How Fast Do We Recognize Familiar Music?

16 Aug

Cognitive neuroscience has given insight at the underlying processes that help us recognize and react to certain auditory stimuli.  With the help of ERPs (event-related potentials), psychologists are able to determine when these processes instantiate.  Several prominent examples include when spoken words can be recognized (after 240 msec from its presentation) and when famous voices can be identified (after 250 msec from its presentation).  Both examples relate to linguistic material but are similar responses seen with non-linguistic auditory stimuli?

Suzanne Filipic, Barbara Tillmann, and Emmanuel Bigand set out to discover how fast similar cognitive processes execute when listening to music.  The experiment consisted of two sections: the first investigated the recognition of a familiar song and the second investigated the emotional response time to a given auditory stimulus.  Students, mainly from non-musical backgrounds, were recruited to listen to short, musical examples that were presented in 250, 500, 1,000, 2,000, 4,000 msec increments.  The students then judged these brief excerpts as “familiar” or “emotionally touching” using a 10-point scale.

In the first section, the students listened to the musical examples and judged the “familiarity” of each excerpt.   The results showed that participants were able to distinguish the familiarity of the musical examples starting with the excerpts that were 500 ms long. In the second section, the students were given a set of musical examples that were hand-selected by musicologists based on their emotive qualities.  High-emotional samples and neutral samples (the neutral samples were derived from the “least emotional” pieces ) were used.  Again, students were instructed to rate each example on a 10 point scale on how “moving” the piece of music was.   Low-dynamic excerpts, samples that held “less energy” in comparison to the high-dynamic excerpts, were recognized as “moving” with just 250 ms of audio playback (such results are a bit puzzling, as low-dynamic excerpts contain less musical information than high-dynamic excerpts).

The results of both experiments show that the recognition of and capacity for an emotional response to music occurs early on after the onset of a non-linguistic auditory stimulus.  The researchers of this study note that the difference in time between these two ERP responses, suggests independence amongst two separate processes that work during our recognition and emotional evaluation of music.  Furthermore, the results from this study contribute to a biological, and utilitarian model of music acquisition, indicating that perhaps all of our music-tendencies are not cultural by-products but might have been derived for adaptive and functional purposes.

Filipic, S., Tillman, B., & Bigand, E. (2010). Judging familiarity and emotion from very brief musical excerpts. Psychonomic Bullentin & Review, 17(3), 335-341.


3 Responses to “How Fast Do We Recognize Familiar Music?”

  1. MinusTheLinus August 16, 2010 at 11:37 pm #

    Hello! I’ve been reading your blog for a little while, and as a fellow musician I very much enjoy it.

    I need a bit of clarification, because you’re hitting on a topic that particularly interests me. To what extent do these results show a “biological” and “functional” response to the music as opposed to cultural? I understand how this is an extension of linguistic reception, and as far as that is concerned, I’m not confused — auditory recognition is essential for survival and all that. But wouldn’t different cultures have a different interpretation of “emotionally touching” music, the way many have different ideas on rhythm and dissonance?

    • freezegelman August 17, 2010 at 11:25 am #


      A great question indeed! Usually when a dichotomy exists (in this instance the question of biological or cultural dominance over our ability to judge familiar music or induce emotional arousal) it is not whether one influence prevails over the other, but rather to what extent does each influence contribute to a musical experience. As you mentioned, much research has demonstrated a cultural impact on the perception of dissonance and rhythm. Additionally, you are more than correct when you claim that culture can affect feelings of familiarity and our emotion experiences with music – listening frequency seems to amplify both of these cases. However, this does not mean these topics are immune to biological predispositions. In the case of rhythm, we have found that infants can detect when a beat is abruptly dropped from a consistent, rhythmic cycle (like a looped drum beat) even before they can speak! Furthermore, several studies have shown that there are universal or biological tendencies for all music listeners. For example, one study examined how the Mafa (a group situated in the mountains of Cameroon, isolated from western-classical music) and Western-Music Listeners perceived emotions in both Mafa flute music and western-classical music. Both groups agreed when music sounded “happy” or “sad” across the two genres (Fritz, Jentschke, Gosselin, Sammler, Peretz, Turner, Friederici & Koelsch, 2009; Juslin & Vastfjall, 2008).
      What the researchers argue in the familiarity/emotion study, is because the majority of the participants were non-musicians who demonstrated no apparent musical expertise and were presented with instrumental classical music (music that is stripped of any semantic cues), the results were untouched from any influence of musical training or cultural experiences. To be fair, one might argue that despite the lack of musical expertise with these participants, their musical environment and listening habits might account for the results shown (perhaps all these participants listened to vast amounts of classical music!). Yet this study addressed how fast, and with how little perceptual information, discernment about familiarity and emotion can be made. Therefore, regardless if you are conditioned to western-classical music, Irish jigs, or euro-techno, you will experience a sense of familiarity or an emotional response with 500ms and 250ms of auditory-information respectfully. In other words, there seems to be biological limits to how fast we can process auditory stimuli, and that culture can not override such limits (for example if 50ms of our favorite song is played, we would not feel any emotional arousal or sense of familiarity). Furthermore, the authors note that repetition effects (or the listening frequency as I mentioned above) can facilitate performance with only unfamiliar melodies, not familiar melodies. Therefore, the researchers in this study would have witnessed an increase of performance with the unfamiliar melodies, yet nothing of the sort occurred. Although, the researchers did note that the density of the musical examples affected their results: as mentioned earlier in the post, the low-density musical segments were identified as “emotionally moving” faster/with shorter excerpts than the high-density segments. Perhaps it is not culture that should be taken into account, but the intra-musical properties of the selected examples (one question I came out of this study was, if low-density samples seem to be more preemptive at eliciting emotional responses, how much musical material is needed to experience any emotional arousal?).

      Finally, the researchers related their study to previous studies on visual perception (information I chose to omit to comply with the brevity of this post). Large amounts of research has already found similar fast-acting processes for facial familiarity and expression (170 and 120 msecs respectfully). These visual processes have been theorized to play a functional role for social reasons (recognizing a friend from an enemy?). Interestedly enough, the processes that identify a visual object work SLOWER than the processes that locate the object in our spatial view. This might have some evolutionary significance: we don’t care if the object that appears in our right field of visual is a dinosaur or a harmless deer, for if we wait around too long to find out what the object is we might get eaten first. Therefore receiving information about an object’s location in our spatial field before its identity is more useful. A similar argument could be made about music listening: fast-acting processes for auditory reception make no distinction between linguistic and non-linguistic stimuli, because it is more important to identify whether the stimuli is strange or familiar, than to determine if it is the French language or quiet, piano key strokes. Thus, the authors argue that the fast-acting processes for music might be derived from the same mechanisms that interact with language.

      Thanks for your comment!

      – Parker

  2. MinusTheLinus August 17, 2010 at 7:47 pm #

    Wow, you definitely nailed that question. This is nearly the same argument against most people who casually claim to be “tone deaf”; recognizing a familiar voice over the phone in their auditory cortices is a similar function to the recognition of pitches and timbres… I know there are exceptions of course (you’ve read “Musicophilia,” I’ll bet?). But I’m no expert. Thanks for your response!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: