Self-Experimentation?.....Sounds Creepy and Weird....
It definitely does. It brings up images of mad scientists consuming strange elixirs and potions. It makes one think about some crazy genius injecting himself with strange medicines. Dr. Jekyll and Mr. Hyde may come to mind.
Over the past two centuries, self-experimentation has been most-documented in medical research, with a peak of self-experimentation occurring during the first half of the 20th century.
Most of this medical self-experimentation was undertaken to examine the effects of infectious diseases, followed by investigations of anesthesiology, physiology, pharmacology, and radiology (Weisse, 2012). For example, Albert B. Sabin administered an oral polio vaccine to himself before administering it to thousands of others in a field trial (Weisse, 2012). Although reports of self-experimentation in the medical field is in decline, many self-experiments have proven invaluable to the medical field in terms of advancing science and saving lives.
Why Many Methodologists and Scientists Hate Self-Experimentation
The scientific community likes nice and neat experiments that are well-designed and thus, accurately tell the researcher whether an independent variable had some type of effect on a dependent variable. To do this, an adequate sample size that has been randomly assigned and selected must be a prerequisite in order to eliminate any rival explanations. When dealing with a sample size of n=1, any findings just simply do not hold up in terms of scientific validity. Not only is the sample size inadequate, but there is obvious self-selection bias happening here. You cannot generalize findings from a self-experiment, nor can it be replicated across different groups, settings, contexts, and times without some major quality of the study being distorted.
Perhaps the main beef against self-tracking and self-experimentation is simply the fact that self-experimenter expectations lead to systematic error in data collection and data interpretation- making any findings completely biased an invalid.
Similar to Mook (1983), who defends external invalidity, I extend the proposition to the defense of internal invalidity, hypothesizing that the benefits of self-experimentation lie not in the outcomes, but in the process. The foundation of this suggestion stems from approaching scientific inquiry with a naturalistic and constructivist paradigm. In the constructivist paradigm individuals construct their own realities and beliefs to make individual and social meaning out of what they subjectively experience (Guba & Lincoln, 1985). There is no single, objective truth, but rather each person has a unique construction that is shaped by his or her own value system. Looking at self-tracking in this light, it is perfectly acceptable to not be able to generalize our data. But we must also not view our data collection and interpretation as being useless to others. More on the prosocial effects of self-experimentation in later segments...
Self-Experimentation & Self-Tracking as a Strategy for Idea-Generation, Decision-Making
From an experimental psychology perspective, Seth Roberts (2004), believes that although there may be potential bias problems, long-term self-experimentation has an enormous strength in helping one generate and develop new plausible ideas. Rather than generating theoretical ideas post hoc using exploratory data analysis (Tukey, 1980), one can begin to gather data on personal assumptions, and through a process of advanced trial and error, uncover their potentially true underlying mechanisms (Roberts, 2004). Through twelve years of self-experimentation and self-tracking, Roberts discovered several surprising cause-effect relationships: standing eight hours a day reduced early awakening and made sleep more restorative, even though more standing was associated with less sleep; and eating a half a stick of butter helped improve the speed in which he could complete mathematical problems (Roberts, 2010).
The quantification and collection of information can also serve as a sort of memory enhancement. It also can be utilized to help one make more informed decisions (Cowley, Lindgren, & Langdon, 2006). Numbering things allows one to test, compare, and experiment. Numbers make problems less resonant emotionally, but more tractable intellectually. In science and in business, quantitative data is often the gold standard of truth.
Why not so when it comes to the individual?
Self-Experimentation & Self-Tracking as an Outlet for Curiosity
Practically, self-experimentation and self-tracking provide an outlet to the curious individual who is not satisfied with studies that describe the means or averages of groups. This individual recognizes that while these studies provide enlightening information on human behavior, they do not offer much help in how a single person should go about using this information. The individual may question, “Well, do I fall into the average, or am I an outlier? What dot would I be?”
Even though authors may suggest different actions towards implementing their findings, without self-experimentation, one must blindly trust that these broad and highly conditional suggestions will improve the quality of one’s life. In most cases, one likely will never be certain of the actual effects. This is no longer acceptable, and now due to advances in technology, individuals can take things into their own hands.
References for Further Reading if Interested:
Cowley, B. J., Lindgren, A. & Langdon, D. (2006). Using self-experimentation and single-subject methodology to promote critical thinking. Critical Thinking, 1, 26-40.
Guba, E. G., & Lincoln, Y. (1985). Fourth generation evaluation as an alternative. Educational Horizons, 139-141.
Mook, D. G. (1983). In defense of external invalidity. American Psychologist, 34(4), 379-387.
Roberts, S. (2004). Self-experimentation as a source of new ideas: Examples about sleep, mood, health, and weight. Behavioral and Brain Sciences, 27, 227-288.
Roberts, S. (2010, August 17). Arithmetic and butter. Retrieved from http://quantifiedself.com/2010/08/arithmetic-and-butter/
Tukey, J. W. (1980). We need both exploratory and confirmatory. American Statistician, 34, 23-25.
Weisse, A. B. (2012). Self-experimentation and its role in medical research. Texas Heart Institute Journal, 39(1), 51-54.