The point is, it was (and in some ways still is) a useful model. You can, for example, use it to make accurate predictions, generate new experiments, etc.

The problem with much psychoanalytical is not that it was strictly superceded by a more accurate iteration of the same model, but that it was not falsifiable in an obvious way to begin with. You like history of science references, so I will contend that it was like Wolfgang Pauli put it - "not even wrong".

Are you sure about that?

What if I made predictions about behaviour based on a psychodynamic model? Surely that's a venue for falsifiability (if you want to go with that notion of what science actually does). It may not readily lend itself to aggregate statistical analysis (which makes certain assumptions about how data are generated, etc.), but it still offers itself for scrutiny all the same.

What if the skillful application of a psychodynamic model generates theraputic results, on par with some psychotropic medication? See the seminal narrative review by Shedler below (and there have been several formal meta-analyses since).

What if a psychodynamic model engenders clinically useful empathy for psychopathology, in a way that raw empiricism could not? I would argue that it's emotionally useful for psychiatrists to conceptualise their patients' illness in theoretical terms, particularly when dealing with interpersonally destabilising pathology, like NPD or BPD.

Under these three rubrics, and many others, psychodynamic models allow themselves to be falsifiable. They can, in fact, be "wrong." But I think the more important question is: can they be useful?

Shedler, J. The efficacy of psychodynamic psychotherapy. Am Psychol. 2010;65(2):98-109. http://www.ncbi.nlm.nih.gov/pubmed/20141265

 

[lymphocyte]

The problem with much psychoanalytical is not that it was strictly superceded by a more accurate iteration of the same model, but that it was not falsifiable in an obvious way to begin with. You like history of science references, so I will contend that it was like Wolfgang Pauli put it - "not even wrong".

Sorry to come back to this, but the Nobel Prize winning chemist Roald Hoffmann made the very astute point that little of his work as a chemist is "falsifiable": what hypothesis, exactly, are you falsifying when you're synthesising a new polymer or drug? Instead, you apply a combination of experience, intuition, and theory to create something useful. You might argue, well, the proof's in the pudding--but I've already cited data that show psychotherapy is non-inferior to or even more effective than medication.

 

[clausewitz2]

Sorry to come back to this, but the Nobel Prize winning chemist Roald Hoffmann made the very astute point that little of his work as a chemist is "falsifiable": what hypothesis, exactly, are you falsifying when you're synthesising a new polymer or drug? Instead, you apply a combination of experience, intuition, and theory to create something useful. You might argue, well, the proof's in the pudding--but I've already cited data that show psychotherapy is non-inferior to or even more effective than medication.

Smaller point first - sure, novel synthesis per se is not really testing hypotheses, but if your mass spec means you assign a structure to a compound that doesn't make a d*mn bit of sense given its physical properties, you are testing hypotheses right there.

Toy example: if you alkylated a long chain hydrocarbon at carbon number 53 and nothing much happened but then you alkylated it at carbon 54 and it became a purple gas at room temperature, best believe you are putting paid to some hypotheses in chemistry.

 

[clausewitz2]

I know. I'm sorry. Even I was a little embarrassed to trot out Bohr (I did study graduate physics), but it's the paradigmatic example in philosophy of science. Perhaps a better example would have been Newtonian physics or special relativity.

The point is, it was (and in some ways still is) a useful model. You can, for example, use it to make accurate predictions, generate new experiments, etc.



Are you sure about that?

What if I made predictions about behaviour based on a psychodynamic model? Surely that's a venue for falsifiability (if you want to go with that notion of what science actually does). It may not readily lend itself to aggregate statistical analysis (which makes certain assumptions about how data are generated, etc.), but it still offers itself for scrutiny all the same.

What if the skillful application of a psychodynamic model generated theraputic results, on par with some psychotropic medication? See the seminal narrative review by Shedler below (and there have been several formal meta-analyses since).

What if a psychodynamic model engendered clinically useful empathy for psychopathology, in a way that raw empiricism could not. I would argue that it's emotionally useful for psychiatrists to conceptualise their patients' illness in theoretical terms, particularly when dealing with interpersonally destabilising pathology, like NPD or BPD.

Under these three rubrics, and many others, psychodynamic models allow themselves to be falsifiable. They can, in fact, be "wrong." But I think the more important question is: can they be useful?

Shedler, J. The efficacy of psychodynamic psychotherapy. Am Psychol. 2010;65(2):98-109. http://www.ncbi.nlm.nih.gov/pubmed/20141265

If we are going to start justifying frameworks based on effects that are not primarily due to the details of the frameworks themselves or any entities they posit but on their social context in the therapeutic interaction, I would strongly prefer to make it clear, at least among practitioners, that that is what is happening.

Replace all psychodynamic principles with arbitrary names if you think the details are not important. Say the patient/client did X because of, say, Bob.

 

[clausewitz2]

At base I think there are two broad categories of folks when it comes to this sort of thing: either you believe Karl Popper was on to something or that Feuerabend was a mensch and really had the right idea. I think this is a clash of axioms and difficult to talk across.

 

[lymphocyte]

If we are going to start justifying frameworks based on effects that are not primarily due to the details of the frameworks themselves or any entities they posit but on their social context in the therapeutic interaction, I would strongly prefer to make it clear, at least among practitioners, that that is what is happening.

Replace all psychodynamic principles with arbitrary names if you think the details are not important. Say the patient/client did X because of, say, Bob.

Maaaaaaaaaaaaan.... Have you read STAR*D? Why was CATIE so shocking? Irving Kirsch? Freakin' this. Most of what psychiatrists do, there is still no clue about the details.* Can we at least make that clear? I suppose what I'm saying is, why can't we justify all frameworks (including medicalization) on proven efficacy (broadly and narrowly defined)? On that basis alone, there's good evidence supporting a psychotherapeutic framework.

I don't know what you mean about "arbitrary names." I think it's pretty clear to practitioners what you're doing when you engage in some kind of psychotherapy, whether that's CBT, DBT, psychodynamics, psychoanalysis, etc. And the details do matter. Different evidence for different psychotherapeutic styles.

At base I think there are two broad categories of folks when it comes to this sort of thing: either you believe Karl Popper was on to something or that Feuerabend was a mensch and really had the right idea. I think this is a clash of axioms and difficult to talk across.

Not really. Clinicians treat based on what's been shown to work. There's strong evidence to suggest that some styles of psychotherapy work well in certain clinical contexts (and they "work" in several meanings of the term).

*Of course, we become more knowledgeable over time: autoimmune encephalitis is everybody's favourite example, but I'm not talking about C/L or neuropsych. I'm talking about bread and butter outpatient work.

 

[clausewitz2]

Maaaaaaaaaaaaan.... Have you read STAR*D? Why was CATIE so shocking? Irving Kirsch? Most of what psychiatrists do, there is still no clue about the details.* Can we at least make that clear? I suppose what I'm saying is, why can't we justify all frameworks (including medicalization) on proven efficacy (broadly and narrowly defined)? On that basis, there's good evidence supporting a psychotherapeutic framework.

I don't know what you mean about "arbitrary names." I think it's pretty clear to practitioners that what you're doing when you engage in some kind of psychotherapy, whether that's CBT, DBT, psychodynamics, psychoanalysis, etc. There's no voodoo magic behind a curtain. There's two people talking, with one person thinking very, very carefully about what to say and what's been said. And the details do matter. Different evidence for different psychotherapeutic styles (again, CBT, DBT, psychodynamics, etc.).



Not really. Clinicians treat based on what's been shown to work. There's strong evidence to suggest that some styles of psychotherapy work well in certain clinical contexts (and they "work" in a variety of ways, as mentioned above). I was just curious about learning more.

*Of course, we become more knowledgeable over time: autoimmune encephalitis is everybody's favourite example, but I'm not talking about C/L or neuropsych. I'm talking about bread and butter outpatient work.

So this is a very different conversation than the one you started earlier where you claimed reproducibility was not especially important or was only one good among many.

What is an example of an interpersonal phenomenon that cannot be explained in a psychoanalytical framework?

 

[lymphocyte]

So this is a very different conversation than the one you started earlier where you claimed reproducibility was not especially important or was only one good among many.

I don't think so. Here's the flow as I see it --

I started off by saying that empiricism strictly in terms of reproducibility is an impoverished empiricism, and that there are other good reasons to support a theory or model, such as explanatory power, predictive ability, and clinical efficacy. You brought up the idea of falsifiability. I argued that indeed psychotherapy is falsifiable in many non-obvious ways. Then you said something I didn't quite understand, so I apologise if I misconstrued you. What I responded with was, in terms of clinical efficacy alone, there's good evidence to support psychotherapy (even psychoanalytic psychotherapy), perhaps just as much evidence as for some psychopharmacology.

Then I had a burrito and sorta lost my train of thought.

What is an example of an interpersonal phenomenon that cannot be explained in a psychoanalytical framework?

I have no clue. But certainly psychoanalysis isn't a thing that you can point to and use to categorise interpersonal interactions in a binary fashion ("psychoanalytically explicable/inexplicable"). It seems like an attitude, a perspective, an entire intellectual tradition with competing and elaborating ideas. It's been protocolised to some extent as @RomanticScience pointed out with Transference-Focused Psychotherapy (TFP). That certainly brings it under the rubric of reproducible empiricism, but perhaps reproducible empiricism isn't the only (or best) way to evaluate it. That was my original point.

I feel like I've taken us far off track. Blame the burrito.

 

[smalltownpsych]

Object relations theory and attachment theory are important for understanding interpersonal development and functioning as well as affective regulation. Next look at basic principles of learning such as reinforcement and conditioning. I don't like to get to deep into the theoretical or go out on an interpretive limb about explaining behavior and prefer to stick to the more neurologically based and observable phenomena.

 

[clausewitz2]

Object relations theory and attachment theory are important for understanding interpersonal development and functioning as well as affective regulation. Next look at basic principles of learning such as reinforcement and conditioning. I don't like to get to deep into the theoretical or go out on an interpretive limb about explaining behavior and prefer to stick to the more neurologically based and observable phenomena.

Yeah, should clarify: attachment theory has a tremendous evidence base, as does classical learning. I hope nothing I said suggests I have any beef with those frameworks.

 

[smalltownpsych]

Yeah, should clarify: attachment theory has a tremendous evidence base, as does classical learning. I hope nothing I said suggests I have any beef with those frameworks.

Not at all. In fact I was getting a bit lost in the discussion so just felt like adding in my two cents.

 

[lymphocyte]

Not at all. In fact I was getting a bit lost in the discussion so just felt like adding in my two cents.

@smalltownpsych Were there any seminal texts or articles that were formative to your development as a psychologist? Especially with regard to personality disorder (but anything really). I'm done with Steps, doing an unexpectedly light rotation, and finally have some time to read...

 

[Ceke2002]

Object relations theory and attachment theory are important for understanding interpersonal development and functioning as well as affective regulation. Next look at basic principles of learning such as reinforcement and conditioning. I don't like to get to deep into the theoretical or go out on an interpretive limb about explaining behavior and prefer to stick to the more neurologically based and observable phenomena.

I love objection relations and attachment theory from a treatment point of view, although right now we're doing some cognitive analytical work (which I'm also thoroughly enjoying). Definitely exploring reinforcement and conditioning has been majorly helpful for certain conditions as well. Obviously, as you already know, my Psychiatrist tends to be fairly eclectic in his approach to psychotherapy (he does like to mix things up, so in 6 years or so we've covered a mix of IPT, CBT, CBT-Mindfulness, ACT, Metacognition, Psychodynamic, Object Relations, Attachment Theory, CAT, and probably several others I'm forgetting off the top of my head), but I actually find that style to be a lot more helpful than someone who seems to searching for the one true way of therapy.

 

[smalltownpsych]

@smalltownpsych Were there any seminal texts or articles that were formative to your development as a psychologist? Especially with regard to personality disorder (but anything really). I'm done with Steps, doing an unexpectedly light rotation, and finally have some time to read...

Amazon product

 

[smalltownpsych]

I love objection relations and attachment theory from a treatment point of view, although right now we're doing some cognitive analytical work (which I'm also thoroughly enjoying). Definitely exploring reinforcement and conditioning has been majorly helpful for certain conditions as well. Obviously, as you already know, my Psychiatrist tends to be fairly eclectic in his approach to psychotherapy (he does like to mix things up, so in 6 years or so we've covered a mix of IPT, CBT, CBT-Mindfulness, ACT, Metacognition, Psychodynamic, Object Relations, Attachment Theory, CAT, and probably several others I'm forgetting off the top of my head), but I actually find that style to be a lot more helpful than someone who seems to searching for the one true way of therapy.

I tend to use the object relations and attachment theory stuff for conceptualizing and guiding my interventions than the actual interventions. That is where it gets confusing and difficult to research. Does my understanding of neurodevelopment and psychological development help me better choose techniques and interventions that are more effective than merely using a diagnostic checklist and selecting the indicated manualized treatment? That is what I tend to believe and it makes sense, but there are problems with acting on belief and what seems to make sense because it could still be incorrect.

 

[whopper]

Was TLP polarizing? Yes but kind of like John Keating, Robin Williams' character from Dead Poet Society. He brought up a lot of problems that are going on in psychiatry that the Ivory Tower academicians never bothered to talk about.

 

[lymphocyte]

Was TLP polarizing? Yes but kind of like John Keating, Robin Williams' character from Dead Poet Society. He brought up a lot of problems that are going on in psychiatry that the Ivory Tower academicians never bothered to talk about.

I tend to agree. But "polarising' is a great cover word, isn't it?

To be fair, his language was... blunt, and some of his posts on rape and gender equality got a bit... out there.

Another point: he stopped writing soon after he was doxxed. To my knowledge, he never deleted or edited what he had already posted, but I've always wondered if he became worried about reputational injury, etc. A sad loss in any case.

I would agree, but you've entirely missed the point, which is that counter to what was claimed there are other models once can use for assessment and conceptualization of personality than a psychodynamic one.

I think I understood that point, but twice you seemed to use empiricism as a reason to favour one kind of an assessment over another (as below).

You did mention forensics at the very end, but forensics is an ancillary function of psychiatry that privileges its own kinds of methods and discourse (citational, reproducible, authoritative, etc.). I was centering the discussion on psychiatry more broadly.

So perhaps I wasn't contending your main point but exploring some of the implicit assumptions of value in your original post.

I see these implicit value assumptions all the time in medicine, and they often go unchallenged, especially in relation to reproducible empiricism. It's nowhere more frustrating than in psychiatry where the empirical evidence is often sparse and difficult to individuate.

You may think this is arm-chair puffery, but I see a similar dynamic unfold in intensive care (my first love). There too you see clashes between theoretical models and reproducible empiricism (and, unsurprisingly, there too the empirical evidence is often sparse and difficult to individuate). When that happens, some intensivists push back on theoretical grounds and are sometimes vindicated: dry vs wet in sepsis, Guyton blood pressure curve and fluid responsiveness, strict vs liberal transfusion criteria, PPI vs gut biota translocation--just read one of the more popular though polarising texts in the field (The ICU Book) and you'll get an idea of what I mean. And here's one of many examples from the contemporary literature: http://www.ncbi.nlm.nih.gov/pubmed/26026358.

If I've completely misconstrued you than I apologise. I think we're in consensus based on your last post.

While personality is an artificial construct and all models have their flaws and limitations there are many useful models that are more empirically based than voodoo

There are several models one can use that are much more empirically grounded and not psychodynamically based.

 

[splik]

You did mention forensics at the very end, but forensics is an ancillary function of psychiatry that privileges its own kinds of methods and discourse (citational, reproducible, authoritative, etc.). I was centering the discussion on psychiatry more broadly.

Forensics is the very basis of the development of the specialty of psychiatry in the first place, so I don't think it is correct to say that it is merely an ancillary function of psychiatry. All alienists were experts in medical jurisprudence. The moral authority and power that psychiatry holds is based on it supposedly being a medical science (which is why in the US at least psychoanalysis was held to be a science), and this gives psychiatrists an extremely privileged position where we have the power to deprive people of their liberty, and determinations of whether one is competent to stand trial, make financial decisions, make medical decisions, make a will, be executed, or whether they should be held criminally responsible for their actions. That the average clinical psychiatrist today is ignorant of the law does not make our work any less mired in law. The very act of diagnosis itself is a political act that has far-reaching consequences and implications. I never stated that an empirical or positivistic approach to psychiatry was necessarily superior to a psychoanalytic approach (and I think most people who read my posts would know my reference to voodoo was somewhat tongue in cheek), but like it or not this is the status of psychiatry today, a status I have lamented elsewhere.

 

[lymphocyte]

Forensics is the very basis of the development of the specialty of psychiatry in the first place, so I don't think it is correct to say that it is merely an ancillary function of psychiatry. All alienists were experts in medical jurisprudence. The moral authority and power that psychiatry holds is based on it supposedly being a medical science (which is why in the US at least psychoanalysis was held to be a science), and this gives psychiatrists an extremely privileged position where we have the power to deprive people of their liberty, and determinations of whether one is competent to stand trial, make financial decisions, make medical decisions, make a will, be executed, or whether they should be held criminally responsible for their actions. That the average clinical psychiatrist today is ignorant of the law does not make our work any less mired in law. The very act of diagnosis itself is a political act that has far-reaching consequences and implications. I never stated that an empirical or positivistic approach to psychiatry was necessarily superior to a psychoanalytic approach (and I think most people who read my posts would know my reference to voodoo was somewhat tongue in cheek), but like it or not this is the status of psychiatry today, a status I have lamented elsewhere.

No... It was indeed the basis but thankfully it's now ancillary to therapy. At least, I hope therapy figures much more prominently in what psychiatrists do than forensics. Are you suggesting that jurisprudence really plays that much of role in what most psychiatrists actually do (outpatient, affective disorders, not in the Ivory Tower)?

The idea of diagnosis being a political act... I thoroughly disagree, but I won't touch that topic with a ten-foot pole.

I take your point about one framework not necessarily being superior to another. It just rustles mah jimmies whenever I see "it's more empirical" as a reason to favour anything.

 

[clausewitz2]

Ummm... It was indeed the basis but thankfully it's now ancillary to therapy. At least, I hope therapy figures much more prominently in what psychiatrists do than forensics. I mean, does jurisprudence really play that much of role in what most psychiatrists actually do (outpatient, affective disorders, not in the Ivory Tower)?

The idea of diagnosis being a political act... I thoroughly disagree, but I wouldn't touch that topic with a ten-foot pole.

But I take your point about one framework not necessarily being superior to another. It just rustles mah jimmies whenever I see "it's more empirical" as a reason to favour anything.

So this is sort of what I was talking about in bringing up Popper v Feuerabend in philosophy of science. Reproducibility is a necessary theoretical commitment for falsifiabilitif you deny the importance of the first -in principle-, you can always come up with a reason to avoid the second.

If you think Popper was on to something with logical positivism, not being falsifiable is very close, or at least somewhat adjacent to, being incoherent. As an aside, for truth-conditional semantics, this is literally true.

If you side with Feyerabend, let a thousand epistemic flowers bloom and don't be too concerned about whether something is falsifiable if it seems to be useful and true-ish. This leaves the door open to rather a lot of woo and questions of well why don't we use vision quests or sing the seidr song to ask for Odin's counsel if all participants feel they derive benefit.

Do you see how it this is a difference in epistemic axioms that makes arguments rather impoverished?

Edit: I should say that I think the latter position is logically defensible - @splik has commented approvingly before about the shamanic aspects of our profession - it just makes a very different set of initial assumptions.

 

[lymphocyte]

Reproducibility is a necessary theoretical commitment for falsifiabilitif

Of course it's not. I can falsify the statement: "Abraham Lincoln was president yesterday." I can never falsify that statement by reproduction.

Do you know the literal definition of the p-value? I have it open right in front of me in Larry Wasserman's "All of Statistics" (which is a graduate-level text): it's the probability, under the assumption of some hypothesis, of obtaining a result equal to or more extreme than what was actually observed. But how bizarre is that definition? Many would argue that what was actually observed was all there ever will be--that very little of any experiment can ever actually be reproduced and that there are no do-overs in life.

I don't mean to launch a discussion on frequentist vs Bayesian interpretations of statistics, but, again, this isn't arm-chair puffery. This philosophical nuance has had tremendous implications for medicine and psychology. It explains why most published research findings are probably false. Here's a great Nature article in the same vein: www.nature.com/news/scientific-method-statistical-errors-1.14700

vision quests or sing the seidr song to ask for Odin's counsel if all participants feel they derive benefit

I don't even...

I've already cited data supporting psychotherapy (if you want to go the route of empiricism). I've also made several non-empirical arguments to support certain psychotherapeutic frameworks. If you want to engage on those grounds--on the grounds of data and reasoned argument--that's great. Otherwise, we'll be stuck at impasse. What you wrote strikes me as an uncharitable caricaturization.

By the way, people who ascribe to these frameworks aren't idiots. And their patients aren't necessarily guilliable. Something real and demonstrable happens when two people engage in a psychodynamic way--sometimes to greater effect than what could ever happen pharmacologically. I think that's one of the points @RomanticScience was making.

I'm definitely not arguing for pharmacological nihilism. But there's a very undeserved confidence that psychiatry, and medicine more generally, has gleaned from physics by slathering itself with "mathiness." I think until we acknowledge how tenuous our evidence-base really is (for what most of what psychiatrists actually do--outpatient, affective disorders, etc.), we'll continue to be humiliated by the results of STAR*D, the unpublished study 329, Irving Kirsch, the new network meta-analysis in BMJ on antidepressants in adolescents, etc. Some might even argue that prescribing most SSRIs in most instances would be just the same as seeking Odin's counsel...

 

[clausewitz2]

Of course it's not. I can falsify the statement: "Abraham Lincoln was president yesterday." I can never falsify that statement by reproduction.

Do you know the literal definition of the p-value? I have it open right in front of me in Larry Wasserman's "All of Statistics" (which is a graduate-level text): it's the probability, under the assumption of some hypothesis, of obtaining a result equal to or more extreme than what was actually observed. But how bizarre is that definition? Many would argue that what was actually observed was all there ever will be--that very little of any experiment can ever actually be reproduced and that there are no do-overs in life.

I don't mean to launch a discussion on frequentist vs Bayesian interpretations of statistics, but, again, this isn't arm-chair puffery. This philosophical nuance has had tremendous implications for medicine and psychology. It explains why most published research findings are probably false. Here's a great Nature article in the same vein: www.nature.com/news/scientific-method-statistical-errors-1.14700



I don't even...

I've already cited data supporting psychotherapy (if you want to go the route of empiricism). I've also made several non-empirical arguments to support certain psychotherapeutic frameworks. If you want to engage on those grounds--on the grounds of data and reasoned argument--that's great. Otherwise, we'll be stuck at impasse. What you wrote strikes me as an uncharitable caricaturization.

By the way, people who ascribe to these frameworks aren't idiots. And their patients aren't necessarily guilliable. Something real and demonstrable happens when two people engage in a psychodynamic way--sometimes to greater effect than what could ever happen pharmacologically. I think that's one of the points @RomanticScience was making.

I'm definitely not arguing for pharmacological nihilism. But there's a very undeserved confidence that psychiatry, and medicine more generally, has gleaned from physics by slathering itself with "mathiness." I think until we acknowledge how tenuous our evidence-base really is (for what most of what psychiatrists actually do--outpatient, affective disorders, etc.), we'll continue to be shocked by the results of STAR*D, the unpublished study 329, Irving Kirsch, the new network meta-analysis in BMJ on antidepressants in adolescents, etc. Some might even argue that prescribing most SSRIs in most instances would just be the same as seeking Odin's counsel...

I am mostly in agreement with you about the tenuous evidence base in much of the field, but I seem to have a different reaction. I think that part of the solution involves more and better trials with more carefully selected populations,e endpoints, and interventions, and you seem to think that is less important. Fair?

Also, you certainly can test the example statement you gave by reproduction if you consider consulting every piece of relevant evidence as a new trial. But that is largely a quibble. At the end of the day, I making a very basic logical point. If you say "in principle, it does not matter if future tests of this hypothesis find it to be true so long as the current test finds it to be true" it becomes very, very difficult to falsify theories predicated on that test, because you always have the dodge of "well it worked this one time let me tell you about it".

I am not sure that basic point requires statistical exegesis.

 

[lymphocyte]

I am mostly in agreement with you about the tenuous evidence base in much of the field, but I seem to have a different reaction. I think that part of the solution involves more and better trials with more carefully selected populations,e endpoints, and interventions, and you seem to think that is less important. Fair?

Fair enough. I think this attitude--as things currently stand--will lead to a tremendous waste of resources and reputation. But probably a continued bonanza for the pharmaceutical industry. (You know I was going to whip out my tin-foil hat at some point...)

Also, you certainly can test the example statement you gave by reproduction if you consider consulting every piece of relevant evidence as a new trial. But that is largely a quibble. At the end of the day, I making a very basic logical point. If you say "in principle, it does not matter if future tests of this hypothesis find it to be true so long as the current test finds it to be true" it becomes very, very difficult to falsify theories predicated on that test, because you always have the dodge of "well it worked this one time let me tell you about it".

Yeah, we'll just have to disagree here, if you can call it disagreement, since I'm definitely not making this point.

 

[clausewitz2]

Fair enough. I think this attitude--as things currently stand--will lead to a tremendous waste of resources and reputation. But probably a continued bonanza for the pharmaceutical industry. (You know I was going to whip out my tin-foil hat at some point...)



Yeah, we'll just have to disagree here, if you can call it disagreement, since I'm definitely not making this point.

So what on earth did you mean when you said the reproducibility was not important? You seem to be using this word in a non-standard way.

Edit: but then I have never met a frequentist in the wild. Re-reading your posts I get the sense that is where you are coming from, so you all probably think about these things very differently. I will admit I am deeply committed intellectually to a Bayesian framework.

 

[clausewitz2]

For the record I think Ioannidis' interpretation of his seminal paper is "don't assign a high degree of confidence to a single published trial" not "empirical evidence is intrinsically suspect".

 

[lymphocyte]

For the record I think Ioannidis' interpretation of his seminal paper is "don't assign a high degree of confidence to a single published trial" not "empirical evidence is intrinsically suspect".

I think that's most people's interpretation. I completely agree with it. Again, I'm distinguishing empirical evidence (generally good) from reproducible empiricism (generally not-awesome).

So what on earth did you mean when you said the reproducibility was not important? You seem to be using this word in a non-standard way.

Not at all.

Again, I can falsify the statement: "Abraham Lincoln was president yesterday." I can never falsify that statement by reproduction (yesterday will never happen again), and yet you and I would still agree that it's false. Why? There are many reasons, but clearly falsifiability is much more than reproducibility. Moreover, do you see how that statement being true 150 years ago doesn't mean that it must be true now? Something other than reproducibility can make a statement true or false.

I'm arguing that theories should stand or fall on much more than just reproducibility (as embodied by the p-value). Simplicity, unification, explanatory force, predictive capacity--all of these other features, often non-empirical, are valuable, perhaps even more valuable than reproduction (if reproduction is indeed possible). And yet, if your only argument in favour of a theory is that "p-value <.05", well, that seems very impoverished indeed.

In my very limited experience as a medical student, "p-value not <.05" has been the only consistent argument I've heard against psychodynamic psychotherapy. Let's use citalopram instead! Oh wait...

 

[lymphocyte]

Edit: but then I have never met a frequentist in the wild. Re-reading your posts I get the sense that is where you are coming from, so you all probably think about these things very differently. I will admit I am deeply committed intellectually to a Bayesian framework.

I'm actually a trained (and once gainfully employed) Bayesian. Unfortunately, the vast majority of doctors interpret evidence like frequentists (even though they're almost certainly not), and they don't know the difference. This confusion has tremendous implications for therapy, research, etc.

 

[clausewitz2]

I think that's most people's interpretation. I completely agree with it. Again, I'm distinguishing empirical evidence (generally good) from reproducible empiricism (generally not-awesome).



Not at all.

Again, I can falsify the statement: "Abraham Lincoln was president yesterday." I can never falsify that statement by reproduction (yesterday will never happen again), and yet you and I would still agree that it's false. Why? There are many reasons, but clearly falsifiability is much more than reproducibility. Moreover, do you see how that statement being true 150 years ago doesn't mean that it must be true now? Something other than reproducibility can make a statement true or false.

I'm arguing that theories should stand or fall on much more than just reproducibility (as embodied by the p-value). Simplicity, unification, explanatory force, predictive capacity--all of these other features, often non-empirical, are valuable, perhaps even more valuable than reproduction (if reproduction is indeed possible). And yet, if your only argument in favour of a theory is that "p-value <.05", well, that seems very impoverished indeed.

In my very limited experience as a medical student, "p-value not <.05" has been the only consistent argument I've heard against psychodynamic psychotherapy. Let's use citalopram instead! Oh wait...

I continue to assert that when you evaluate the Abraham Lincoln statement, you are testing the hypothesis and it's truth conditions against whatever pieces of evidence you have (newspapers, other people's memories of who was president yesterday, etc). The time dependence introduced by the word "yesterday" makes this claim somewhat unlike most scientific claims - very few people want to publish papers demonstrating phenomena that only happen on the third Tuesday of every month!

I happen to agree with you about p-values being problematic, and I think there is glacially slow movement towards not reporting them, instead using effect sizes with confidence intervals.

 

[lymphocyte]

I continue to assert that when you evaluate the Abraham Lincoln statement, you are testing the hypothesis and it's truth conditions against whatever pieces of evidence you have (newspapers, other people's memories of who was president yesterday, etc). The time dependence introduced by the word "yesterday" makes this claim somewhat unlike most scientific claims - very few people want to publish papers demonstrating phenomena that only happen on the third Tuesday of every month!

I happen to agree with you about p-values being problematic, and I think there is glacially slow movement towards not reporting them, instead using effect sizes with confidence intervals.

1. I think you're correct; but nothing that you cite has anything to do with reproducibility. In slogan form: reproducibility is not falsifiability. P-value does not science make.

2. All frequentist claims are claims about the past; in the medical context, they obtain from trials that have already been conducted and can never be fully reproduced (that's what the p-value means). Only Bayesians make their claims accessible to new information (and information that isn't strictly limited to empirical claims).

3. I think we're in consensus with only a few theoretical quibbles. Seriously the best discussion I've had in weeks.

 

[clausewitz2]

1. I think you're correct; but nothing that you cite has anything to do with reproducibility. In slogan form: reproducibility is not falsifiability.

2. All frequentist claims are claims about the past; in the medical context, they obtain from trials that have already been conducted and can never be reproduced (that's what the p-value means). Only Bayesians make their claims accessible to new information (and information that isn't strictly limited to empirical claims).

3. I think we're in consensus with only a few theoretical quibbles. Seriously the best discussion I've had in weeks.

Agrees about the quality of the discussion. Reproducibility in the context of the Abraham Lincoln example is just the tendency of the sources of evidence you consult to continue to support the hypothesis on repeated consultation. That is, if i find a line in a history of the US presidency saying Lincoln was no longer president yesterday, you can find the same line when you consult the source, and that line will still be there when I read it again next week.

Now, if there is a photograph of the president from yesterday and sometimes it shows Lincol and sometimes it shows Taft, there is a reproduction failure. Also, if you build your theory on evidence like that photograph without at least a principled commitment to finding reproducible evidence, you can see how it would be difficult to falsify.

 

[lymphocyte]

Reproducibility in the context of the Abraham Lincoln example is just the tendency of the sources of evidence you consult to continue to support the hypothesis on repeated consultation.

But that's not reproducibility. Reproducibility is reproducing data to support a claim. ("Look Ma! I did the experiment over again"!) What data can you reproduce to support the claim that: "Abraham Lincoln was president yesterday"? I contend that yesterday can never truly be reproduced, just like any trial can never truly be reproduced.

You talk about consultation, citation, memory, etc. None of them reproduce data, and all of them are context-dependent. The paper from 1862 that says, "Abraham Lincoln was president yesterday" will be true until it's not. The only way to adjudicate when it's not true is to update with new kinds of information. That's the Bayesian insight.

Sorry to keep saying this, but this isn't arm-chair puffery. There are Bayesian techniques that capture this insight and generate statistical estimators that 1) are more intuitive to interpret, 2) tend to have a smaller standard error, and 3) account for all of those other, sometimes non-empirical, reasons to believe a theory. Meanwhile, I hope the p-value dies a swift and miserable death.

Again, the only consistent argument I've heard against dynamic psychiatry has been "p-value not <.05." Meanwhile descriptive psychiatry has been repeatedly humiliated with STAR*D, CATIE, study 329, multiple meta-analyses, Irving Kirsch, etc.

There needs to be a new paradigm.

 

[clausewitz2]

But that's not reproducibility. Reproducibility is reproducing data to support a claim. ("Look Ma! I did the experiment over again"!) What data can you reproduce to support the claim that: "Abraham Lincoln was president yesterday"? I contend that yesterday can never truly be reproduced, just like any trial can never truly be reproduced.

You talk about consultation, citation, memory, etc. None of them reproduce data, and all of them are context-dependent. The paper from 1862 that says, "Abraham Lincoln was president yesterday" will be true until it's not. The only way to adjudicate when it's not true is to update with new kinds of information. That's the Bayesian insight.

Sorry to keep saying this, but this isn't arm-chair puffery. There are Bayesian techniques that capture this insight and generate statistical estimators that 1) are more intuitive to interpret, 2) tend to have a smaller standard error, and 3) account for all of those other, sometimes non-empirical, reasons to believe a theory. Meanwhile, I hope the p-value dies a short and miserable death.

Again, the only consistent argument I've heard against dynamic psychiatry has been "p-value not <.05." Meanwhile descriptive psychiatry has been repeatedly humiliated with STAR*D, CATIE, study 329, multiple meta-analyses, Irving Kirsch, etc. (I'm only referring here to outpatient, mostly affective disorders, etc.).

There needs to be a new paradigm.

I think inserting "yesterday" is throwing is off a bit. "Abraham Lincoln is president" is very easily susceptible to experimental validation. By putting the temporal conditional on it, you are removing the ability to generate new and relevant data from experimental methods in the current moment, because by stipulation it is not currently relevant. After all, we are talking about yesterday, not today.

Replace my talk of consulting with "sampling," and assume you are looking at the whole corpus of texts that happen to mention who the president was yesterday. Now we are getting into meaningful reproducibility even as you define it, I think.

 

[lymphocyte]

I think inserting "yesterday" is throwing is off a bit. "Abraham Lincoln is president" is very easily susceptible to experimental validation. By putting the temporal conditional on it, you are removing the ability to generate new and relevant data from experimental methods in the current moment, because by stipulation it is not currently relevant. After all, we are talking about yesterday, not today.

Replace my talk of consulting with "sampling," and assume you are looking at the whole corpus of texts that happen to mention who the president was yesterday. Now we are getting into meaningful reproducibility even as you define it, I think.

Yes, but you ought never report "the p-value is <.05" (at least, not if you were speaking precisely) but rather "the p-value was <.05." Remember the formal definition: "the p-value is the probability, under the assumption of some hypothesis, of obtaining a result equal to or more extreme than what was actually observed."

The p-value is always with respect to an experiment that has already been conducted. No new information. No updating. No nothing. Finito.

"Yesterday" is supposed to throw you off. The temporal condition is precisely the point. It shows why reproducibility is a very misleading concept and leads to an impoverished kind of empiricism. Bayesian theory is dynamic and permits pre-test probabilities to be updated. Frequentists are static, beholden to experiments that can never truly be reproduced. This distinction carries forward to what kind of research is privileged, what kind of evidence is valued, etc.

Most psychiatrists are Bayesian but act like frequentists. This confusion has tremendous implications for research, therapy, and policy.

 

[splik]

Again, the only consistent argument I've heard against dynamic psychiatry has been "p-value not <.05." Meanwhile descriptive psychiatry has been repeatedly humiliated with STAR*D, CATIE, study 329, multiple meta-analyses, Irving Kirsch, etc.

There needs to be a new paradigm.

Ah but psychoanalytic treatment has been shown to be effective in terms of p-values (though the obsession with p-values is usually by people who are trying to sound clever but have no understanding of statistics whatsoever, a pox on them, but I hate people who say "correlation does not equal causation" even more as it is the refuge of the stupid). The argument against psychoanalytic treatment is the same as that against homoepathy or EMDR - which is that even if the treatment works, the theoretical basis is hocus pocus and constitutes a deception. Deliberately peddling a treatment based on theories that are untrue, pseudoscientific, unfalsifiable, unverifiable is why people meet psychodynamics with scorn, not its efficacy as it undermines the credibility of the profession. I'm not necessarily saying I agree with this as I'm a pragmatist but I certainly sympathize.

In the Tavistock Adult Depression Study psychoanalytic psychotherapy did quite well http://onlinelibrary.wiley.com/doi/10.1002/wps.20267/abstract
Falk Leichsenring's work has also shown efficacy for psychoanalytic treatment with obligatory (but largely useless) p-values thrown in.

 

[lymphocyte]

Ah but psychoanalytic treatment has been shown to be effective in terms of p-values (though the obsession with p-values is usually by people who are trying to sound clever but have no understanding of statistics whatsoever, a pox on them, but I hate people who say "correlation does not equal causation" even more as it is the refuge of the stupid). The argument against psychoanalytic treatment is the same as that against homoepathy or EMDR - which is that even if the treatment works, the theoretical basis is hocus pocus and constitutes a deception. Deliberately peddling a treatment based on theories that are untrue, pseudoscientific, unfalsifiable, unverifiable is why people meet psychodynamics with scorn, not its efficacy as it undermines the credibility of the profession. I'm not necessarily saying I agree with this as I'm a pragmatist but I certainly sympathize.

In the Tavistock Adult Depression Study psychoanalytic psychotherapy did quite well http://onlinelibrary.wiley.com/doi/10.1002/wps.20267/abstract
Falk Leichsenring's work has also shown efficacy for psychoanalytic treatment with obligatory (but largely useless) p-values thrown in.

Hoboy.

Let's distinguish psychoanalytic as a subset of psychotherapy.

I've already argued that psychotherapy (even psychoanalytic psychotherapy) is falsifible in several non-obvious ways. I've already argued that it is verifiable as well. No need to rehash that.

Truth is a value judgment that no scientific model can rightly attain on a scientific basis alone. Science just doesn't work that way.

Pseudoscientific... well, I've read the studies on all sides, and I'm generally not impressed. So-called "biologically psychiatry" has also peddled its own theoretical basis about how for example 1) SSRI's work or 2) how atypicals are therapeutically different from typical antipsychotics... until, oh wait, 1) they probably don't, and 2) they aren't that different at all.

I think there's plenty of hocus pocus and deception to go around (almost literally with the unpublished study 329 and others).

There's more I want to say here, but I'm starting to flag... blerg.

 

[splik]

Pseudoscientific... well, I've read the studies on all sides, and I'm generally not impressed. So-called "biologically psychiatry" has also peddled its own theoretical basis about how for example 1) SSRI's work or 2) how atypicals are therapeutically different from typical antipsychotics... until, oh wait, 1) they probably don't, and 2) they aren't that different at all.

well biological psychiatry has its own brand of pseudoscience

 

[PsyDr]

IMO: Psychoanalysis is a useful tool for self discovery for some people. Probably a pretty poor tool for treatment of axis I disorders. For some personality disorders, it may be useful. Probably not for most. The theory behind the technique has constantly evolved. There is even theoretical explanations about why the technique works (for some) that use a completely different theoretical explanation (i.e., it works, but not how you think it works). There are other tools out there. Some professional, some not. Some better, some not.

If you're interested in CBT applications of psychodynamic techniques, the Functional Analytic Psychotherapy (a 3rd wave CBT) literature may be of interest.

From a financial standpoint, psychoanalysis is awesome for the provider.

 

[lymphocyte]

From a financial standpoint, psychoanalysis is awesome for the provider.

May I ask some tangential and somewhat rude questions? How does renumeration work? Per hour? Cash? Do any insurance plans even reimburse? How do you feel good enough to charge X? Is fellowship worth it financially or more for building a theoretical framework (or set of frameworks)? Advertising? How quickly do you build a patient panel? I've always wondered about the actual details of running a PP psychoanalytic shop... Please don't feel obligated to answer any or all.

 

[lymphocyte]

@whopper I found a TLP post that was deleted! Polarising indeed, but it struck me as one of his most insightful...

http://trilema.com/wp-content/uploads/2014/11/tlp.html

 

[PsyDr]

May I ask some tangential and somewhat rude questions? How does renumeration work? Per hour? Cash? Do any insurance plans even reimburse? How do you feel good enough to charge X? Is fellowship worth it financially or more for building a theoretical framework (or set of frameworks)? Advertising? How quickly do you build a patient panel? I've always wondered about the actual details of running a PP psychoanalytic shop... Please don't feel obligated to answer any or all.

1) an analyst sees an analysand. The analysand pays cash for the analysts time.
2) it's per hour
3) I guess the provider actually believes that the service is worth what they charge.
4) no idea how it works out financially for the median practitioner.
5) insurance does not cover psychoanalysis.
6) no idea
7) no idea
8) the details are likely very different than most medical and psychological practices. Highly dependent on prestige, location, etc. I went to a big name person for a while who charged around $400/hr and required 3x a week.