Home > Memory > Being able to sing along: Semantic priming and familiar songs

Being able to sing along: Semantic priming and familiar songs

Sing3Have you ever heard the saying, “If I could remember school work like I remember lyrics, I’d be a genius?” It is true that many people remember an immense number of songs throughout their lifespan. Melodies for popular songs are almost unforgettable, and learned lyrics can stay in memory for a lifetime (Bartlett and Snelus, 1980). Memory for songs is contained in two stores that have two separate functions: episodic and semantic memory. Episodic memory allows you to remember the “when,” and “where,” of things, so recalling the first time you ever heard “Hey Jude” by the Beatles would use episodic memory. Semantic memory refers to remembering the facts and vital information about something – the “what” – but not being able to specifically recall when you learned that information. Remembering the lyrics and tune to “Hey Jude,” uses semantic memory. It is not necessary for you to remember the first (or last) time you heard the song in order for you to be able to sing along.

In 2012, Johnson & Halpern studied the effects of semantic priming on the recognition of popular songs. “Priming” refers to when a stimulus effects the response to a following stimulus. A “prime” is the first stimulus introduced, and then a follow-up stimulus usually follows shortly after. Priming effects are well established; response times to stimuli are greatly increased when the prime and the target object are related. For example, in a lexical decision task, where a participant must identify the target as a “word” or “non-word,” response times are much faster when the prime and target match (e.g., prime = “table,” target = “chair”) than when the target and prime are unrelated (e.g., prime = “table,” target = “fox”). Johnson & Halpern’s study examined whether this priming effect could be extended to tune and lyric recognition.

Previous studies have examined the effects of priming within one song (e.g. Peretz et. al. 2004), but Johnson & Halpern examined the effects of priming across multiple songs. They separated songs into related categories – for example, the “Christmas” category included “Deck the Halls,” and “Jingle Bells;” the “Patriotic” category included “America the Beautiful,” and “Yankee Doodle” – and tested priming in four separate experiments: tune primes, tune targets; category primes, tune targets; category primes, lyric targets, and tune primes, lyric targets.

In the tune primes: tune targets experiment, related primes and tunes were from the same song category. The data showed much faster response times to songs (participants had to determine if the song was real or not) when the prime and tune were related. This supports the spreading activation model, a model in which one thought can activate another thought based on their similarity to one another. They then explored the priming of melodies when the category was explicitly named, in the category primes: tune targets experiment. In this second experiment, related primes stated the category that the tune was from. Interestingly, no priming effect was noticed in this experiment. This is most likely due to modality differences, displaying that lexical primes do not help when identifying an auditory stimulus.

In the third experiment, the effects of priming from lexical category to lyrics (both verbal modalities) were examined (category primes: lyric targets). Not surprisingly, an almost identical priming effect to Experiment 1 was found. This supports the conclusion stated above that priming in songs occurs within, but not between, verbal and auditory modalities. Experiment 4 questioned if a related tune prime would have an effect on response time to the lyrical target (tune primes: lyrical targets). The results of this mirrored Experiment 2, with no significant priming effect found.

Johnson and Halpern’s study is significant because it shows that memory for songs, and their components (melody and lyrics), is organized based on semantic (or conceptual lines). We unconsciously create category associations between related songs in order to able to identify them faster and remember them better. If you sing a Christmas song, it probably will make more Christmas songs pop into your head. This shows that music memory follows the spreading activation model, reducing cognitive load and making it easier to recall things when they are relevant to the context. More broadly, this experiment shows that we are perhaps more cognitively invested in music that previously thought. Melodic recognition is not simply based on familiarity with that particular melody; instead, we have an elaborate conceptual network that allows us to recognize familiar tunes. Johnson and Halpern’s study shows that remembering melodies and lyrics is as complex as any memory process. So, in reference to the quote in the beginning, the way you remember your schoolwork is very similar to the way you remember your favorite song.


Here is a link to the original article: http://0-link.springer.com.library.colby.edu/article/10.3758/s13421-011-0175-z/fulltext.html


Bartlett, J.C., and Snelus, P. (1980). Lifespan memory for popular songs. American  Journal of Psychology, 93, 551-560. doi: 10.2307/1422730

Johnson, S. K., & Halpern, A.R. (2012). Semantic priming of familiar songs. Memory and Cognition, 40, 579-593. doi: 10.3758/s13421-011-0175-z.

Peretz, I., Radeau, M., & Arguin, M. (2004). Two-way interactions between music and language: Evidence from priming recognition of tune and lyrics in familiar songs.  Memory and Cognition, 32, 142-152. doi: 10.3758/BF03195827

Categories: Memory Tags: ,
  1. May 2nd, 2013 at 12:02 | #1

    Really interesting post, Molly! I find it fascinating how the spreading activation model can be applied to our memories of music and melodies, in addition to other forms of semantic information. It’s pretty amazing how we are able to make unconscious connections between songs and song lyrics in order to retrieve them faster and increase our memory of them. Unfortunately, I still feel like I can remember songs and lyrics better than some of my school work! Maybe I should spend more time practicing retrieval processes for my school work instead of for my favorite songs…

  2. May 9th, 2013 at 17:26 | #2

    I wonder if the melody and rhythm of a song also helps you remember it. A lot of songs also rhyme, so you know what the next line needs to end with. When I have to memorize lines for a play, memorizing Shakespeare (in verse) is much easier than just memorizing lines that aren’t in a pattern and a rhythm. I feel like this also has to play in somehow.

  3. May 14th, 2013 at 14:22 | #3

    Hi Anna! As I mentioned in the beginning of my article, melodies play a huge part in remembering a song! It is actually basically impossible to forget a familiar melody once you learn it (the way they test this is based on recognition of the melody, NOT being able to recall it without any cues). I liked your point about rhythm and rhyme, too. I didn’t encounter anything about rhyme when I was researching this article, but I bet that your idea is right!

You must be logged in to post a comment.