by Michael D. Anestis, M.S.
The summer before I began graduate school, I received a package in the mail from Florida State University. The package contained my first assignment as a graduate student, which consisted of reading two articles by Paul Meehl and a selection of work by Karl Popper. A page into the Popper document, I found myself wondering what I had gotten myself into. The text was fairly dense, the material was hard to connect with, and I couldn't help wondering what the heck it had to do with mental illness and psychotherapy.
Fortunately, we devoted a good amount of time to discussing the paper early in those first few weeks of graduate school and it resurfaced on a number of other occasions, including during the oh-so-fun preparation for comprehensive exams. That being said, my understanding of Popper's work has expanded and my appreciation for its importance has become almost limitless.
The Popper piece we read was entitled "The logic of scientific discovery," and its central aim was to explain how to apply logic in science while outlining the dangers of straying from these principles. My goal here today is to attempt to talk about this paper in a way that will be engaging and to hear your thoughts on Popper's conclusions. Admittedly, this stuff is not the kind of thing most of us turn to when we are seeking entertainment, so I'll do my best to keep it interesting, but it might be a difficult task. I'll break the article down into sections in order to try to make the material more manageable.
Inductive logic
Popper opened his work by discussing the nature of inductive logic. To use his own words (p.27), "it is usual to call an inference inductive if it passes from singular statements (sometimes called particular statements), such as accounts of the results of observations or experiments, to universal statements, such as hypotheses or theories." So, in other words, inductive logic is when we see something small and assume it applies on a grand scale: I see one white swan and conclude that all swans are white.
How might we apply this to mental illness? Well, for example, a therapist might see a client who engages in non-suicidal self-injury (e.g., cutting) who also has a history of childhood sexual abuse. From there, the therapist might conclude that self-injury is a sign of having been sexually abused as a child.
A lot of times, our inductive conclusions are correct, but Popper noted that any conclusion drawn from this type of thinking may turn out to be false upon further investigation. In other words, no matter how many white swans we see, the next one might be black and no matter how many clients we see who self-injure and have a history of childhood sexual abuse, the next client we see who self-injures might not have any such history. Still, the more times we see things, especially things that are intuitively appealing to us, the more we believe they are true. Of course, as your sample size grows, your confidence becomes increasingly justifiable, as the likelihood that the results are due purely to chance decreases, but inductive conclusions are not based upon a systematic observation of repeated events, but rather from singular statements that stand out in our mind.
Is inductive logic always a problem?
You might look at what I wrote just above this heading and think "you know, I need to think that way sometimes, so we can't just dismiss it." Popper would agree with you. His point is not that inductive logic is bad, but rather that it has a specific role in the scientific process and that it's role should not expand beyond that.
When we are coming up with new ideas, we need to think this way. The creative process that leads to new ideas and drives science forward is based upon people looking at smaller phenomena and wondering whether or not they generalize on a broader scale. Popper calls this the "context of discovery," which he believes is the first step in the scientific process. The problem, as he sees it, is that many people stop there. They develop a theory and assume that it is true.
The better way forward, according to Popper, is to take that new theory and test it, using deductive logic in what he terms "the context of justification." So, to put it more simply, on any given day, a lot of smart people make observations about the world and develop theories about what those observations tell us. Some of those theories are right, some are wrong, and some are somewhere in between. If we simply stick with our theory and go no further, than a number of competing ideas will never compete and we will not uncover the truth. If, however, we take our theory and test whether or not it holds up, we can really get somewhere.
Truth?
I used a small word with heavy meaning in that last paragraph: truth. If we accept the idea that some theories are more accurate than others, we need to have some sort of system for determining which is which, otherwise people are left to simply persuade others to believe them. Popper argues that we step closer to "truth" each time a single observation is "inter-subjectively tested." In other words, the more times people independent of ourselves can replicate our finding, the more confident we can become that it is true (although we can never actually prove that this is the case, but we'll get to that more in a bit).
So how does this apply to mental health? Simply because one study supports the use of a particular form of therapy for a particular condition does not mean the results will always be so favorable; however, as more and more studies repeat the finding, we can become more certain that the particular form of therapy is, in fact, useful in treating that condition. Of course, some studies are better than others, so we have to be careful in drawing our conclusions, but the point still remains. Some forms of therapy have a lot of inter-subjectively tested support for their utility in certain contexts and some do not. This does not mean that those with support are perfect, but it does mean that we have built up an evolving foundation of support to believe that that therapy is taking us in a good direction.
What this essentially comes down to then is hypothesis testing. We develop a sense that we can use a particular piece of information (e.g., how uncomfortable an individual is with bodily sensations associated with anxiety) to predict an outcome (e.g., number of panic attacks over a particular period of time). Rather than just accepting that this is true, however, we test it. So, using the example in parentheses, we might recruit a large group of individuals and assess their levels of anxiety sensitivity (fear of the symptoms of anxiety, particularly bodily sensations). We might then bring those folks back six months later and assess the number of panic attacks they had over that period. If folks with higher levels of anxiety sensitivity on average had a greater number of panic attacks, our hypothesis will be supported. This is only one study though, so before we became confident, we would need to see other people not associated with our research group reproduce that finding repeatedly and reliably.
So what makes for a good theory?
Popper laid out four main principles we need to keep in mind when developing a theory:
- The predictions are internally consistent. In other words, if you are predicting that individuals with high levels of anxiety sensitivity will have higher rates of panic attacks, you can not then predict that they have lower rates of panic disorder.
- The theory is falsifiable. This one is a big deal. Every year, my beloved Pittsburgh Pirates vow that they will "improve," however, they do not define improvement, meaning that when they somehow lose more games than they did the previous year, they can claim improvement in "attitude," "clubhouse atmosphere," "potential" or some other elusive and difficult to measure variable. We then can not hold them accountable to any real extent. This is a big deal in mental health too and you've seen the issue come up in some of the back and forths about recent PBB articles. Some theories of mental illness (e.g., your symptoms stem from some unconscious conflicts) and psychotherapy (e.g., therapy must work to resolve unconscious conflicts) can not be falsified. In other words, there is no way to tell if they are doing what they claim or not, so we are left with subjective evaluations of abstract ideas. Now, this does not mean that those ideas are wrong, but it raises very serious questions about the degree to which the theory is helpful. In other words, if there is no way to test whether or not a theory is wrong, it's value is drastically decreased, particularly if alternative theories exist that do allow such testing.
- The theory is not redundant. A good theory provides us with new information rather than simply repackaging something we already know.
- The theory stands up to empirical tests. In other words, if the theory predicts a particular outcome will occur more often in particular circumstances, we should be able to demonstrate that this is, in fact, the case.
So, if a theory does all this, does that mean we know that it is the "truth?"
Popper was very careful to clearly explain that, when a hypothesis is confirmed, that does not mean that it is definitively true. It simply means that, for now, there is reason to dismiss it. Popper envisioned a marketplace of competing ideas. These ideas are pitted up against one another in battle. Typically, one idea proves itself superior to the other, but flaws in the winning idea are revealed as well. In response, the superior idea is adjusted to account for those flaws and the newer refined version is tested. This process repeats itself over and over again without end, with each new test producing slightly more knowledge.
So, to put it more simply, scientific research in clinical psychology does not tell us things are definitely true, but the more often a finding is replicated independently using strong methodology, the more confidence we become that it is not entirely false. As weaknesses are revealed, we adjust our guesses and try out new models to see if we can attain better results. So, in the long run, science is a never ending march towards an unobtainable goal: perfection. We build piece upon piece to create a more well developed picture, but we have to use caution to make sure that what we say is true does not step beyond the bounds of what the evidence actually tells us - an issue that has also come up quite a bit in recent PBB articles and the comments that followed.
Popper explained this as follows (p.33): if "the singular conclusions (e.g., predictions) turn out to be acceptable, or verified, then the theory has, for the time being, passed its test: we have found no reason to discard it. But if the decision is negative, or in other words, if the conclusions have been falsified, then their flasification also falsifies the theory from which they were logically deduced. It should be noticed that a positive decision can only temporarily support the theory, for subsequent negative decisions may always overthrow it. So long as a theory withstands detailed and severe tests and is not superseded by another theory in the course of scientific progress, we may say that is has "proved its mettle" or that it is "corroborated."
So what kinds of predictions should we be making?
Popper notes that the more straight forward and specific the prediction, the more valuable it is. In other words, if I say it will be "warm" tomorrow, that offers up a wide range of temperatures as fitting my criteria, some of which others might not consider warm. On the other hand, if I say it will be 74 degree tomorrow, there is not much wiggle room.
Let's put this in the context of mental health. If I say that talk therapy will be useful for a client, there are any number of ways I can look at any result and claim that they support my case. On the other hand, if I say that cognitive behavioral therapy for binge eating disorder will reduce the number of binge eating episodes, help the client develop a set pattern of healthy daily eating habits, and reduce the client's self-reported difficulty with using healthy behaviors to decrease negative emotions, there are very clear cut, well-tested methods for evaluating whether or not my prediction was correct. In other words, in the world of science, the more vague we get, the worse off we are and the more subjective we get, the worse off we are.
Summary
So what's my point in all of this? Clearly I chose a bit of a difficult topic today, so hopefully there has to be a reason, right? My intention today was to walk you through the manner in which scientifically-minded folks approach their understanding of mental illness and psychotherapy. We do not claim to have special knowledge based upon intuition and philosophies. Instead, we develop ideas, test whether or not they translate to the real world, and adjust them according to the facts as they present themselves through systematic observations. Our conclusions are only temporary, as every year somebody develops something new that causes us to shift our focus a bit and think about things a little more clearly. We do not twist results though vague commentary to make it seem as though they support our results. Instead, we present the data and, while errors in interpretation are made, the availability of the data empowers people to call our conclusions into question. This process is often a slow grind full of esoteric results on topics that are uninteresting to most, but each of those little pieces contributes to how we understand things.
By all means, there are aspects of life that are difficult or even impossible to measure and, as such, it is difficult to apply these principles to those issues. That being said, we can - and regularly do - apply these principles successfully in our studies of psychotherapy and mental illness and the efforts to do so have resulted in amazingly positive outcomes. Not everyone is going to think like a scientist, but hopefully as you figure out your own viewpoint, you will do so on the basis of a clear and accurate understanding of what science actually is and what scientists actually do.
I'll leave you today with one last quote from Popper himself (p.38):
"I hope that my proposals may be acceptable to those who value not only logical rigor, but also freedom from dogmatism; who seek practical applicability, but are even more attracted by the adventure of science, and by discoveries which again and again confront us with new and unexpected questions, challenging us to try out new and hitherto undreamed-of answers."
************
If you would like to learn more about the topics discussed on PBB, we recommend that you consult our online store of scienifically-based psychological resources.
Mike Anestis is a doctoral candidate in the clinical psychology department at Florida State University
Subscribe in a reader
