McGurk Effect
Hear "ba" while watching lips say "ga," and you may hear "da." The McGurk Effect shows that what you see can change what you hear.
Your eyes can change what your ears hear. That's the McGurk Effect, first described by Harry McGurk and John MacDonald in 1976. When you watch someone talk, your brain blends the sound with the movement of their lips into a single perception, and if the two disagree, the result can surprise you.
Here's the classic demonstration. A video shows someone whose lips are forming "ga," but the audio track plays "ba." Faced with the mismatch, your brain often settles on "da," or some other in-between sound.
The sound coming out of the speakers really is "ba." But seeing the lips form "ga" nudges your brain toward "da" as it tries to make sight and sound agree. Close your eyes and you'll hear "ba" again.
So understanding speech isn't only about listening. What you see shapes what you hear, often without you noticing.
Nobody disputes that the McGurk illusion is real; you can try it yourself. The debate is about what it tells us, and whether researchers have leaned on it too heavily.
It's an odd window on everyday speechMcGurk stimuli pair a sound with lip movements that contradict it, usually as isolated syllables like "ba" and "ga." Real conversation almost never presents that kind of mismatch. Kristin Van Engen and colleagues (Journal of the Acoustical Society of America, 2022) argue that this makes the illusion a poor model of how we normally combine sight and sound, and call for tasks using matching words, sentences and stories instead.
Susceptibility doesn't predict real-world benefitMany studies treat how often someone "falls for" the illusion as a score for how well they integrate what they see and hear. But in a 2017 study of 39 young adults, Van Engen, Xie and Chandrasekaran found that McGurk susceptibility, which ranged from 0% to 100%, did not predict how much people were helped by seeing a speaker's face when following sentences in noise. If the score doesn't track the everyday skill, it's a questionable measure of it.
Huge individual differencesIn the same 2017 study, some people experienced the illusion on every trial and others never did, and researchers have documented considerable variability in the effect more broadly. The classic "everyone hears da" demonstration overstates how uniform the effect is.
Language and culture matterWork by Kaoru Sekiyama and Tohkura (1993) found that Japanese listeners rely on lip movements much less than American listeners, and native Japanese speakers show a markedly weaker McGurk effect than native English speakers. So the strength of the illusion is partly learned rather than a fixed feature of human hearing.
The illusion remains a vivid, reliable classroom demonstration that vision shapes hearing. The pushback is narrower: it's a striking special case, and researchers increasingly caution against treating it as a general-purpose yardstick for audiovisual speech perception.
Resources
Play a recording of someone saying "ba," dub it over a video of lips saying "ga," and many listeners will confidently report hearing "da." That is the McGurk Effect, a multisensory illusion that exposes how tightly sound and sight are woven together in speech perception. Harry McGurk and John MacDonald described it in 1976 under the memorable title "Hearing lips and seeing voices": when what you see and what you hear conflict, the brain can settle on a speech sound that was in neither.
The effect sits at the crossroads of several research areas:
Multisensory IntegrationThe McGurk Effect is a textbook case of the brain combining information from more than one sense (Stein & Meredith, 1993). That integration underpins many cognitive processes and is how we build a coherent picture of the world.
Audiovisual Speech PerceptionThe effect shows that understanding speech depends on visual cues as well as auditory ones (Rosenblum, 2008). Research in this area has found that the brain draws on both sound and sight to interpret spoken language accurately.
Cross-modal PerceptionHere information in one sense (vision) changes what we perceive in another (hearing) (Spence, 2011), a vivid demonstration of how interconnected our sensory systems are.
Top-down ProcessingThe effect can also be read as top-down processing, in which prior knowledge and expectations shape perception (Frisby & Stone, 2010). Seeing the speaker's lips sets up an expectation that bends the sound we hear, much as the brain's assumptions about light and shadow fool our eyes in the checker shadow illusion.
All four threads meet in this one illusion, which is why researchers keep returning to it. It offers a clean, repeatable window into how the senses interact and into the neural machinery that turns raw input into perception.
References
- Frisby, J. P., & Stone, J. V. (2010). Seeing: The Computational Approach to Biological Vision (2nd ed.). MIT Press.
- McGurk, H., & MacDonald, J. (1976). Hearing lips and seeing voices. Nature, 264(5588), 746-748.
- Rosenblum, L. D. (2008). Speech perception as a multimodal phenomenon. Current Directions in Psychological Science, 17(6), 405-409.
- Spence, C. (2011). Crossmodal correspondences: A tutorial review. Attention, Perception, & Psychophysics, 73(4), 971-995.
- Stein, B. E., & Meredith, M. A. (1993). The Merging of the Senses. MIT Press.