Skip to main content

Verified by Psychology Today


The Mysterious Effects of Repetition on Music Perception

New research sheds light on how the brain differentiates speech from song.

Intuitively, it may seem that singing is a completely different type of experience than speaking. Singing is melodic, rhythmic, and manifests in artistic expression; speech, in contrast, is more analytical, symbolic, and serves purposes of communication.

In fact, new research published in Science found evidence that speech and song are also processed by different regions of the brain. In their study, Philippe Albouy and colleagues compiled a database of a cappella recordings of short phrases, and edited them in two different ways. First, they removed the melodic contour of the speech segments, essentially making them monotonic but still intelligible. The second manipulation involved changing the temporal aspects of the sound waves making the words unintelligible, but keeping the melodic contour intact. They then scanned participants in an fMRI experiment while listening to the two types of sound clips. When participants heard the monotone (but intelligible) speech, the left auditory cortex lit up. When participants heard the melodic (but unintelligible) speech, the right auditory cortex was much more active.

This research suggests that the brain draws a hard distinction between speech and song, dedicating different hemispheres to process each type of sound. However, sometimes the very same sound can be heard as song in one context but as speech in a different context.

In 2008, Diana Deutsch of UC San Diego discovered that simply repeating a spoken phrase can lead to a perceptual transformation of the phrase from speech-like to song-like. In her most famous example, a looping sound clip of her saying “sometimes behave so strangely” initially sounds normal, as if it is being spoken in the context of a regular sentence. However, by the third, fourth, or fifth repetition, one starts hearing it differently, as if it is a melody, part of a song. The very same pitch contour that is initially interpreted as prosodic and contributing to the meaning of the phrase is gradually perceived more and more as musical. In Deutsch's study, participants rated the phrase as mostly speech-like after one repetition, but mostly song-like after the tenth repetition.

In addition to measuring people’s overt ratings of phrase, Deutsch also asked participants to repeat back the phrase exactly as they heard it, either after listening to the recording just once (Group 1), or after hearing it loop 10 times (Group 2). As ratings by a separate group of participants showed, the two groups differed greatly in how they reproduced the phase. Participants in Group 1 produced something that sounded like speech, with some participants matching the original pitch of the recording more than others. However, participants in Group 2 produced something that sounded much more like song, following the melodic contour and rhythm of the original recording much more closely than participants in Group 1.

What was most intriguing is that participants in Group 2 (who had heard the phrase 10 times before repeating it) exaggerated the musicality of the original phrase; they produced systematic distortions of the original recording toward a more musical rendition; their reproductions were more rhythmic and more melodic than the original. The timing of the syllables fit better into a standard meter, and the pitch of each syllable was more stable and closer to one of the notes of the musical scale than the original recording.

It is still unknown how the brain responds to the repetition of speech. One possibility, considering the results of Deutsch et al. (2008, 2011) and Albouy et al. (2020) is that when a phrase is first heard, it is processed for meaning by the brain’s left auditory cortex. Upon a few repetitions, once there is no more meaning to extract, the brain may shift its strategy and begin processing the phrase for its musical meaning, now in the right hemisphere. Thus, the locus of brain activity may shift from left to right as the listener becomes more and more familiar with the phrase.

Future research may investigate whether such shifts of laterality as a function of repetition are fundamental to the human ability to understand, interpret, and derive pleasure from music.


Albouy, P., Benjamin, L., Morillon, B., & Zatorre, R.J. (2020). Distinct sensitivity to spectrotemporal modulation supports brain asymmetry for speech and melody. Science 28: 1043-1047.

Deutsch, D., Lapidis, R., & Henthorn, T. (2008). The speech-to-song illusion. The Journal of the Acoustical Society of America, 124(4), 2471.

Deutsch, D., Henthorn, T., & Lapidis, R. (2011). Illusory transformation from speech to song. The Journal of the Acoustical Society of America, 129(4), 2245-2252.

More from Nicolas Davidenko Ph.D.
More from Psychology Today