35

In the field of Cognitive Science, when studying about the Stroop effect, the claim is often made that it had been used during the Cold War by the American counter-intelligence units in order to identify Russian spies.

For example, this blog article How To Catch A Russian Spy quotes from the book Willpower:

the Stroop task became a tool for American intelligence officials during the cold war. A covert agent could claim not to speak Russian, but he’d take longer to answer correctly when looking at Russian words for colors.

The idea is that you can't control the effect that your native language has on your performance on this task, so it can be used to uncover your native language. Is there any historical evidence of this usage?

Oddthinking
  • 140,378
  • 46
  • 548
  • 638
Vitaly Mijiritsky
  • 1,073
  • 10
  • 14
  • 1
    Explaining the claim here, via a quote, would be helpful. – LangLаngС Jun 30 '18 at 14:23
  • 4
    Of course that would not work to find Americans who spied for the Russians. And if someone admits he comes from Russia but claims not to be a spy, that test cannot tell if he is a spy, although it might possibly indicate he came from Russia at a different time than he says. – M. A. Golding Jun 30 '18 at 19:02
  • 1
    The claim isn't that the Strood test is universally useful in catching all spies, but that it can reveal whether someone is a native / fluent Russian speaker, which may be a useful test in the case of a suspected spy who claims not to speak Russian. So the question is whether there are any confirmed cases where the test did provide critical information in uncovering a spy. – PhillS Jun 30 '18 at 20:40
  • 8
    Re Stroop, the claim is that it can reveal whether someone is a native/fluent speaker of any language in which the test is given, it's not specific to Russian. The question is whether it was even attempted, let alone really yielded results. – Vitaly Mijiritsky Jul 01 '18 at 03:05
  • I found the claim repeated [on a PT blog](https://www.psychologytoday.com/us/blog/play-in-mind/201204/when-red-looks-blue-and-yes-means-no) but with no more specifics, alas. – Fizz Jul 01 '18 at 05:05
  • Any spy worth a nickle would always lower his baseline for any test given – Andrey Jul 06 '18 at 14:44
  • My impression, from far away in both space and time, during the McCarty era everyone was a communist spy until proven perhaps not. It would not surprise me if they used tests based on the Stroop effect at least on one supposed spy. – ghellquist Oct 04 '19 at 20:42

1 Answers1

9

As a partial answer, according to the 1987 book Systems and Theories of Psychology, volume 2, page 410:

We conclude our discussion of the Stroop effect with a story for whose veracity we cannot vouch. It seems a man was accused of being a Russian agent, a charge which he vehemently denied. He claimed to know nothing even of the Russian language. His questioners gave him the Stroop test in the Russian language, which he duly failed; that is, he said "blue" much more slowly when the word was the Russian for "yellow!" If the story is true, the Stroop test has been used to reveal something considerably more applied than the nature of unconscious cognitive processes.

So this story existed by at least 1987.

DavePhD
  • 103,432
  • 24
  • 436
  • 464
  • Not definitive but interesting. Could you please quote a larger excerpt to get the entire context? – gaspar Jul 02 '20 at 15:17
  • Chapter name would also be helpful to look this up in a different edition. – gaspar Jul 02 '20 at 15:40
  • 1
    @gaspar I expanded the quote on both ends. Not sure of chapter name, subchapter name is "The Stroop Effect" which starts on page 409. – DavePhD Jul 02 '20 at 17:15
  • 1
    I just did the stroop Test myself in English and failed miserably although I’m not even a native speaker. This doesn’t prove the test was actually used to identify Russian spies but makes it very plausible. Anyone who falsely claims not to know any Russian will have a hard time passing the test. – Hartmut Braun Jul 02 '20 at 18:16
  • 1
    "It seems" does not suggest a confident source. – Jack Aidley Jul 03 '20 at 07:59
  • 2
    @JackAidley Neither does "for whose veracity we cannot vouch." – Cristobol Polychronopolis Jul 08 '20 at 18:14