Thursday 6 January 2011

How did we know that he was going to say it before he said it?

From NPR:

In one of Professor Daryl Bem's studies, 100 college students were shown a list of 48 common nouns flashed on a computer, one at a time, for three seconds each. The instructions said: Look at the word, try to visualize it (see "tree;" imagine "tree") and then go on to the next word.

Afterward, they were told, Surprise! We're going to give you a quick memory quiz. How many of the words we just showed you can you recall? Students typed in the words they remembered.

Then a computer went through the same list of words and chose 24 — totally randomly; no human was involved. Before you leave, the students were told, we still want you to scan and then type the words the computer selected. As they typed, the students were, of course, committing those randomly selected words to memory. But who cares? The test was over.

Now comes the surprise. When Dr. Bem checked the original surprise recall test, a weird pattern emerged. He noticed the students for some reason turned out to be better at recalling the words they had scanned and retyped after the test. A second group of 24 words served as a control. The computer never asked students to retype them. Those words weren't recalled as often.

Then Bem drops his bomb: "The results show that practicing a set of words after the recall test does, in fact, reach back in time to facilitate the recall of those words."

2 comments:

Nick Drew said...

somebody has been reading Childhood's End (Arthur C Clarke)

View from the Solent said...

I knew Nick Drew would say that ....

A (mathematically heavy) lambasting of Dr Bem's article here http://www.ruudwetzels.com/articles/Wagenmakersetal_subm.pdf.

The summary finishes "We conclude that Bem’s p-values do not indicate evidence in favor
of precognition; instead, they indicate that experimental psychologists need
to change the way they conduct their experiments and analyze their data."