Monday, 3 July 2017

Aging and Motor Memory

In my spare time, I am slowly learning Bach’s English Suite in A minor, movement by movement. I take a break from writing or studying and I sit down at the piano, pick a small section, and work through it, trying to get it under my fingers. I always make progress throughout a practice session, and leave the piano feeling like I’ve accomplished some learning. But the next day, when I come back to the piano, and try the same section, I’m usually disappointed. The improvements from the previous day don’t seem to last in the same way they did when I was a younger musician.

Unfortunately, this is a natural effect of aging.

Memory and learning differ between young adults and older adults in a number of ways. Older adults tend to have more difficulty with declarative memory – conscious memory for facts and events – than younger adults. This is often attributed to age-related loss of neurons in the hippocampus, a structure in the brain that is the site of declarative memory formation. So if I tell you that Ulaan Baatar is the capital of Mongolia, you would store that in your declarative memory. If you’re twenty, you are more likely to remember this fact tomorrow than if you are seventy. From a psychological point of view, deficits in declarative memory seem to be related to decreased attentional resources in older people. In other words, older people can’t pay attention to as many different things at once as a younger person can, so they can’t spend as much time committing any one fact to memory. That means the capital of Mongolia won’t be stored in their memory as stably as it would in a young person.

Motor learning uses a different type of memory known as procedural memory, and has its own difficulties for older adults. Older people are able to improve at a new skill over a practice session, but they show a different pattern from young people when it comes to retaining that skill. When young research volunteers learn a new motor skill, such as a finger-tapping pattern, they improve significantly over a practice period, with a decrease in the number of errors and a gradual increase in speed. When the volunteers come back the next day for a second session, their performance often has improved overnight, without any further practice. This is due to sleep-dependent consolidation, in which our motor skills both improve and become resistant to interference from other memories, while we’re sleeping. People can literally improve their motor skills, such as playing a musical instrument, just by sleeping.

As we age, things start to change. Older adults just don’t show the same between-sessions improvement in motor skills, and in fact their performance on the second day of training on a task starts out much lower than where they left off the day before, just like when I practice Bach. When researchers compared the brainwaves of sleeping young adults with sleeping older adults, it became clear that older adults spend less time in slow-wave sleep and show a decrease in sleep spindles, a particular type of brainwave that is believed to be important in motor memory consolidation.

Research from the University of Montreal suggests that the hippocampus, even though its main role is in declarative memory, is important for sleep-dependent consolidation of motor memory. This implies that decreased hippocampal function in aging leads to problems with motor memory consolidation. The key point here is that there are important interactions between the declarative and procedural memory systems. Which means that the declines in declarative memory which happen naturally with age also affect procedural memory.

Understanding this gives us a hint at a solution to the problem of motor learning in older adults. One of the ways in which we learn motor skills is by using declarative memory to help us along the way. For example, when we’re learning a new piece of music, we consciously read the music and try to be aware of things we need to remember:  cross finger 4 over here; F# in left hand here, and so on. Using declarative memory to bolster motor learning is a poor strategy when declarative memory isn’t working so well. In order to counteract the effects of poor declarative memory on motor learning, we should choose practicing strategies that rely more on implicit, procedural memory, strategies based on repetition of the movements we want to learn rather than our cognitive appraisal of the notes and movements required to make them.

The obvious candidate for this type of learning is a technique called errorless learning.  This technique suggests that if you can simplify a task somehow so that it can be practiced without making errors (or at least as few as possible), then you engage procedural memory systems, leading to more automatic performance. For example, a 2012 study by Chauvel and colleagues tested older and younger adults in two techniques to learn golf putting skills. One group used “infrequent error” learning, where the people practiced putting into a hole from a short distance away. The other group practiced putting from a larger distance, while led to more frequent errors because the task was harder. This second group had to develop declarative strategies about how to improve. Then both groups were tested on putting from an intermediate distance, with and without distractions. The researchers wanted to see if the older adults fared better with one type of learning than the other. And the results were clear: when using the “infrequent error” approach, older and younger adults performed equally well on the test. Using the “frequent error” approach older adults performed worse than younger adults.

In music practice, errorless practice can be achieved by practicing at a slow enough tempo to avoid mistakes in pitch and rhythm. I’ve adopted this approach recently, playing everything extremely slowly and accurately and I found that it improved my retention of the pieces both within and between practice sessions. 

Of course, the idea of practicing slowly is not new or mind-shattering. I often tell my students that if they’re practicing their pieces so quickly that they’re making mistakes, then they’re actually practicing their mistakes. But since reading about errorless learning, I’ve been encouraging them more and more to practice error-prone sections using “slow-motion” practice, and this has been very helpful for them. It’s revealing to see the reasoning behind why this works: because errorless practicing is training our procedural, automatic memory.

While my sense of pride wants me to point out that I don't actually fall into the category of "older adult", I'm not as young as I used to be, and clearly that makes a difference in how my memory processes function. As in the story of the tortoise and the hare, slow and steady wins the race, especially if you're no spring chicken. 

Albouy, G., King, B.R., Maquet, P., and Doyon, J. (2013). Hippocampus and striatum: Dynamics and interaction during acquisition and sleep-related motor sequence memory consolidation. Hippocampus.

Chauvel, G., Maquestiaux, F., Didierjean, A., Joubert, S., DieudonnĂ©, B., and Verny, M. (2011). Use of nondeclarative and automatic memory processes in motor learning: how to mitigate the effects of aging. GĂ©riatrie et Psychologie Neuropsychiatrie du Viellissement 455–463.

Chauvel, G., Maquestiaux, F., Hartley, A.A., Joubert, S., Didierjean, A., and Masters, R.S.W. (2012). Age effects shrink when motor learning is predominantly supported by nondeclarative, automatic memory processes: Evidence from golf putting. The Quarterly Journal of Experimental Psychology 65, 25–38.

Craik, F.I.M., and Rose, N.S. (2012). Memory encoding and aging: A neurocognitive perspective. Neuroscience & Biobehavioral Reviews 36, 1729–1739.

Fogel, S.M., Albouy, G., Vien, C., Popovicci, R., King, B.R., Hoge, R., Jbabdi, S., Benali, H., Karni, A., Maquet, P., et al. (2014). fMRI and sleep correlates of the age-related impairment in motor memory consolidation. Hum Brain Mapp 35, 3625–3645.

King, B.R., Fogel, S.M., Albouy, G., and Doyon, J. (2013). Neural correlates of the age-related changes in motor sequence learning and motor adaptation in older adults. Front Hum Neurosci 7, 142.