Project Core – Demarcation and Departure

Project Core – Demarcation and Departure
by guest blogger,
Tyler A. Kokjohn

Paranormal phenomena encompass a diverse array of experiences that are often considered to be entirely outside the boundaries of science.  However, the tools and techniques of scientific investigation are so powerful because they can be applied to many situations and this demarcation is not necessarily sharp.  Some cryptozoology research efforts and the search for extraterrestrial intelligence (SETI) are solidly within the scientific mainstream in terms of approach and methodology.  While the often fleeting and unpredictable nature of paranormal events complicates their study, it is possible to investigate certain aspects of them scientifically.  Project Core was created to initiate that process.

Investigations begin with a question, or as a scientist might put it more properly, a hypothesis.  The heart of the scientific method is asking questions that can be answered through direct observation or experiment.  The art of the process is ensuring that your approach and answers are unbiased and valid.  Some of the challenges are illustrated by a simple experiment I conducted in my own back yard.

The video (below) combines a series of photos taken at one minute intervals over about an hour which shows flower blooms opening up in the morning and moving to face and follow the sun.

Flowers1


How do plants respond to the environment like that?  That is a complex problem, but it is possible to reduce the scope of inquiry to an answerable question. Scientists often employ a reductionist approach to break down complicated problems into more manageable smaller pieces.  In this case the correlation between sunlight and flower opening leads to a testable hypothesis; exposure to visible light induces flowers to open.

To test my hypothesis I collected two yellow flowers in the morning before they had opened.  I placed one under a light indoors and the other inside an adjacent container (a coffee cup with a lid) which I could keep dark.  The experiment was started at 8:47 a.m. and was stopped at 9:34 a.m., when I noted almost all the flowers outside had opened up.

 

Screen Shot 2015-03-04 at 12.18.04 PM

The experimental outcomes:

Screen Shot 2015-03-04 at 12.19.14 PM

Unfortunately, it seems the original hypothesis that light exposure triggers bloom opening was not supported by the experimental evidence.  The flower blossom incubated in light opened up, but because a similar flower held in the dark (the control) also opened up, it seems light exposure had no impact on the bloom opening response.

Case closed?  Not exactly, I made a couple of mistakes.  The tricky part was trying to do my best to debunk my own investigation to discover any errors in method or logic.  To start, I reviewed the rationale for the experiment design.  Both orange and yellow flowers were available to me, but I used only yellow flowers from the same small garden area to study.  Hopefully, these actions provided experimental subjects that were as similar to each other as possible.  That means any activity differences observed during the experiment could confidently be attributed to the differences in the manipulated variables, in this case, light exposure.

The flowers I tested were cut and that meant they were physiologically different from the ones out in the yard with functioning root systems.  However, I synchronized my experiment with the bloom opening behavior of the rooted flowers.  The cut flowers, including the one in the dark, had also opened at the same time of day.  In addition, the fact that cut flower blooms opened suggests that all the necessary systems functioned like those active in the rooted plants.  If the cut flowers did not open or did so in a very different time frame it would be a lot more difficult to make a convincing case for functional physiological equivalence.  So it appears my reductionist approach using cut flowers mimicked reality closely enough to be useful.

The work began with an obvious correlation between light exposure and bloom opening.  However, even the most apparently clear-cut correlations do not necessarily reveal the true underlying causes.  To avoid falling prey to the logical fallacy ‘with this, therefore because of this,’ deceptive correlations are rooted out by subjecting them to additional confirmatory tests.   However, these tests must be done carefully and their results interpreted skeptically.

There are problems with my study because I was too lax with the self-criticism from beginning to end.  I set up and performed the experiment improperly, but failed to notice the error until after the work was completed.  Suppose I wanted to submit a manuscript describing this work and my conclusions to a scientific journal for publication.  Before it is published, it goes through a peer review in which it is evaluated by anonymous experts chosen by the journal editors.  Editors rely on peer reviewers to expose shortcomings in approach or execution.

It is not hard to imagine if I had not noticed the errors and submitted my work for publication, an alert reviewer would probably returned a polite, but devastating, critique of the work and results to the journal editor and me.

The author has failed to test his central hypothesis that light exposure induces closed flower blooms to open.  Unfortunately, the collection of flowers seems to have taken place after dawn when all the blooms had already been exposed to indirect light.  Perhaps this exposure was sufficient to catalyze petal motion.  If so, the observation that a flower incubated in the dark could open is easily explainable as nothing more than experimental artifact.  To answer the question, the investigator will need to collect flowers at night and take precautions to ensure the dark controls are incubated without any light exposure.    

Without this system of external checks, you can see how easy it would be to make and publish errors.  In this case, I did not have a true controlled experiment because of an unnoticed oversight in how I performed the work.  That small mistake was the difference between being able to reach a scientifically valid conclusion or not.             

With experience a person can get better at developing controlled experiments, but a more insidious danger, confirmation bias, sometimes confounds investigators.  Humans are natural story tellers.  Accepting the experimental results at face value it was easy to conclude that light exposure has nothing to do with bloom opening.  That idea is so seductive because it seems to fit all the observations so well.  The problem is that the way I did the work makes it impossible to reach any experimentally-validated conclusion at all about the impact of light on bloom opening.  I quit being skeptical too soon.  Believing the results of a single shoddy experiment, I then built a tidy global narrative as to how the whole process worked.  But note how I began by investigating one hypothesis (light exposure induces bloom opening) and ended up answering a different one (light exposure has nothing to do with bloom opening). These subtle shifts and circular logic traps can be extremely hard to recognize because few stories appeal to us more than the ones we create ourselves.  Poor experiment design is troublesome, but lethal confirmation bias may emerge at any point in the investigation process.  Even liberal applications of Occam’s Razor may not prevent biased investigators from extracting preconceived answers out of uncorroborated correlations and poorly executed experiments.  If a mainstream scientist is skillful or simply lucky, he or she recognizes any errors and corrects them before proceeding to publication.  If not, a peer reviewer will probably be delighted to expose them.

Creating and evaluating hypotheses by confronting them with data demands proficiency, patience and determined ruthlessness.  Hypotheses for which no properly controlled, reproducible supporting evidence can be developed must be modified, put aside or rejected.  The process is sometimes so tricky that scientists do not trust themselves to do it properly every time.  That’s why they employ peer review to check their work.

Aspects of paranormal phenomena can be reduced to experimentally tractable questions.     The Project Core survey is best viewed as a departure point to construct hypotheses to be confirmed or refuted through additional work.  Some potential lines of inquiry are already clear.  We invite you to explore the questions posed in the Project Core survey as well as develop new testable hypotheses, the ultimate peer review.

The inherent challenge of paranormal phenomena investigation suggests that many attempts to describe and comprehend them may be deemed crude and unsatisfactory.  We should anticipate needing to systematically eliminate biases, refine ideas and explore alternative approaches as new data and insightful constructive criticisms dictate.

Advertisements

2 thoughts on “Project Core – Demarcation and Departure

  1. This is a clear, nicely laid-out description of the scientific method, the basis for most of modern knowledge. When a hypothesis is examined as you describe, then we can be fairly confident that, yes, now we “know” something. There’s reason, evidence, retesting and peer review to support that the thing is true or a valid statement of what has been learned. And so human knowledge is built, one brick at a time.

    The sadness for me in observing ufology over the past two decades is that we are learning little or nothing that is confirmable; no new knowledge is being built about a phenomenon that I believe has a genuine core of experience. It would be wonderful if researchers could find a way to apply the research techniques and methods you describe in studying these mysterious human experiences. Thanks for offering us your own professional insight…along with your blooper!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s