Bipolar 2 From Inside and Out

Did you hear the news?

Bullying, inflammation, anger, low self-esteem, abuse, biochemicals, unsettled gender identity, cat parasites, and anything bad causes depression/bipolar disorder/PTSD. (Double-jointedness, too, except I don’t know if that’s bad or not.)

But don’t worry. Reading, happy memories, cat videos, a new vaccine, or Tylenol can help!

Science reporting these days is confusing, deceptive, and sometimes just plain wrong. Perhaps science reporters don’t mean to mislead, but that’s exactly what they do. Among the problems are publishing demands, lack of knowledge, logical fallacies, and the difference between correlation and causation.

Let me explain.

Publishers demand big, catchy headlines, and they prefer “New Hope for Bipolar on Your Grocer’s Shelf” to “Experiments on Genes and Diseases Continue.” Add to that the fact that editorial budgets have been slashed and personnel shuffled around so much that todays’s “science reporter” may have been last month’s “political correspondent” – and trained only in basic journalism, if that.

Science is complicated and difficult to understand, unless you’ve got special training. Even then, your expertise is likely to be only in one area – the microbiology of prostate cancer in mice, for example. And most people’s understanding of how scientific research works is, well, not understanding so much as knowing that DNA is somehow like a fingerprint.

Here’s a website with videos that tackle the subject quite nicely. My personal favorite is the one about animal trials in research, which explains (among other things) why my father, who had bone cancer, always said he was tired of being compared to a white rat.

Another problem is argument by analogy, which appears more in opinion pieces than in stories labeled as science. But here’s a sample, damning research on psychotropic drugs, written by Kelly Brogan, MD, ABIHM (American Board of Integrative Holistic Medicine).

The most applicable analogy is that of the woman with social phobia who finds that drinking two cocktails eases her symptoms. One could imagine how, in a 6 week randomized trial, this “treatment” could be found efficacious and recommended for daily use and even prevention of symptoms. How her withdrawal symptoms after 10 years of daily compliance could lead those around her to believe that she “needed” the alcohol to correct an imbalance. This analogy is all too close to the truth.

Well, no it’s not, for a number of reasons. Analogies always break down after a while, some sooner than others. For example, that hypothetical six-week trial would be longer than six weeks, come only after years of animal studies (including ones that focused on unwanted side effects like, I don’t know, hangovers or liver damage). The trial would have included control groups, placebos, and other research protocols. The ten-year “withdrawal” effect wouldn’t show up in six weeks; people who drink only two drinks per day are not generally considered alcoholics or go into withdrawal (indeed they may be drinking wine for heart health).


And so forth. Having two drinks per day is not analogous to taking a medication for social anxiety disorder. It’s associating the disliked thing (psychotropic meds) with a thing known to be bad (alcoholism) and damning by association.

Here’s another flaw: the principle that “correlation does not equal causation.” The classic example comes from the 1950s, when it was claimed that rock and roll music would lead to teenage pregnancies. It’s true that some teenagers who listened to rock and roll became pregnant (correlation). But some didn’t. And some teenagers who listened to country or jazz became pregnant. And I think by now we know what really causes pregnancy (causation).

This problem is illustrated by an article, “Scientists: The ‘Tortured Artist’ Is a Real Thing.”
The first thing to notice is that the headline is misleading, or possibly completely untrue. The article explains a study that supposedly shows that “creative genius and mental disorders are connected at a genetic level,” then goes on to debunk it:

“Any particular set of genes is only going to explain a very small part of variation in any psychological trait,” says Scott Barry Kaufman, a psychologist at the University of Pennsylvania. Indeed, the variants in the new study have a tiny, miniscule impact on creativity – less than 1 percent.

The rest of the article waffles back and forth and concludes inconclusively. Are creativity and “madness” linked somehow? Possibly. Does one cause the other? We don’t know, but there are a lot of theories. There are lots of other possible factors. Without the headline, would anyone read that? How do you define “creativity,” anyway?


Here’s a selection of recent articles that purport to have some relevance to mental illness or mental health.

Can Reading Make You Happier?
Answer: Bibliotherapy helps some people, possibly because of changes in the brain.

Artificial Recreation Of Happy Memories May Become The Next Big Weapon Against Depression
Thesis: “Urging a depressed person to stay positive by remembering the good things in life is unlikely to be helpful advice. That is because depression blocks access to happy memories. But what if we could somehow artificially recreate such memories to allow for some more positive thinking? A study suggests that this is indeed possible – at least in rats….However, more research will be necessary to obtain a clearer picture of how this might work in humans.”
(Again with the rats.)

Science Shows that Watching Cat Videos is Good for You
The article, which says “research suggests that the pleasure you derive from watching cat videos can often outweigh the guilt of procrastination,” is largely tongue-in-cheek, but that headline is a grabber. Headlines that use “waffle words” like may, can, might, possibly, someday, appears to usually indicate a story that says nothing significant. They build up hope, but if a study comes along that disproves the theory, it will never be reported.

Researchers Are Developing A Vaccine For Post-Traumatic Stress Disorder
It begins: “New studies are suggesting a link between the immune system and the way the body reacts to stress. Research with rodents are raising hopes that one day, tweaking a person’s immune system could be a way to treat or even prevent conditions like PTSD, Nature reports.” How many warning signs can you spot there?

Double-Jointedness Is Linked to Anxiety
One of my favorites. Correlation/causation much?

Your pain reliever may also be diminishing your joy
Actually it says that acetaminophen blunts both positive and negative emotions. And, it adds, “this study offers support to a relatively new theory that says that common factors may influence how sensitive we are to both the bad as well as the good things in life.” Gee, who would have guessed?

Is Depression a Mental or Physical Illness? Unravelling the Inflammation Hypothesis
This is actually a good article. The headline question is a valid and interesting one, and the author states, “But while there may be a connection between inflammation and depression, one doesn’t necessarily lead to the other. So it’s too simplistic to say depression is a physical, rather than a psychiatric, illness.” Hooray for correlation/causation!

What does all this reporting prove? Almost nothing. Except “let the reader beware.”

Comments on: "New Hope for Mental Illness" (3)

  1. I think all students should be required to take a course in scientific method and critical thinking in high school. It might not fix the understanding science problem, but it could at least give everyone a common frame of reference.
    My favorite reference for “the plural of anecdote is not data” education is You can even create your own science conspiracy theories!


  2. brightonbipolar said:

    Excellent post – very informative!


Comments always welcome!

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Tag Cloud

%d bloggers like this: