An Unwarranted Hidden Assumption in Research on Personality Disorders

One of the major reasons I became interested in family systems theory, tribalism, family myths, social psychology, and other manifestations of collectivism was because I noticed a big problem with the major forms of psychotherapy practiced on individuals: psychodynamic and cognitive-behavior therapy, and, though to a lesser extent, humanistic therapies like Gestalt therapy.  

All of these forms of individual therapies pay way too much attention to the way patients are reacting, and not nearly enough attention to what it is they are reacting to.

It’s a bit like looking at someone who is falling apart after recently having personally witnessed their entire family being beheaded by terrorists, and concluding that he or she has “poor distress tolerance coping skills.” Well, maybe not quite that bad, but you get the idea.

Some psychologists talk about something called the fundamental attribution error. According to Richard Nisbett and Lee Ross in their 1980 book, Human Inference: Strategies and Shortcomings of Social Judgment, this is defined as “the assumption that behavior is caused primarily by the enduring and consistent disposition of the actor, as opposed to the particular characteristics of the situation to which the actor responds.”

Richard E. Nisbett, Ph.D.

Of course, internal predispositions, one's past history of learning due to environmental reinforcement, and free will are very important in determining how people are going to respond to a given situation. With people who have personality disorders in particular, however, to say that their living in a family war zone, as frequently described in this blog, is not a huge part of the problem seems to me to be the height of absurdity.

I thought of this issue recently after reading an article entitled “Ecological Momentary Assessment in Borderline Personality Disorder: A Review of Recent Findings and Methodological Challenges” (Santangelo, Bohus, & Ebner-Priemer, Journal of Personality Disorders 28 (4), pp. 555-576). 

Ecological Momentary Assessment (EMA) is a research technique designed to look at behavior and internal processes outside of the confines of what is called retrospective reporting. Retrospective reporting is the subjects' response to questionnaires about the way they normally respond in their daily lives - in hindsight.

People in studies using this technique are given a diary to fill out several times per day at regular, fixed intervals as they live their normal lives. They are instructed to record certain feelings and reactions they are experiencing. In the article’s abstract, it says that EMA is “characterized by a series of repeated assessments of current affective, behavioral, and contextual experiences or physiological  processes while participants engage in normal daily activities.”

As the authors reviewed the results of prior studies using this methodology in subjects with borderline personality disorder (BPD), one of those hidden assumptions I defined in a previous post just jumped out at me. The authors were inherently ignoring issues created by the fundamental attribution error. 

The definition of EMA in the article's abstract mentions “context,” by which I assume they mean the environmental context, but in the studies and in their discussion about them, the issue of environmental context seemed to be missing in action. The subjects were always asked about how they were responding, but almost never asked about what it was that they were responding to!

The authors’ literature review focused on five of the DSM’s (the official diagnostic manual of the American Psychiatric Association) criteria for BPD: 1. Affective instability. 2. Dissociation and transient paranoid ideation. 3. Interpersonal disturbances. 4. Self esteem disturbances. 5. Suicidality.

Now, one legitimate reason for doing these studies is to check on the validity of the diagnostic criteria for BPD, in which case descriptions about how the subjects’ families were behaving would be somewhat irrelevant. Since the diagnostic criteria were used to establish the diagnosis of BPD before the studies were even done, if the studies seemed to indicate that the criteria are turning out to be invalid, that would have to mean one of two things:

1.       Patients with BPD have been invariably lying through their teeth - on an impossibly consistent basis - in giving even superficial descriptions of their personal symptoms and experiences during diagnostic interviews ever since the syndrome was first recognized, or 
2.      The experimenters in the various earlier studies were lousy diagnosticians and were not applying the criteria in a valid manner.

Now, since I would assume that neither of these things was generally true, a finding that the subjects did not experience these symptoms would be most surprising. Of course, generally the subjects did experience the symptoms, although perhaps in some cases not quite in the generally accepted way. This sort of a conclusion is very close to being a tautology – that is, “a rose is a rose.”

But I digress. The authors clearly mention that some of the symptoms they are looking at occur in response to stress, but generally the subjects are not asked to describe the actual stresses to which they are responding. For instance, they say that subjects with BPD were found to be “more prone” to experience stress than controls. 

The problem with this is that it that assumes that the stressors that the controls are responding to are of equal frequency, severity, and nature as the stressors to which the subjects are responding. But no descriptions of those essential factors are presented. Perhaps if the controls were living in a more stressful environment, they would experience the stresses in a fashion more similar to that of the BPD subjects. 

Why are the subjects not also asked in their diaries to describe the stressors to which they are reacting? Is it all in their heads?  (It’s All in Your Head was the original title of my last book. Damn those academic publishers who thought that title was too colloquial). Or is it because therapists, like a lot of people these days, don’t want to look at what is actually going on in families?

Another issue is that, even if the diaries did ask about stressful interactions with intimates, and even if patients described them honestly and included their own behavior in their descriptions, the experimenters would still be in the dark about how severely stressful they were. That is because these interactions have subtexts, as I described in my post The Obvious Secret of Interpersonal Interactions Within Families. 

Words and behaviors during family interactions take on additional shades of meanings within the context of all prior interactions, and these meanings can significantly add to the stress level of the involved parties. In fact, without knowing the entire history of the patient's family interactions, the experimenter's judgments about the severity of the stress would by necessity be extremely flawed. 

As far as I know, there is only one method by which a mental health professional can obtain this data: long term psychotherapy with the involved individual. This should also include occasional conjoint sessions with the patient and family members, to get their sides of the story. The stressors of every single patient have qualities that are unique to them.

Without any descriptions of the nature of the stressors, we can not really come to valid conclusions. Of course, a possible assumption that should be made is this: people who are under severe stress are undoubtedly more likely to respond with more severe reactions than people who are under far less significant stress. 

Duh!

Dumb Hidden Assumptions in Drug Abuse Research

The mental health professions these days seem to want to blame their patients' repetitive problematic or self-destructive behavior on just about anything except what is, in the large majority of cases, the primary causes: family dysfunction and adverse childhood experiences (ACE’s). And I mean, they would rather it be almost anything else.

In my post of February 26, 2011, I discussed how a slight increase in aggressive thoughts following the playing of violent  video games by adolescents was translated by researchers into the games being a major risk factor for the development of youth violence. The fact that most compulsive video game players are inveterate couch potatoes who do not get out much never entered into discussions.

Not surprisingly, a recent longitudinal study (Fergus0n et. al., J. Psychiatr Res 2012; 46: 141-146), showed that, by taking other variables into account such as intra-family violence, the correlation between video games and even short-term aggression could no longer be established.  Another older paper from the same lead author (Ferguson and Rueda,  J Exp Criminol, 2009; 5:121-137) showed that aggressiveness in the laboratory, as expected, did not correlate with violent acts in real life.

Focusing on minor targets like video games risks leading social activists and public policy makers to ignore the far more important causes of youth violence like child abuse.

So of course, now that the tide is turning against the insane drug war against  marijuana, which has turned a significant percentage of the population into criminals (who tend to only be prosecuted if they happen to be African American), the folks who refuse to look at reality are now publishing "studies" that attribute a host of problematic behavior almost entirely to the devil weed – while all the while making the most ridiculous hidden assumptions imaginable. 

People who feel the need to be stoned all the time have enough problems; we do not need to make up a bunch of other ones.

In Carl Hart’s book High Price , he recounts his adventures as a reviewer of potential drug abuse studies for funding from the National Institutes of Health. He mentioned that the research agenda was being controlled by the National Institute on Drug Abuse (NIDA). He makes it clear that they were only interested in studies that showed the dangers of street drugs, not on studies which countered the many myths in the field that he had described in the rest of the book. (NIDA also ignores the dangers of the very same drugs they demonize when Pharma sells them for conditions such as "ADHD").

Now comes a study out of Australia and New Zealand: “Young adult sequelae of adolescent cannabis use: an integrative analysis” by Edmund Silins and others. (Lancet Psychiatry, 2014;

1: 286–93). Here is the abstract:

Methods: We integrated participant-level data from three large, long-running longitudinal studies from Australia and New Zealand: the Australian Temperament Project, the Christchurch Health and Development Study, and the Victorian Adolescent Health Cohort Study. We investigated the association between the maximum frequency of cannabis use before age 17 years (never, less than monthly, monthly or more, weekly or more, or daily) and seven developmental outcomes assessed up to age 30 years (high-school completion, attainment of university degree, cannabis dependence, use of other illicit drugs, suicide attempt, depression, and welfare dependence). The number of participants varied by outcome (N=2537 to N=3765).

Findings: We recorded clear and consistent associations and dose-response relations between the frequency of adolescent cannabis use and all adverse young adult outcomes. After covariate adjustment, compared with individuals who had never used cannabis, those who were daily users before age 17 years had clear reductions in the odds of high school completion (adjusted odds ratio 0·37, 95% CI 0·20–0·66) and degree attainment (0·38, 0·22–0·66), and substantially increased odds of later cannabis dependence (17·95, 9·44–34·12), use of other illicit drugs (7·80, 4·46–13·63), and suicide attempt (6·83, 2·04–22·90).

Interpretation: Adverse sequelae of adolescent cannabis use are wide ranging and extend into young adulthood. Prevention or delay of cannabis use in adolescence is likely to have broad health and social benefi ts. Efforts to reform cannabis legislation should be carefully assessed to ensure they reduce adolescent cannabis use and prevent potentially adverse developmental effects.

Funding: Australian Government National Health

The authors claimed to control for confounding variables, but most of these "controlled" variables were unrelated to ACE’s or ongoing family chaos. They were such things as age, sex, ethnicity, socioeconomic status and mental illness. The authors did control for a few possibly relevant parental variables like alcohol use, tobacco use, divorces, and history of depression. But not for how the parents actually behaved around their children, how they treated their children, child abuse or neglect, how chaotic the home environment was, or how and how consistently the children were or were not disciplined.

What on earth makes people who draw the conclusion that the drug was the primary cause of the lower achievement become so stupid that they don't see that frequent drug use is a sign that the teens already had emotional problems before they even started smoking - and that it was these problems that predate the drug use that were the real cause of both the drug use AND the poor performance?

The authors used exactly one rather vague sentence in their discussion to refer to this possibility, which most readers will miss: “…cannabis use in adolescence could be a marker of developmental trajectories that place young people at increased risk of adverse psychosocial outcomes.” (p. 291). 

Ya think?