As a clinical psychologist, I thought that a good way to learn about psychology education at UNC-Chapel Hill would be to look at the research interests of its psychology professors.
I started with the chairman of the department, Jonathan Abramowitz, who is also director of the university’s anxiety treatment program. In March 2010, Dr. Abramowitz and a graduate student he advises, Noah Berman, and two others published an article in Behaviour Research and Therapy entitled “The Relationship between Religion and Thought-Action Fusion: Use of an In Vivo Paradigm.” The title caught my eye because it seemed relevant to my interest in how religious views are affecting higher education. But I was unprepared for the prejudice against Judaeo-Christian faith and the psychological mistreatment of undergraduates committed by this research.
The article describes an experiment to test the hypothesis that highly religious Protestants are more likely to have an obsessive-compulsive tendency than agnostics and atheists. To test that hypothesis, undergraduates were instructed to name a beloved family member and then imagine experiencing incest with that person. They were also told to imagine the family member being in a car accident.
The experiment was designed to examine two kinds of a mental process called thought-action fusion (TAF). The researchers associate TAF with obsessive-compulsive disorders (OCD), and define it as “faulty beliefs about the relationship between mental events and behaviors.” “Moral TAF” is the belief that thinking unacceptable thoughts is the moral equivalent of performing unacceptable behaviors. “Likelihood TAF” is the belief that thinking about an event increases the probability of the event or even causes it to occur.
In the experiment, more Christians were found to have “dysfunctional” TAF than agnostics and atheists. If Christianity correlates with TAF, and TAF correlates with OCD, presto! Berman, Abramowitz, et al. conclude that their experiment supports “the idea that Protestant Christianity includes certain beliefs that may be, in the clinical literature, associated with obsessional complaints (i.e., OCD).”
Here’s how the experiment worked. Two hundred fifty-three undergraduates took a battery of tests, including testing for “religiosity.” Forty-three were selected as highly religious Protestants (HRP), and 30 were selected for an atheistic/agnostic (AA) group. The following description of the experimental procedure is reproduced verbatim from the article. After the participants had rated their level of anxiety at the moment,
participants were asked to think of a close (and beloved) relative, such as a parent or sibling, and write the person’s full name on a note card that was provided. The experimenter then placed the note card next to the computer monitor. Participants were then presented with one of two thoughts that were in the form of a sentence displayed on the computer screen (in counterbalanced order) designed to activate TAF beliefs: “I hope I have sex with __________” and “I hope __________ is in a car accident today.”
Participants were instructed to copy the sentence onto another note card, inserting their close relative’s name into the blank. Immediately after writing each sentence, participants were asked to close their eyes and think about the event occurring. Then, they were asked to . . . indicate (a) their current level of anxiety, (b) their perceived likelihood of the event occurring because of thinking and writing down this thought, and (c) the perceived moral wrongness of thinking and writing down the thought.
After completing these three in vivo ratings, a prompt appeared on the screen that read, “You may now do anything you wish to reduce or cancel the effects of writing or thinking about the sentence.” . . . The experimenter recorded whether or not the participant then performed any such neutralizing behavior (e.g., crossing out the sentence, turning over or tearing the note cards, etc.) in response to this prompt.
Ethical Problems
The American Psychological Association provides the guiding standards for protecting subjects from psychological harm. The experiment appears to have violated at least four fundamental requirements put forth in the APA’s “Ethical Guidelines for Research with Human Subjects.”
First, participation as a research subject is mandatory for an undergraduate degree in psychology at UNC-Chapel Hill. The APA guidelines specifically cite the need for extra safeguards when “the participant is a student, client, or employee of the investigator.” The Abramowitz et al. article does not specify how many participants were psychology majors. For those who were, their participation was especially ethically suspect.
Second, before participating, the undergraduates were told that the experiment was “ostensibly about ‘thoughts and feelings’.” If they were not informed that they would be thinking and writing about disturbing—possibly traumatizing—thoughts, the statement “informed consent was obtained before beginning each participant’s session” is misleading.
The article also states that the participants were fully debriefed following the experiment. “After completing the procedure for both thoughts, participants were debriefed about the actual purpose of the study.” Were they told that the research was designed to find a relationship between Biblical teaching and mental illness? That as Protestants they are more likely to suffer from OCD? The article does not address these important aspects of full disclosure and a participant’s right to decline or withdraw based on “physical and mental discomfort, harm, and danger that may arise from research procedures.”
Third, the authors characterize their method as evoking “harmless unwanted thought[s].” It is remarkable for psychologists to characterize abhorrent thoughts and words as harmless. Incest is one of the most shame-producing and traumatic forms of sexual abuse, with permanent and far-reaching psychological effects.
The authors state that this was the first time vivo procedures, which are designed to elicit actual emotional distress, have been used in this kind of experiment. The long-term psychological effects of this research are unknown. What if one of these students develops misgivings about having submitted to this research, as happened in the infamous Stanley Milgram experiments testing submission to authority? If Dr. Abramowitz has children, does he believe it is harmless to imagine and write about having sex with them? Would he ask them to imagine and write about having sex with him?
As a clinician, I have worked with many incest survivors. Incest is not merely a negative life experience; it is usually a shattering trauma. Statistics on the rates of sexual abuse and assault suggest that several of the 73 undergraduates participating in this research have been or will be victims of some form of sexual trauma. The article, however, does not indicate whether any effort was made to screen out such victims. Also, with a mean age of 20, these young people may not be able to fully understand or report traumatic history.
And what if one of these students or their loved one is involved in a car wreck? These psychologists cannot know whether their methods will worsen post-traumatic effects, including guilt.
Fourth, the article states, “All procedures had been approved by the university IRB.” The IRB (Office of Human Research Ethics Institutional Review Board) is the body that approves research procedures at UNC-Chapel Hill. If the IRB reviewed detailed procedures, it needs a refresher course in the ethical guidelines for psychology research. If the procedures reviewed were not complete, the researchers have morally and legally exposed UNC.
Biased Assumptions
The authors’ anti-Christian bias becomes evident in the second paragraph of their article. They are looking for TAF in “religious institutions which impose explicit moral standards for thinking and behaving, which are inculcated by authority figures (e.g., clergy) and include the possibility of punishment (e.g., damnation) [that] might foster the development of rigid and maladaptive beliefs about thoughts and their influence.” To demonstrate the relationship between Christian morality and maladaptive TAF, they cite the 10th commandment against coveting and Jesus’s words from the Sermon on the Mount, “I say to you that everyone who looks on a woman to lust for her has committed adultery already in his heart”(Matthew 5:2-28). Later, they cite two more examples: John 3:15, “Anyone who hates his brother is a murderer,” and Proverbs 23:7, “For as he thinks in his heart, so is he.”
The participants were given a chance to use “neutralizing behaviors,” to recover from their “harmless unwanted thought[s].” But the use of neutralizing behaviors was viewed by the researchers as a significant indicator of “maladaptive” TAF in Christians. And the HRP group used more neutralizing behaviors than the AA group. If imagining incest and car wrecks is harmless, why did the participants have to neutralize the effects in order to recover? One subject was so traumatized she “took a picture of her loved one out of a purse and held it to her heart.”
The modern field of psychology was based as much as anything on Freud’s discovery of the anxiety associated with unwanted sexual thoughts, especially those related to parent/child incest. If it is harmless to write and think about having sex with your mother or sister, than why is any pornography or violent image harmful? Why not express every crude or cruel thought that comes to mind?
In looking for the ill effects of religion, the authors ignored other plausible inferences. The same experiment could have been used to test the hypothesis that neutralizing behavior is a positive effect of religion, an attempt to negate repugnant and violent ideation. For example, repentance, rather than being a “compulsive ritual” associated with TAF, may be a mentally healthy strategy.
Finally, the use of the word religion in the article’s title, “The Relationship between Religion and Thought-Action Fusion: Use of an In Vivo Paradigm,” is over-inclusive. The authors themselves appear uneasy about it. In an amazing paragraph at the end of the article, they wrap their anti-Christian bias in a disclaimer and acknowledge that religion in general may not be the only cause of TAF. But, unable to shake their bias, they then get back to their conclusion that Protestants are especially prone to OCD.
From a scientific perspective, the quasi-experimental design of this study precludes causal inferences. . . . Yet in spite of this, even if we presume that religion does promote TAF, this is not exclusive to religion. . . . Lastly, while Protestant doctrine may emphasize certain TAF-like beliefs and attitudes, other religions may not . . . . Thus, the strongest conclusion that can be drawn is that the data are in concert with the hypothesis that certain aspects of Protestant Christianity correspond with TAF beliefs, and may therefore be correlated with obsessive-compulsive symptoms.
Conclusion
Using the same data as this study did, the case could as easily be made that feeling uncomfortable about incest and a statement wishing harm to a family member is a positive indicator of mental health. Having little or no anxiety in response to highly threatening thoughts or feelings is generally considered to be symptomatic of mental illnesses called dissociative states. Minus the anti-Christian bias, this study could have been used to show not that Protestant Christians are obsessive, but that atheists/agnostics are dissociative and unable to respond in a healthy manner to extremely negative ideation.
The UNC-Chapel Hill psychology department spent public funds and risked the psychological well-being of undergraduates to discover that devout Christian students exhibit more discomfort after thinking about incest than do other students, and therefore may be more prone to mental illness. Do parents send their children to UNC to participate in this kind of research? Do North Carolinians understand the biases of the psychologists they are supporting? Do North Carolina taxpayers want to pay for this stuff?