Imagine the unimaginable. Suffering from Alzheimer’s Disease (AD), the person with whom you shared most of your life has forgotten who you are, and even worse, can no longer remember their own experiences, their relationships, and how to behave appropriately in everyday situations. But although most of their long-term memory is heavily impaired, they may continue to relate astonishingly well to autobiographically relevant pieces of music.
After several years of this disease, Turner’s mother-in-law “ED” (a clergyman’s widow) and Jacobsen’s partner’s grandmother “SB” had each lost almost all ability to recognize and recall. But although barely able to identify their children, or even themselves, they could both clearly recognize familiar music. In church, ED joyfully led the congregational singing of less well-known hymns, and SB sang along with great enthusiasm to recordings of nursery rhymes that she used to sing to her children. Even more strikingly, their response to the music gave the impression that they were touched by something that was no longer accessible by looking at once-familiar faces, nor by reminders of autobiographical events, nor by involvement in often-encountered everyday situations. Music seemed to establish a secure connection and willingness to actively communicate that was hard to achieve by other means. The act of hearing and singing familiar songs perhaps induced a sense of security, confident social interaction, and simple pleasure that Alzheimer’s patients rarely experience. Their moments filled with familiar music could be their only times without great doubts and anxious questionings of their state.
This almost paradoxical ability of ED and SB, also described by Oliver Sacks in his book Musicophilia, together with conflicting scientific views on the preservation of musical memory in Alzheimer’s, gave us the initial impetus for our study.
To begin with, we needed to face the challenge that musical memory itself is not yet well understood. While the brain’s temporal lobes have often been suggested as crucial for musical memory, as well as for autobiographical memory and memory of facts, these brain areas are generally affected early and severely in the progression of Alzheimer’s Disease. So this suggestion was quite inconsistent with the frequent anecdotal observations of musical memory preservation in Alzheimer’s. Because there are few neuroimaging studies of long-term musical memory, and none that met our rigorous standards, we had to devise a novel experimental paradigm, and an appropriate analysis strategy, to locate brain areas most crucial to this human capability.
To achieve this, we played carefully controlled excerpts of well-known, recently acquired, and previously unknown musical pieces to young healthy human subjects, while recording their brain responses with ultra-high-field functional magnetic resonance imaging (fMRI). This gave us exceptionally high sensitivity and precise spatial localization. From the fMRI recording during music listening, we could determine where patterns of brain responses were distinctive for these different levels of music familiarity. Using a machine learning technique, we could identify areas crucial to musical memory that provided significant information regarding whether the musical excerpt was well-known, recently acquired, or unknown. Of particular importance were areas close to the brain’ s midline, remote from the temporal lobes, specifically the caudal anterior cingulate gyrus and the ventral pre-supplementary motor area. This was rather surprising, because these areas are not usually considered strongly relevant to any kind of memory, although some data already exist suggesting that they are involved with music familiarity judgments. On further consideration, however, we came to recognize that because most conventional music depends on sequences of expectations and their fulfillment, it is at least plausible that these regions — often associated with sequence planning, sequence evaluation, and complex motor sequences — could indeed be fundamental to the formation of musical memory.
The second part of the study investigated the fate of these musical memory areas in patients suffering from Alzheimer’s Disease. We hypothesized that such areas would be relatively spared in the progression of this disease. Using 20 Alzheimer’s subjects and 34 healthy controls, we evaluated three essential Alzheimer’s biomarkers within the musical memory regions that we had discovered in healthy young subjects, and compared these values to biomarker severity in the rest of the brain. This clearly showed that the areas we found to be crucial to musical memory are among the least affected by advanced Alzheimer’s in the entire brain.
To our knowledge, this is the first study showing that regions crucial to musical memory coding are strikingly well preserved in advanced Alzheimer’s disease. Furthermore, our results support the idea that music is remembered implicitly, akin to remembering a complex series of movements, rather than as a discrete entity or specific countable events.
We hope that this work raises many interesting new questions, and that it provides insights that can help in the management of those who suffer from Alzheimer’s Disease and other forms of dementia, where every moment of joy and peace is valuable. If music as a tool can create such states, we need to understand and use it.
Featured image: Music, Notes, Clef, Sound. CC0 Public Domain via Pixabay.