Has it ever seemed to you that less competent people rate their competence higher than it actually is, while more competent people humbly rate theirs lower?

It’s not just your imagination. This is a genuine cognitive bias called the Dunning-Kruger Effect.

The Dunning-Kruger experiments behind the research focused on cognitive tasks (logic, grammar, and evaluating humor), but similar disparities exist in other areas. In self-assessment of IQ, below-average people overestimated their score and those above average underestimated.

Studies of healthy and unhealthy behaviors are handicapped when they rely on self-reporting because test subjects tend to improve their evaluation. In self-evaluations of driving ability, job performance, and even immunity to bias, we tend to polish our image.

This is called the Lake Wobegone Effect, named after the town where “all the children are above average.”

Notice that there are two different categories of error:

(1) the error where there is a preferred answer and most people are biased toward giving that answer (“How much snack food do you eat?” or “How popular would you say you are?” or “How good a driver are you?”), and

(2) the error where bias changes depending on actual competence, with the less and more competent groups rating themselves too high and too low, respectively.

Let’s look at the second category, where the two extremes make opposite errors. The Dunning-Kruger research hypothesizes that the competent overestimate others’ skill levels. But the error is more complicated for the incompetent—they overestimate their own skill level and they lack the metacognition to realize their error. In other words, they were too incompetent to recognize their own incompetence. Improving their metacognitive skills drove down their self-assessment scores as they became better evaluators of their own limitations.

The original paper was titled, “Unskilled and Unaware of It,” for which the authors won an Ig Nobel Prize in 2000.

The trouble with the world
is that the stupid are cocksure
and the intelligent are full of doubt.
— Bertrand Russell

(This is an update of a post that originally appeared 8/25/12.)

Photo credit:  Robert Orr, flickr, CC

I was a fan of the “Skeptics Guide to the Universe” podcast in the summer of 2007 when Perry, one of the hosts, died at about age 44. I was listening to a memorial retrospective podcast of Perry’s contributions as I walked through Seattle’s streets, on my way to an Alpha group meeting. (Alpha is a series of classes that discuss various aspects of Christianity, so God is in this story somewhere.)

One of the quirks of the show was an ongoing argument over whether birds or monkeys were more impressive animals (don’t ask). Perry would often spar with another regular on the show, enthusiastically arguing for the monkeys.

So there I am, listening to this poignant reflection as I walk to my Alpha meeting, and I turn a corner onto Queen Anne Avenue in Seattle. I see a sign that says “Monkey Love Rubber Stamps” (above).

This sign didn’t have just any name, but a name with an animal. Not just any animal, but a monkey (not an ape or a chimpanzee). Not any common monkey idiom (monkeyshines, monkey business, monkey’s uncle) but monkey love. What more is necessary to indicate a celestial blessing on the memory of a departed friend?

Apparently a lot more than that, because it’s been a memorable coincidence to me and nothing more.

Fortune telling through pie

Here’s another small but surprising event that stuck in my mind. On my daughter’s wedding day, I was making two pies that called for six eggs. The last egg had a double yolk.

This was my daughter’s wedding day. Clearly that had to mean something. Jesus was clearly telling me that they would have a happy marriage. Or that they would get married. Or that they would have kids. Or, because I beat the eggs for the pie, that they would be mangled in a horrible accident. Or something.

Other coincidences

This curiosity of the double egg yolk is like the New York state lottery picking the digits 9 1 1 on the one-year anniversary of the 9/11 attacks. Sure, that’s cool and even memorable, but what does it mean?

Noteworthy and even startling coincidences are easy to find.

• The Apollo 13 mission to the moon nearly ended in disaster, but some clever extemporaneous engineering saved the crew. Look for thirteens in this story, and you find them in abundance. Not only was the mission #13, but the time of launch was 13:13. The disaster happened on April 13. 1970 was the 13th year of the space program.
• In the weeks leading up to the June, 1944 D-Day landings, the code words Utah, Omaha, Overlord, Mulberry, and Neptune appeared in different crossword puzzles in London’s Daily Telegraph. No, not espionage—just coincidences.
• In 2001, an English girl released a helium balloon with her contact information. It landed 140 miles away and wound up in the hands of another girl. Not only did the girls each have as pets a gray rabbit, a guinea pig, and a black Lab, they were the same age and each girl was named Laura Buxton.
• The coincidences between the Kennedy and Lincoln assassinations are famous: Lincoln’s secretary was named Kennedy and Kennedy’s was named Lincoln, both presidents were succeeded by Southerners named Johnson, both assassins were known by three names, Booth shot in a theater and ran to a warehouse while Oswald shot from a warehouse and ran to a theater, and others.
• Everyone knows about the 1912 sinking of the Titanic. Fewer know about a novel written 14 years earlier about another “unsinkable” ship, the largest ever, that struck an iceberg on its maiden voyage and sank with great loss of life. The ship’s name was Titan.

If God is simply that which is unexplained or curious, then sure, that God exists. But God is then no more supernatural than an interesting pattern on a grilled cheese sandwich. Without good reason to think otherwise, a coincidence is just a coincidence.

Self-validating miracle claims

Survivors of a disaster—tsunami, plane crash, whatever—can look at the long odds for their surviving and read that as evidence of God’s providence. The problem with this analysis is that all the naysayers—those who could puncture that bubble with their own stories of how God didn’t care enough to save them—are all dead. The result is a monoculture of survivors who could look imagine God’s action.

Littlewood’s Law

Littlewood’s Law says that a “miracle”—a once-in-a-million event—happens once a month. To make this calculation, he assumed one “event” per second. Obviously, most of those are mundane. These monthly “miracles” are the surprising (but not supernatural) events that we tell our friends, like the two that happened to me, above.

We’re pattern-seeking animals. We see a man in the moon, shapes in clouds, and the face of Jesus or Mary in a tortilla. Someone determined to see supernatural agency in life can find examples, but the evidence doesn’t back them up.

Whatever Nature has in store for mankind,
unpleasant as it may be, men must accept,
for ignorance is never better than knowledge.
— Enrico Fermi

Photo credit: yelp

In 1917, two girls spent much of their summer playing by a stream. Repeatedly scolded for returning home wet and muddy, they said that they were playing with fairies. To prove it, they borrowed a camera and returned claiming that they had proof. That photo is shown here.

A total of five photos were taken over several years. The fairies were called the Cottingley fairies after Cottingley, England, the town where the girls lived.

A relative showed two of the photos at a 1919 public meeting of the Theosophical Society, a spiritualist organization. From there, Sir Arthur Conan Doyle, a devotee of spiritualism, took the baton. He wrote a 1920 article in The Strand Magazine that made the photos famous. (The Strand was also where Conan Doyle first published his Sherlock Holmes stories.) To his credit, Conan Doyle asked experts to evaluate the authenticity of the photos. The opinions were mixed, but he decided to support the story anyway.

Spiritualism, the popular belief that we can communicate with the spirits of the dead, was waning at the time of the article. Magician Harry Houdini, annoyed by fakers using tricks to defraud the gullible, devoted much time to debunking psychics and mediums in the 1920s until his death in 1926.

Houdini and Conan Doyle had been friends, but the friendship failed with their opposite views on spiritualism. Conan Doyle believed that Houdini himself had supernatural powers and was using them to suppress the powers of the psychics that he debunked. (I’ve written more about Conan Doyle’s susceptibility to magical thinking here.)

Research in 1983 exposed details of the Cottingley hoax, and the two principals finally admitted that they had faked the fairies using cardboard cutouts of drawings copied from a book.

The Amazing Randi 

Magician James Randi masterminded Project Alpha, a scheme to plant two fake psychics (Steve Shaw and Michael Edwards, actually talented amateur magicians) in the McDonnell Laboratory for Psychical Research in 1979. Randi contacted the researchers before planting his fakes to caution them how to avoid being deceived. The advice was thorough and genuine, and if they’d followed it, they would have uncovered the trickery.

Two years later, after the lab’s impressive successes were well known within the psychic community and the fake psychics were celebrities, the deception was made public. The press was so bad that the McDonnell laboratory shut down.

The moral of the story is that unless you’re a magician, don’t pretend that you can expose a magician. Said another way, just because you’re smart (and let’s assume both that the researchers were smart and most skeptics are smart), don’t think that you can’t be duped. This was Conan Doyle’s failing.

Magician Ricky Jay said, “The ideal audience [for a magician] would be Nobel Prize winners. … They often have an ego with them that says, ‘I am really smart so I can’t be fooled.’ No one is easier to fool.”

If you believe in the existence of fairies at the bottom of the garden,
you are deemed fit for the [loony] bin.
If you believe in parthenogenesis, ascension, transubstantiation and all the rest of it,
you are deemed fit to govern the country.
— Jonathan Meades

(This is an update of a post that originally appeared 6/2/12.)

Photo credit:Wikimedia

A chapter in Carl Sagan’s The Demon-Haunted World is titled “The Dragon in My Garage.” In the spirit of Sagan’s story, here is an imagined exchange between you and me about my unicorn.

Me: I have a unicorn in my garage!

You: Wow—let’s see!

Me: You don’t want to just take my word for it?

You: Of course not—I want to see.

(I open the garage door.)

Me: Okay, here you go.

You: Uh … this garage is empty.

Me: No … uh, the unicorn is invisible. They can do that, you know.

You: Okay … can you make him make a sound?

Me: No. He’s silent, too.

You: Can we see food vanish as he eats it?

Me: Of course not—he’s magic. He doesn’t need food.

(You wander through the garage with your hands out in front.)

Me: What are you doing?

You: Trying to feel for it.

Me: Uh … no—he’s really small and he scampers away.

You: Can you hear him running? Like the sound of hooves on concrete?

Me: No—I told you he’s silent.

You: Well, how about spreading flour on the floor so we can see the footprints.

Me: Nope. He can float. And I’m sure he would, because he doesn’t like to be detected.

You: Can we can catch him with a net and weigh him? Can we put a sheet over him so I can see him moving underneath? Could we spray paint and see it on his body?

Me: No—he’s … he’s noncorporeal. Yeah, that’s it. Noncorporeal.

Of course, by now you’ve lost interest in this “unicorn.” Still, you haven’t falsified my claim, and I win!

But no one would accept this conclusion. By slithering away from every possible test, this supernatural claim has no evidence to support it. Any unicorn that has this little impact in the world is the same as no unicorn at all. We can’t prove it’s nonexistent, but it’s functionally nonexistent.

“You haven’t been able to falsify my claim” is true but irrelevant. This is backwards reasoning. The proper conclusion is: There is no evidence to support this claim, so there’s no reason to accept this claim.

Isn’t this how Christians evaluate the miracle claims of other religions? Let’s handle those of Christianity the same way.

Jesus is Santa Claus for adults
(seen on a bumper sticker)

(This is an update of a post that originally appeared 6/17/12.)

Photo credit: Wikimedia

I wade through many Christians’ comments and blog posts in which the point boils down to something like, “I sense God’s presence; therefore, God exists.” Or, “I got that job after I prayed for it; therefore, God exists.” Or, “I just know that Grandpa is in heaven; therefore, God exists.”

These Christians imagine a situation like this:

where the arrow indicates causation. That is, God exists, and this causes my sense of God’s presence.

The argument can be expressed more formally:

1. If God existed, I would sense his presence

2. I sense God’s presence

3. Therefore, God exists.

But any argument of this form (If P then Q; Q; therefore, P) is a logical fallacy. Specifically, this is the fallacy of affirming the consequent.

That this is a fallacy is easy to see. For example:

1. If it’s raining, then I have my umbrella

2. I have my umbrella

3. Therefore, it’s raining.

The conclusion in step 3 doesn’t follow because I could have lots of other reasons for having my umbrella. Maybe it completes my outfit. Maybe I want to fly like Mary Poppins. Maybe I need it to act out a Monty Python silly walk or Gene Kelly’s “Singing in the Rain.” Maybe it’s a weapon. Maybe I always carry it, just in case.

The same is true in the “I sense God’s presence” case. The beginning of a more complete map of causes might look something like this:

where HAAD = Hyperactive Agency Detector, a brain trait that natural selection could have favored in early humans. Those who imagined agency (intelligence) behind a rustling in the bushes would run away and live, while those who thought, “Not to worry—probably just the wind” might pay for an error with their lives. A sound might be only the wind or a squirrel … or it might be a leopard. Those who survived (our ancestors) would be the ones with a hyperactive agency detector, which occasionally saw agency where there wasn’t any. For example, this HAAD might assign agency to thunder, drought, and illness.

In this diagram, two possibilities are shown that could create the Christian’s sense of God’s presence, and there might be many more.

Learning correct logical inferences and the long list logical fallacies won’t hurt anyone eager to think more rationally, but if you only learn one, this might be a good one to understand and avoid.

This crime called blasphemy
was invented by priests for the purpose of defending doctrines
not able to take care of themselves.
― Robert G. Ingersoll

Photo credit: Enno Lenze