This American Life: Confessions

This American Life is a radio show on NPR that explores interesting personal stories.  I listen to the podcasts of this show because it is sometimes topical, sometimes general interest, and often thought-provoking and entertaining.  An episode I heard lately turned out to have a story that would be very useful in the science classroom.

The episode titled “Confessions” aired on October 11, 2013, and it focused on (you guessed it) different types of confessions, both religious and secular.  The story from Act One (Kim Possible) centers on a police investigation of a murder.  The detective in charge has a suspect in custody, a decent amount of evidence, and a confession.  After believing that the case was pretty much closed, the detective finds new evidence that contradicts the confession.  Eventually the case is thrown out of the courts.

Years later, the case still haunts the detective.  Why did he obtain a confession if there’s no way the suspect could be guilty?  After revisiting the video, he realizes what went wrong: he had inadvertently provided the witness with enough details for her to provide a convincing confession.  Over the course of a 14 hour interrogation, he had shown the witness credit card statements, pictures of the crime scene, and other details about the case.  Exhausted, the witness told the officers what she thought they wanted to hear.  Luckily, there was evidence to support her original alibi.  Because the case went to the courts, however, the suspect lost custody of her children and had the charges on her criminal record.  The end of the story features the detective apologizing for his botched investigation and devoting his time to educating law enforcement officers about false confessions.

So how does this relate to science?  In a laboratory, you formulate a hypothesis (and possibly alternative hypotheses).  Then you design experiments to test your hypothesis.  There is a fine line between objectively testing a hypothesis and carrying out experiments to prove your hypothesis correct.  A strong hypothesis will withstand any tests to prove it wrong.  Indeed, it is a scientist’s duty to continually test every caveat and alternative.  That being said, it is difficult not to get caught up in believing your hypothesis is true, especially when grant money and publications are depending on it.

The investigator in the Confessions story fell for this same trap.  He was so dogged in his pursuit of his one suspect that he did not investigate other possibilities.  He ran himself so ragged that he let crucial details slip and obtained a false confession.

For students, this story illustrates pitfalls in the scientific method.  Here are some reflection questions to ask of students while or after listening to the story:

  • Phrase the detective’s investigation in terms of the scientific method.  What is the detective’s hypothesis?  What evidence supports his hypothesis?  What evidence contradicts it?
  • How could a researcher fall into the same trap as the detective in this story?  What would be the far-reaching effects of these mistakes on a research project?  On a graduate student’s project?  On a lab’s publication record and ability to obtain research funds?  On a scientist’s career?
  • Is it better to find data to support your hypothesis or to contradict it?
  • Can you ever prove that a hypothesis is true?  Should the same standard be used in law enforcement?  What legal policies already exist that come close (students could reply with “innocent until proven guilty”)?
Advertisements
This entry was posted in Podcast and tagged , , , , . Bookmark the permalink.

Post a Comment

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