Jerry Burger set out to investigate whether a contemporary study of obedience would yield results comparable to what Stanley Milgram found four decades ago. The January American Psychologist provides a thought-provoking account of Burger's results.
I write this month to encourage "Ethics Rounds" readers to find their January American Psychologist and examine a fascinating series of articles. Jerry Burger, of the Santa Clara University department of psychology, set out to investigate whether he could replicate the findings of Stanley Milgram's studies on obedience. Burger's thoughtful and thought-provoking article, along with five commentaries, should be read by students of psychology at every level of professional development. I hope Burger's work will lead and to a renewal of investigations into this area of human behavior and a thoughtful review of the ethical parameters of such experimental studies. Three of the many themes that emerge from these articles involve attention to ethical issues, the relationship between ethics and methodology, and the role of institutional review boards (IRBs) in regulating social science research.
Burger titles the first section of his article "Ethical Concerns." That Burger chose to begin his article in this manner underscores the centrality of ethics in discussions regarding post-Milgram studies on obedience, and ethics emerges as a major theme in four of the five commentaries on Burger's article. Burger's steps to safeguard participants' welfare began with an initial screening. A research assistant asked questions including whether the prospective participant had ever been diagnosed with a psychiatric disorder, was currently receiving psychotherapy or taking medication for emotional difficulties, had ever had problems with alcohol or substance abuse, or had experienced serious trauma such as child abuse, domestic violence, or combat. A positive response to any of these questions excluded the individual from participating; approximately 30 percent of prospective subjects were precluded on this basis. Prospective subjects were next given a series of scales that included depression and anxiety inventories. A clinical psychologist reviewed the responses and conducted interviews to identify individuals who might have a negative reaction to participating. Of the individuals who made it to this second stage of the process, nearly 40 percent were excluded. A clinical psychologist was instructed to end a participant's involvement immediately if any sign of excessive stress emerged. Burger emphasized repeatedly that participants were free to withdraw at any time and still receive their $50 payment. Debriefing took place immediately following a participant's session, so that participants learned within moments of their session's end that the confederate had not actually been shocked.
Beyond these safeguards, Burger altered Milgram's methodology in a manner that commentators found especially intriguing. Burger terms this alteration "The 150-Volt Solution." Burger observed that nearly four out of five people who passed the 150-volt mark in Milgram's study continued to administer shocks in 15-volt increments to the end of the scale, 450 volts. Burger reasoned that the 150-volt mark, the point at which the "learner" protests loudly and demands to end the study, could therefore serve as something of a watershed, the point which, if crossed, likely signals that the study participant will go the entire range of the generator's value. By assuming (with caution) that individuals who proceed beyond 150 volts would continue obeying instructions to administer up to the maximum shock, Burger concluded there was no need to continue the study after this point. He thus avoids the conditions under which Milgram's subjects experienced their most intense levels of distress. With the 150-volt solution, Burger attempts to accomplish two goals: employ the science of psychology to examine the phenomenon of obedience to authority, and simultaneously remove what many consider the most ethically problematic aspect of Milgram's study, the distress induced in the participants.
A second theme to emerge in the articles involves the relationship between methodology and ethics. Burger begins his discussion section by noting "People learning about Milgram's obedience studies often ask whether similar results would be found today. Ethical concerns prevent researchers from providing a definitive answer to that question." Advancing science and protecting the individuals with whom psychologists work are two of psychology's core values. Because every interaction with another human being raises at least the possibility of harm, avoiding harm entirely would end all research. On the other hand, allowing research to move forward with no restrictions whatsoever would risk harms that psychologists agree could not be justified by the knowledge gained. For this reason, our profession must find a point on that continuum that strikes the correct balance. One of Burger's important contributions is highlighting in an explicit manner the complexities of meeting this challenge.
In their commentary, Ludy Benjamin and Jeffry Simpson, while accepting that "most of the recent rules and regulations imposed by IRBs are reasonable, legitimate, and necessary," suggest it may be time to adjust where psychology has struck this balance for "studies high in experimental realism:"
Every day, distressing, difficult and discomforting events happen to tens of thousands of people from all walks of life. To better understand how bad events can be translated into good or better outcomes, we occasionally need to study how individuals manage and cope with negative or adverse situations in well-controlled experiments. The pendulum may have swung too far away from the use of well-designed, thoughtful, carefully developed, and potentially illuminating high-impact studies in social and personality psychology. One implication of Burger's (2009) article is that it may help reopen this important discussion.
Noting in his commentary that "Research on (behavioral) obedience to authority has been virtually nonexistent in recent years, a peculiar cessation given the unparalleled and nonabating interesting in Milgram's research," Arthur Miller discusses the impact of Burger's methodology on his results. Miller calls the 150 volt level "both imaginative and convincing," yet addresses how Burger's alterations of Milgram's methodology—both the extensive screening and the 150-volt solution—limit comparisons of the two studies by precluding one of Milgram's most striking findings, the intensity of the stress participants experienced as they obeyed even to administer shocks of apparently dangerous voltage:
It seems likely that if Burger's (2009) participants had, in fact, been taken past the 150-volt level, the indication of stress and emotion reported by Milgram would have appeared. But one cannot be certain. Here, Burger's use of intense screening comes into play. Because of that screening, participants who would have shown more emotion or stress could have been precluded from being in the study. So, the "half-full, half-empty glass" metaphor seems relevant. Yes, we have a procedure that seems to fulfill some aspects of "replicating Milgram," but at a rather serious cost.
Methodology can be the primary vehicle through which the ethical tension between competing values is negotiated. "Better" methodologies from this perspective are those that accommodate both goals of advancing science and protecting subjects' welfare. Burger's innovation, while not without useful critiques, is to have found a methodology that suggests a better accommodation between competing values.
A third theme to emerge is that of IRB approval. Alan Elm's commentary puts the question most starkly: "So how did Jerry Burger (2009) get his research approved by the IRB at a respectable American university?" Much attention has been focused on the shortcomings of IRBs in providing reasonable and informed review of social science studies. That IRB approval receives so much attention in these articles is an indication of the level of this concern among researchers, a concern sufficiently great that APA has this year created a new committee under its Board of Scientific Affairs whose primary focus will be IRBs and research-related regulations. Arthur Miller offers an optimistic possibility in his commentary: "it is conceivable that Burger's study could now be cited—for example, for IRB review committees—as empirical evidence for the ethical propriety of his modified procedure."
I have addressed only a few of the many thought-provoking contributions of Burger's study and the five commentaries. I hope psychologists will discuss Burger's methods and results from many perspectives, especially that of ethics. And I will be very interested to see whether Miller's suggestion, that Burger's study may affect how IRBs conduct their work, ultimately proves accurate.
Stephen Behnke, PhD, JD, directs APA's Ethics Office.
"Ethics Rounds" columns have been converted into CE programs. Visit www.apa.org/ce.