You must turn off your ad blocker to use Psych Web; however, we are taking pains to keep advertising minimal and unobtrusive (one ad at the top of each page) so interference to your reading should be minimal.

If you need instructions for turning off common ad-blocking programs, click here.

If you already know how to turn off your ad blocker, just hit the refresh icon or F5 after you do it, to see the page.

Psi man mascot

Subliminal Perception

Perception without awareness is not the same thing as "subliminal perception." Subliminal perception occurs when a stimulus is too weak to be perceived yet a person is influenced by it.

As discussed in the section on psychophysics, the word limen was used in the 19th Century to refer to the absolute threshold. That was defined as the point at which a stimulus could be detected 50% of the time.

By that definition, a stimulus detected 49% of the time would be subliminal. Obviously this is not what people mean when they discuss subliminal perception.

By subliminal most people mean a stimulus is too weak or distorted to be detected through conscious effort. A better word for that is undetectable.

What did "subliminal" mean in psychophysics? What do most people mean by using the term?

Undetectable stimuli can be created by flashing a visual stimulus then quickly masking it with another stimulus. If the masking stimulus appears quickly, such as 20 msec after the first stimulus, it prevents the first stimulus from being consciously perceived.

As reported in Chapter 3 (States of Consciousness), Robert Zajonc (1980) showed that people could judge a stimulus as pleasant or unpleasant even if it was too brief to see because it was masked after 20 msec.

That was a surprising effect, so it led to all sorts of related research. One of Zajonc's students, John Bargh, became the most prominent researcher on controversial social priming effects. More about them below.

How can an undetectable stimulus be created?

The effect Zajonc discovered is real and can be replicated, but some other so-called subliminal perception effects are legends or hoaxes. An example is the fictitious 1950s experiment in which the message "DRINK COCA-COLA" was supposedly flashed to an audience in a movie theater, too rapidly to be consciously perceived.

The result was said to be a dramatic increase in Coke sales. Many people heard about this study, but the person reporting it, James Vicary, said in a 1962 inter­view in Advertising Age "the original study was a fabrication intend­ed to increase customers for his failing marketing business." (Pratkanis, 1992)

What was the true story behind the famous "Drink Coca-Cola" study?

Despite the lack of evidence for genuine effects, subliminal perception became the basis for a multimillion-dollar industry. Cassette courses using "subliminal suggestion" claimed to bolster self-esteem, help people stop smoking, improve memory, and more, all using messages too weak to be consciously detected, on audio cassettes.

Knowing that Zajonc had demon­strated a genuine subliminal effect, the United States Department of Defense funded research to determine whether subliminal suggestion techniques worked. They hired a group of psychologists headed by Robert Bjork of UCLA to investigate.

The group issued a 269 page report in 1990. They concluded there was "neither theoretical foundation nor experimental evidence" for effectiveness of subliminal self-help tapes (Swets and Bjork, 1990).

What did the Bjork committee determine, in its review of research?

Another study (Greenwald, Spangen­berg, Pratkanis, and Eskenazi, 1991) found that "neither memory nor self-esteem tapes produced their claimed effects." This study was discussed in Chapter 1 in the context of placebo effects because "more than a third of subjects had the illusion of improvement."

Half the tapes were deliberately given the wrong label. When the illusion of improvement occurred, it corresponded to the label, not the contents of the tape.

This was good evidence that subliminal tapes worked through placebo effects, not through subliminal perception. No effects of undetectable ("subliminal") stimuli were found.

Another mythical form of influence is backward masking or backmasking in which messages are put into music backwards. Repeated tests showed people could not decipher messages played backwards, nor did the mes­sages have any detectable influence on people (Vokey and Read, 1985).

What is "backmasking"? Does it have any detectable effect?

There is also no evidence for the idea that hiding little demon faces in pictures of ice cubes will influence readers to buy liquor, or that sex and death symbolism in general can influence buyers of commercial products.

That thesis was discussed in a video titled, "The Ad and the Id" (1992). In the video, a sociology professor (Dr. Bernard McGrane) gave examples of advertisements that contain hidden images of demons, sex, or death symbols.

Some of the examples were dubious or nearly invisible, but some were clearly intentional. This indicates a few advertisers still believed claims found in Vance Packard's 1957 book, The Hidden Persuaders.

Packard proposed that hidden symbols of sex and death, not consciously perceived, could be found in magazine advertisements. They supposedly aroused deep needs in consumers, leading them to buy products.

Packard argued that advertisers were using hidden symbolism to activate the id, the animal part of the psyche (according to Freud). Packard called this "subliminal mind control," based on arousing "eight hidden motivations we all have."

Although Freudian ideas have faded from currency in exper­imental sychology, McGrane showed enough examples of hidden sex and death symbols in advertising to suggest Vance Packard's ideas were still alive and well 35 years after publication. Packard's book came out in a 50 year anniversary edition in 2007, and it sold well again.

The notion that sinister forces can manipulate us with uncon­sciously-perceived symbols receives no experimental support. However, it is apparently an entertaining idea, so it is passed on to new generations.

A few advertisers apparently believe in the efficacy of hidden sex and death symbols outlined in Hidden Persuad­ers. If so, hidden sex or death symbols will occasionally appear in advertise­ments, alert sleuths will detect them, and the belief system will continue.

The Priming Brouhaha

One type of subliminal perception is easy to replicate: the type demon­strated by Zajonc in 1980. A stimulus is flashed for a split second then quickly masked, in a split second, with another stimulus.

The stimulus in this case cannot be seen, even if the subject tries, so it is literally undetectable. Then (for example) another word can be shown and the experimental subject asked to identify it as quickly as possible.

If the priming stimulus bears a close relationship to the target word, the subject will be faster to identify the target word. The priming effect lasts about a tenth of a second (Greenwald, Draine, & Abrams, 1996).

What is an example of undetectable stimuli changing behavior?

That type of priming is a robust (easily replicated) effect. It is useful to language researchers for investigating relation­ships between word meanings.

However, in the 21st century, a different type of priming dominated the news: one that fits with the topic of the previous page: perception without awareness.

Between 2000 and 2010 (and since) there were thousands of published studies about priming where the word priming meant influencing behavior using subtle stimuli. These stimuli were perceivable but not noticed.

Why are social priming studies better classified as perception without awareness?

The classic example was an experiment by Bargh, Chen, and Burrows (1996). Participants were exposed to words about old people while doing a task that involved unscrambling sentences.

The words included "forgetful, retired, wrinkle, rigid, traditional, bitter." All were related to cultural stereotypes of old age.

Afterward, when the experiment was supposedly over, participants were videotaped walking down the hall away from the research room. The group exposed to elderly stereotype words walked more slowly, like an older person.

What was the Bargh et al. (1996) experiment?

The result was surprising, so replications were attempted. Doyen, Klien, Pichon, and Cleeremans (2012) tried to replicate the original, but they made a crucial change: experimenters were not informed about the intentions or expected outcomes of the experiment.

The replication failed. But when the student experimenters were informed about the original study, before attempt­ing another replication, the effect came back. This strongly suggested an experimenter effect at work.

Bargh responded by defending his study. He pointed out differences between the original study and the Doyen et al. replication. He also launched what some called a personal attack on a blogger who publicized the failed replication.

The blogger responded with a detailed post about the whole issue. He noted that another psychologist (Daniel Simons) called Bargh's angry response "a case study of what NOT to do when someone fails to replicate one of your findings."

As charges flew back and forth, this was officially dubbed the priming brouhaha (Srivastava, 2012). It turned into a useful, stimulating discussion of issues relevant to science and replication in general.

Social priming is discussed in detail in the Social Psychology chapter. After the "brouhaha" in 2012, several other famous and unlikely-sounding priming effects failed to be replicated, so the social priming phenomenon is not always dependable.

However, there is little reason to doubt that some priming effects could occur. All that is really claimed, in the name of social priming, is that a bit of information in the environment influences a person to think or act differently a few seconds or minutes later.

The Bargh study (and thousands of similar priming studies in social psychology) fall under the heading of perception without awareness. The priming stimuli are not undetectable. They are simply not drawn to the subjects' attention.

In fact, some priming effects disappear if people notice them. This was prob­ably Bargh's most valid criticism of the Doyen et al. replication failure.

Doyen et al. used so many "old people" words that most of their subjects noticed them. This was unlike the original Bargh et al. study, where not a single participant noticed the priming.

This difference could well explain the replication failure. Unfortunately, the unruly dispute (brouhaha) distracted attention from this key factor and its relevance to the outcomes.

What might have been the real reason the replication of the elderly stereotype study failed?

Perception without awareness includes any influence on mind or behavior by stimuli outside of attention. That includes the entire universe of events not dominating attention at the time. Once attention is given to a pattern of stimulation, the effect might be altered.

All scientists would accept the idea unattended stimuli could influence thoughts and behavior. That is not even controversial. Therefore (to me) the only argument with regard to priming is not "is it real" but rather "when and how does it occur."

An example might come from our discussion of aromatherapy. Many people reported psychological effects from aromas, but at least one researcher could not find these effects under controlled conditions. She suggested they were placebo effects.

Would the effects change if people were unaware they were being stimulated by an odor? It is possible. Several other priming effects proved to occur only if people were not informed about them and did not notice them.

Meanwhile, subliminal perception (defined as influence from a stimulus you cannot detect even if you try) is real enough, but only at tiny time scales: a tenth of a second for word priming. Nobody has demonstrated long-lasting effects of undetectable stimuli.


Bargh, J. A., Chen, M. & Burrows, L. (1996). Automaticity of social behavior: Direct effects of trait construct and stereotype activation on action. Journal of Personality and Social Psychology, 71, 230-244.

Doyen, S., Klien, O., Pichon, C., & Cleeremans, A. (2012, January 18) Behavioral Priming: It's All in the Mind, but Whose Mind? Retrieved from:­plosone/article?id=10.1371/journal.pone.0029081

Greenwald, A. G., Draine, S. C., & Abrams, R. L. (1996) Three cognitive markers of unconscoius semantic activation. Science, 273, 1699-1702.

Pratkanis, A. R. (1992). The cargo-cult science of subliminal persuasion. Skeptical Inquirer, 16, 260-272.

Srivastava, S. (2012, March 12) Some reflections on the Bargh-Doyen elderly walking priming brouhaha. The Hardest Science. [blog] Retrieved from:

Swets, J. A., & Bjork, R. A. (1990). Enhancing human performance: An evaluation of 'New Age' techniques considered by the U.S. Army. Psychological Science, 1, 85-96.

Vokey, J. R. & Read, J. D. (1985) Subliminal messages: Between the devil and the media. American Psychologist, 40, 1231-1239.

Zajonc, R. B. (1980) Feeling and thinking: Preferences need no inferences. American Psychologist, 35, 151-175.

Write to Dr. Dewey at

Don't see what you need? Psych Web has over 1,000 pages, so it may be elsewhere on the site. Do a site-specific Google search using the box below.