by Michael D. Anestis, M.S.
****************
A man was running laps around a track at his local high school on a Saturday morning. It was early, so most people were still asleep, putting off exercise or errands for another couple of hours in order to catch up on some sleep. The man was excited about a quiet run in the cool air as a healthy start to his day. Part way through his first lap, however, he noticed something odd. Another man was standing alongside the track by himself, clapping his hands. This other man was not watching him run, he was simply standing there clapping. He decided to ignore him and just keep running. After a few laps, however, his curiosity became too much and he stopped beside the clapping man and approached him.
"Is everything okay" he asked?
"Oh yes. Very good," said the clapping man.
"That's good. Can I ask why you're standing here clapping?"
"Certainly. I'm clapping to keep the tigers away." This response surprised the running man, who took a few seconds to ponder his next move before replying.
"But there are no tigers in this area of the world."
"Exactly," said the clapping man.
*****************
I often tell some variation of this story to my Abnormal Psychology students at some point during the course of the semester and, quite quickly, they get the point. Oftentimes, we have a tendency of seeing relationships between phenomena when, in fact, no such relationship exists (or if it does, it is substantially weaker than we believe it to be). In this case, the clapping man believed that clapping would keep tigers away and, in fact, no tigers appeared while he was clapping. Of course, no tigers would have appeared had he not clapped either, but a sloppy glance at his data - clapping present/tigers absent - led to the formulation of an illusory correlation. Now, in reality, very few of us make logical errors as absurd as that one, but our vulnerability to illusory correlations is not as small as you might think and it plays a pivotal role in many areas of life, including mental illness.
Day-to-day life
One example of illusory correlations that impacts many people on a daily basis is the impact of choice in the lottery. Studies have shown that, when an individual is given the option to choose his or her numbers, that person is more likely to play the lottery than if the numbers are assigned arbitrarily. Additionally, if an individual chooses his or her numbers, that individual is less likely to be willing to sell back the lottery ticket - even above cost - than if the numbers had been assigned arbitrarily. Of course, there is absolutely no correlation between choosing your numbers and winning the lottery unless you have inside information on what the numbers are going to be, so in this case, behavior is driven by a false belief in a relationship between choice and outcome.
Another common example of this is the phenomenon of "odd match." The best example of this is another scenario I frequently present to my students. Have you ever thought about a friend with whom you have not spoken in a long time and then, right after thinking about that person, he or she called/emailed/etc...? You probably have and, when it happened, you probably thought "wow, that was amazing." Now...have you ever thought about a friend with whom you have not spoken in a while and then, immediately after thinking about that person, he or she did not call/email/etc...? I suspect the answer is yes and that, in fact, this has happened far more often than the first scenario. Sometimes, we see a connection between two things - thinking about a person and hearing from them - that does not really exist, but because the occurrence was novel or exciting, we forget all about all of the times that a different result occurred and we fail to consider all of the available data before drawing conclusions about the phenomenon.
Recently, I have been reading a lot about obstetrics and natural child birth. Before I began this reading, I had no idea that this component of the medical world is plagued by the same battle facing clinical psychology: empirical evidence versus folklore and intuition. Generally speaking, when we think of the medical profession, we assume it consists of well-tested, scientifically-based decision making procedures and intervention protocols. I certainly do. In most cases, this assumption is valid. In obstetrics, however, this is often not the case. The degree to which this is true is way beyond the scope of this article (I recommend reading "The Thinking Woman's Guide to a Better Birth" or viewing "The Business of Being Born" if you are interested in learning more, as both are based heavily upon data and present both sides of the story despite obvious slants on what they believe to be the best approaches). Let me provide you with an example though.
Henci Goer, the author of "The Thinking Woman's Guide to a Better Birth," who includes an extensive summary of the literature on every topic she covers, devoted an entire chapter to the topic of electronic fetal monitoring. Because this topic is outside my area of expertise, I will demonstrate the impact of illusory correlations here by quoting the author herself:
"The basic premise behind electronic fetal monitoring (EFM) is that insufficient oxygen (hypoxia, asphyxia) in labor is a common cause of severe mental retardation, death, and especially cerebral palsy, and that changes in the fetal heart rate precede brain damage. Based on that premise, obstetricians reasoned that intermittent listening had been unable to prevent brain injuries because it provided too little information too late. The solution, then, became a machine that made a continuous tracing of the fetal heart rate and how it reacted to contractions. Unfortunately, however, the premise was wrong on both points."
"To begin with, less than 10 percent of cases of cerebral palsy or mental retardation during or shortly after birth result from oxygen deprivation during labor. Even when doctors suspect that lack of oxygen played a role, often other factors did as well. Consider the following: the number of cases of oxygen deprivation in labor has declined steeply since 1979. If birth asphyxia were largely responsible for cerebral palsy, the cerebral palsy rate should have declined too. It hasn't. Moreover, continuous EFM has become nearly universal since the 1980s. If it worked, it should have affected the cerebral palsy rate by now, but the rate remains unchanged." (p.87)
"Second, there really isn't much connection between what the baby's heart rate does during labor and measures of the baby's condition at birth, such as the pH of the baby's blood or the baby's Apgar score. Few babies diagnosed with fetal distress are born in poor condition. A bedrock truth of EFM is that if the monitor says the baby is fine, the baby is almost certainly fine, but if the monitor says the baby is not fine - that is, that she has nonreassuring heart rate patterns - the baby is also probably fine." (p87-88)
"EFM's high false-positive rate also reinforced the belief that it works. "Fetal distress" appears, the obstetrician performs a cesarean or forceps delivery, the baby is born healthy, and everybody thinks that EFT saved the baby - only, of course, the baby was fine all along."
So...even highly educated individuals motivated to ensure the health of their clients see illusory correlations and subsequently engage in illogical behavior. In this case, a worthy cause - reducing rates of mental retardation and cerebral palsy - led to the creation of an intervention. That intervention, it turns out, does not produce the desire effects; however, a crude look at the data without an understanding of the high rate of false positives - EFM indicates distress/emergency procedures implemented/healthy baby born - leads to the belief that the intervention saved the day when, in fact, the fate of the day was never in question and the intervention, in reality, actually introduced a substantial amount of unnecessary risk to both the mother and the baby.
Mental health
The impact of illusory correlations on mental health is actually hinted at in the clapping man example, as it is a particularly prominent issue for individuals with anxiety disorders. Anxiety is characterized by avoidance behaviors and safety aids. We believe that some horrible consequence is looming, so we look for ways to change the situation. As it turns out, however, the horrible consequences that anxious individuals fear almost never actually occur, regardless of what steps are taken to avoid them.
For example, individuals with social anxiety disorder are characterized by a fear of being negatively evaluated by others. While working in a community campus, I encountered several clients who believed that, if they walked near large groups of people, others would look at them and dislike them, harshly judging their movements, outfit, and general persona. As a result, they avoided walking in areas of campus in which students typically congregated. As you can imagine, this led to a fairly restrictive routine, but the individuals also never experienced the overt judgment they had feared, so they came to believe that their avoidance of crowded areas was the cause of their lack of exposure to such ridicule. Being a scientifically-minded therapist, I engaged these clients in cognitive behavioral therapy (CBT), which includes exposing clients to their feared stimuli and testing their hypothesized fears. In this case, I not only had clients walk through crowded areas of campus, I also had them do things like put a spot of ketchup high up on their face where it made no sense for ketchup to be or walk oddly or do a spin move halfway across the courtyard or approach an individual directly in front of the library and ask them if they know where the "book place" is. In almost every case, my clients found that other people completely ignored them and, in the other cases, they found that other people simply noticed the odd behavior briefly and moved on. In other words, even when they chose not to avoid and, in fact, grossly exaggerated normal participation, nothing bad happened. Until they were asked to test their beliefs, however, these individuals were driven by illusory correlations - the false belief in the connection between avoiding crowds and preventing humiliation.
What does this say about the importance of science in mental health care?
To wrap this topic up, let me bring it back to our ongoing PBB conversation: why science is such a vital component of clinical psychology. No matter how smart we are, how well educated, experienced, or well-trained we are, we all fall victim to illusory correlations. We believe things, attend to information that supports our beliefs, and discount evidence that contradicts our beliefs. As a result, we see relationships where they don't exist and we act on faulty premises. What this means is that, without empirical evidence that objectively evaluates whether or not what we believe is true, we are vulnerable to making frequent mistakes, some of which will be larger than others in scope. In our day-to-day life, this is often not a big deal. After all, the phone call example I gave in the beginning of the article does not hurt anyone and often leads to a great story. In mental health care, however, it is a very big deal. If we let our intuition overrule data as we make treatment decisions, we are putting the health of our clients at risk. In doing this, we are assuming that, for some reason, we are not vulnerable to the weaknesses inherent in the human condition. The way we view things before we look at clear results (and interpret them using legitimate methods) is often not a reflection of the way things really are. That is not an insult, but rather a simple acknowledgment of what it means to be human. Fortunately, by building off data accumulated through years of research, we can counteract those limitations and enhance health care. This is, in fact, one of the underlying principle of empirically supported treatments.
*******************
If you would like to learn more about how errors in thinking play an important role in clinical psychology, we recommend the following items, all of which are available through our online store of scientifically-based psychological resources:
- Science and Pseudoscience in Clinical Psychology
by Scott Lilienfeld, Steven Lynn, and Jeffrey Lohr
- 50 Great Myths of Popular Psychology: Shattering Widespread Misconceptions about Human Behavior
by Scott Lilienfeld, Steven Lynn, John Ruscio, and Barry Beyerstein
- The Great Ideas of Clinical Science: 17 Principles That Every Mental Health Professional Should Understand by Scott Lilienfeld and William O'Donohue
- How To Think Straight About Psychology by Keith Stanovich
Mike Anestis is a doctoral candidate in the clinical psychology department at Florida State University
Subscribe in a reader
