The Experimental Design of Paranormal Claims

Topics covered
Popular Clips
Episode Highlights
Evaluation Process
The Independent Investigations Group (IIG) employs a structured process to evaluate paranormal claims. Applicants begin by submitting an online form, which is then assigned to a member who refines the testing methodology through email exchanges 1. notes that many applicants fail to provide a clear explanation of their alleged powers, often dropping out during preliminary discussions 2. He explains, "They thought their power was working and it just didn't work" 2.
  Â
Fair Testing
Designing fair tests for paranormal claims is a meticulous process at IIG. The group uses rigorous statistical standards, requiring a probability of success by chance to be below one in 5000, much stricter than the typical scientific threshold of one in 20 1. highlights the importance of empirical validation, stating, "If we're to take anyone seriously, we have to compare what they can do to how often someone could do the same thing purely by chance" 3. This approach ensures that any claim of paranormal ability is thoroughly vetted.
  Â
Testing Challenges
Conducting paranormal tests presents unique challenges, including potential cheating and logistical issues. describes a case where an applicant claimed to discern if a person in a photo was alive or dead, requiring careful normalization of images to prevent bias 4. Despite these efforts, no applicant has passed even preliminary demonstrations, highlighting the difficulty of proving such claims 5. "We don't really have any experience with this because, as I said, nobody's ever passed a preliminary demonstration," Jerry remarks 4.
Related Episodes

Artificial Intelligence, a Podcast Approach
Answers 383 questions

Bayesian A/B Testing
Answers 383 questions
Game Theory
Answers 383 questions

AI Decision-Making
Answers 383 questions
Introduction
Answers 383 questions

AI in Industry
Answers 383 questions
Artificial Intelligence and Auction Design
Answers 383 questions

Neuroscience from a Data Scientist's Perspective
Answers 383 questions

Evolutionary Computation
Answers 383 questions

Animal Decision Making
Answers 383 questions

Video Game Analytics
Answers 383 questions
[MINI] The T-Test
Answers 383 questions

Human Detection of Fake News
Answers 383 questions

Accessible Technology
Answers 383 questions
[MINI] Natural Language Processing
Answers 383 questions
