Home > Attention, Memory > Can’t Stop The Music: Individuals with Alzheimer’s are spared their memory for music

Can’t Stop The Music: Individuals with Alzheimer’s are spared their memory for music

I guess I jumped the gun and gave the answer away in my title, but… does the loss of memory due to diseases like Alzheimer’s affect memory for music?

Typically in research fields, losing something allows us to understand how that something works. In this case, losing memory as a result of Alzheimer’s disease provides information on what is cognitively impaired as well as what is not impaired. So does losing your memory due to dementia impair memory for music?  If evidence points to the answer no, perhaps there exists a unique memory system solely for music; but if evidence points to the answer yes, perhaps there still exists a unique memory system for music. That’s pretty confusing. I’ll make a valiant effort to explain what I mean by analyzing a psychological study. Kerer et al. (2013) seek to answer this main research question by examining explicit memory for music in individuals with and without cognitive impairment, including those with Alzheimer’s disease.


Memory is a complex system – including but certainly not limited to – neurons, electrical and chemical signals, proteins, and many different storage systems. To scratch the surface, explicit memory is basically the conscious search to recall prior experiences, events, facts, or information. To be clear, the following study measured explicit memory as opposed to implicit memory, which is the involuntary and unconscious memory system. (For example, stating what you ate at your last meal relies on your episodic memory; whereas knowing how to walk is typically an implicit memory.) The Kerer et al. study involved individuals with dementia, characterized by the deterioration of cognitive functions – memory, language, attention – which leads to impairment in the performance of daily activities. Other individuals had mild cognitive impairment (MCI), which is the transitional state between normal cognition and development of dementia. A previous blog post also describes the use of music for individuals with dementia.

The purpose of the present study was to examine music memory by testing individuals with different stages of dementia in their cognitive ability to identify familiar music excerpts. In other words, is there any part of explicit memory for music that remains intact despite having dementia?

The study included 10 individuals with Alzheimer’s disease (AD), 10 MCI individuals, and 23 healthy (control) individuals. Everyone took the MMSE (mini-mental state examination) to assess mental status, covering five areas of cognitive function: orientation, registration, attention, recall, and language (Kurlowicz & Wallace, 1975). For reference, lower MMSE scores are correlated with higher cognitive impairment. As for experiment measures, the researchers created a test with 24 highly familiar music excerpts. Before this experiment, the researchers created a pilot study to make sure the songs were indeed familiar; the chosen songs ended up mostly being childhood and folk tunes. The reason why the researchers chose songs that were familiar was because, granted you’ve heard the song before, whether you recognize the tune depends on your episodic memory. So when most Americans without memory impairments hear a song like “the itsy bitsy spider,” the episodic memory is ignited, and they think, “oh boy, this takes me back to the good ol’ days,” or something analogous to that statement.

OK so now let’s pretend you’re taking the experiment. You walk into a room, go through initial screening process like signing consent forms and filling out other papers, then you’re instructed to put on headphones. You get randomly assigned to listen to a bunch of songs – some originally vocal, others originally instrumental, but all made into piano renditions. Also, you may hear some songs that are distorted, as in the pitches of the notes are changed to make it sound different from how you originally remember hearing the song.

After every excerpt, you’re asked to name the song. If you’re thinking, “uh oh. I feel like I’ve heard this song before, but I can’t remember the exact name,” then the researchers give the option of letting you choose the correct song name from multiple choices. If you’re still thinking, “uh oh. I’m drawing a complete blank on the name of the song,” and can’t choose between the multiple choices, then the researchers give yet another option to let you briefly write down any familiar aspects of the song… where you’ve heard it, what you associate with it, etc. These series of tasks measure your verbal explicit memory for music, as in your ability to search your brain for a particular sliver of information, and verbally spit it out.

Next you’re asked whether the original melody is altered (are the pitches different from the original tune?), and then whether you remember the original version of the song being instrumental or vocal. These last two questions measure nonverbal explicit memory for music, or your ability to discern the quality of music based on your memory for it.

Now that we’ve covered the procedure of the study, statistical tests were used to analyze the data. Data revealed that healthy individuals performed better on the naming task than did individuals with MCI and AD. Aka healthy participants would be more correct in naming the childhood song “Mary had a little lamb” than would the affected participants. Data also showed that a higher MMSE score correlated to better performance on the naming task. Surprisingly, the clinical groups performed better on the music quality tasks than did the control group! Individuals with MCI and AD were more accurate to say something along the lines of: “I remember this song being an instrumental tune,” (differentiating between quality of the song) as well as “This version of the song that I’m listening to for this experiment sounds different from how I remember it” (detecting pitch distortion). Thus, a lower MMSE score correlated with a better performance in these two tasks. To wrap this all up, the study revealed that the control group performed better in music tasks requiring verbal memory, whereas the clinical groups performed better in nonverbal discrimination tasks.

So are musical recognition and music memory spared in individuals with dementia? Yes and no. It seems that – parallel to what we’ve discussed in class with amnesiac individuals – episodic and verbal explicit memory are impaired in AD individuals. However, nonverbal explicit memory, which in this case is the ability to discern the quality of sound, remains intact. This means that we have a specialized memory system for music independent of verbal and visual memory, but, going back to the “pretty confusing” bit earlier in the post, not all music memory is equal. The present study shows that we may have separate memory systems for music within explicit memory: nonverbal and verbal systems, or one distinct mechanism for identification and another for discrimination of music. Hopefully not so confusing anymore.

Going off on a related, but undeniable tangent, I’m currently taking a neurobiology course. A couple of weeks ago we watched a video of a man diagnosed with both anterograde and retrograde amnesia. The gentleman could not remember anything after about 10 seconds, indicating an intact short-term memory but the loss of long-term memory. Additionally, he could not remember anything from his past except for recognizing his wife. Despite the near complete loss of memory, he remained a proficient pianist (music was his profession before his amnesia). This unscathed memory for music is fascinating because it provides further evidence that music has a unique storage system in our brains; or at least that the sensorimotor and nonverbal aspects of the brain that are linked to music remains intact.

Back to the current study. As for why the clinical groups actually performed better and not just equally as well as the healthy subjects for the nonverbal musical memory tasks, attention is a possible explanation. Perhaps the healthy subjects directed more of their attention on naming the music excerpt, leaving less attention available to detect distortion of the pitches and judgment of timbre. This makes sense, as we learned in class that the capacity of attention is limited, especially when attending to multiple tasks that require the same sensory system. In this case, participants were asked to perform two tasks, identifying and discriminating, involving the auditory system. Furthermore, since the clinical groups did not have the information available to identify the music, they focused their full attention on the differentiating task, leading to better performance.

There are countless opportunities to utilize music in medical institutions and research knowing that memory for music can be spared even in individuals with Alzheimer’s disease. Some ideas include: using music to relax patients struggling with anxiety or trauma, music to probe memory systems and facilitate remembering, and rehabilitation of motor skills using music.  This website offers more information on music therapy and neurology.

This is a neat blog post on the use of music after post-op recovery.

Here’s a link to the original study.



Kerer, M., Marksteiner, J., Hinterhuber, H., Mazzola, G., Kemmler, G., Bliem, H. R., & Weiss, E. M. (2013). Explicit (semantic) memory for music in patients with mild cognitive impairment and early-stage Alzheimer’s Disease. Experimental Aging Research, 39, 536-564.

Kurlowicz, L. & Wallace, M. (1975). Mini-Mental State: A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12 189-198.

Categories: Attention, Memory Tags: ,
  1. December 10th, 2015 at 11:34 | #1

    I really enjoyed this interesting article! I was struck by how the level of processing differed between the MCI and AD participants and the controls. The MCI and AD participants processed the music more shallowly, focusing on the surface features like the phonology. The controls, on the other hand, processed the music more deeply on a semantic level. In class, we learned that the deeper one processes the input, the better quality the memory, especially if the test asks questions regarding the deeper meaning. In the test phase of this experiment, there were questions that got at the surface level and the meaning/semantic level, and both groups were able to better responds to the questions that most closely aligned to their level of processing.

    I was also wondering how muscle memory/automatic processes played into the amnesic man continuing to play the piano. Such lifelong dedication to his craft has shifted the process from a controlled, effortful process to a very automatic process. I play a little piano (but cannot read sheet music fluently) and one thing that has always struck me as interesting is that once that I learn a piece by heart, I can play it very well with little effort because playing the song had become an automatic process. In fact, if I try to think about what notes I’m playing or try to start in the middle, I often get confused and make mistakes. This phenomenon was discussed in another blog post (Emily Moslener, “The Mind of a Champion). In that post, Emily talks about how athletes sometimes choke because they direct their attention towards every step in a process that has already become automatized (conscious processing theory). Maybe this proficient pianist is purely relying on an automatic process and procedural memory, and as such, couldn’t learn any new pieces of music.

  2. Lily Landry
    December 10th, 2015 at 19:54 | #2

    I found this post really interesting for a number of reasons. Of course, I was initially drawn to your piece because of its similarity to my own study, which also looked at music and Alzheimer’s disease. However, I thought it was interesting that your study focused more on memory for music itself, rather than the enhancement of other forms of memory using music. I think their proposal of a separate memory store for music is intriguing, but I wonder if there could be another explanation for the results they found. For instance, we talked in class today about how breakdowns in attentional selection — a feature of DAT — has a negative impact on language comprehension; thus, perhaps DAT individuals have difficulty comprehending lyrics, which in turn affects their ability to assign meaning and ultimately encode (and later retrieve) the verbal information, resulting in deficits in their verbal memory.

You must be logged in to post a comment.