More on “Behavior Priming” and Unconscious Influences

By Neuroskeptic | August 16, 2017 10:00 am

Last year, psychologists B. Keith Payne and colleagues breathed new life into the debate over ‘social priming’ with a paper called Replicable effects of primes on human behavior.

Behavioral or social priming – the idea that subtle cues can exert large, unconscious influences on our behaviour – was a major topic of research for many years, but it’s since been largely discredited. The field’s reputation suffered when Diederik Stapel, a leader in the field, was exposed as a fraud. Many researchers have since failed to replicate well-known social priming effects.

So it surprised many when Payne et al. claimed to have a found a genuine, replicable example of behavior priming. They showed that flashing up words like “bet” and “gamble” made people more likely to bet in a computerized gambling task.


In my post about the paper, I praised Payne et al.’s work, but I questioned whether the gambling task was really comparable to the kind of ‘social priming’ as practiced by Staple and others. Payne et al.’s claims just seemed more believable than, say, “professor priming” in which volunteers asked to describe a typical professor supposedly did better on a quiz later.

Now, prominent psychologist David Shanks offers his thoughts on the Payne et al. paper: Misunderstanding the Behavior Priming Controversy.

Like me, Shanks doesn’t think Payne et al.’s priming can be compared to the controversial kind of priming studies. For one thing, there’s the difference in timescale: Payne et al.’s priming effect took place over a fraction of a second, but many social priming claims involve minutes or hours. Shanks also says that many social priming effects just seem psychologically implausible:

To activate the stereotype ‘professor’ and observe a downstream effect on an individual’s ability to answer the question “What is the total number of spots on a die?” would require a rich pathway of interconnecting concepts and processes which is orders of magnitude beyond what is typically revealed in more conventional (e.g., repetition) priming experiments, where priming is commonly found to transfer remarkably narrowly across tasks and time.

Shanks questions other aspects of Payne et al.’s argument, in particular the idea that the priming effect operated unconsciously. The idea that primes can influence us without our being aware of it is central to many theories that seek to explain social priming. Shanks is a well-known critic of “unconscious processing” claims. He concludes that

Perhaps the mind is not an iceberg much of whose modular structure is below the waterline of awareness, and perhaps the unconscious can not in reality perform all of the high-level functions that the conscious mind can.

CATEGORIZED UNDER: blogging, methods, papers, select, Top Posts
  • Uncle Al

    Youtube v=_zHN4vCfwh4 “Just like you and me.” Sanctuary!

  • Jenny H

    There is certainly ‘unconscious behaviour priming’. THAT is how advertising works.
    The more people think they are unaware of advertising generally the more effect it has on them

    • OWilson

      Spot on!

      A Rose by any other name, is still, Brainwashing!

    • bing181

      I don’t know that I’d be making any sweeping generalisations about how advertising works, it’s just not that homogenous. But if I was looking for something like this that plays a key role in advertising, it wouldn’t be priming, it would be availability (as in the Availability Heuristic).

  • David Littleboy

    “perhaps the unconscious can not in reality perform all of the high-level functions that the conscious mind can.”
    This is, of course, spot on. But I think that the “unconscious” is probably better than he thinks. There are probably a lot of Minskyian “frame” or Schankian “script” sorts of things that can be done really fast by the unconscious, and then checked to see if the results make sense by the conscious. (E.g. you don’t have to think when you are handed a menu, unless the menu is somehow batty.) But those sorts of things aren’t testable with a “priming” based model. Of course, the “scruffy” AI types didn’t figure out anything better in terms of a methodology for testing how the frame, script, or even declaratory knowledge we clearly have is used. Sigh.

  • David Littleboy

    By the way, you are exactly right about this work being “more believable”. The linguistic priming research is solid. There’s no question that seeing the word bug makes it faster to process stuff about insects, spying devices, and software glitches that follows than it is to process unrelated stuff that follows. So if you already like to gamble, being reminded that you like to gamble clearly is going to get you thinking about that behavior. Which is all consistent with the recent subliminal work showing that if you flash “Coke” at a thirsty person, they’ll be more likely to buy a Coke, but doesn’t affect folks who aren’t thirsty.
    How interesting this work is, though, is another question. (As an ex-scruffy AI person, it doesn’t seem to have a whole lot of anything to do with how we think and solve problems consciously.)

  • Pingback: Priming or non priming – What is behavioral? A blog of recent updates to behavioral economics()



No brain. No gain.

About Neuroskeptic

Neuroskeptic is a British neuroscientist who takes a skeptical look at his own field, and beyond. His blog offers a look at the latest developments in neuroscience, psychiatry and psychology through a critical lens.


See More

@Neuro_Skeptic on Twitter


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Collapse bottom bar