Home > Aging, Memory > Aging and Metamemory

Aging and Metamemory

Everyone has to get old, and if you aren’t old yet, you likely have an older friend or relative in your life with whom you are close. Many seniors, like my grandmother, complain that their memory is failing them in their old age. Nani forgets where she placed her keys, has trouble recalling recently-learned names, and sometimes even forgets childhood facts. It can be difficult watching someone you love lose bits and pieces of their memory, and it’s even more upsetting to hear their sadness when they talk about how much they think they’ve lost. Because of these difficulties associated with age, and because there is such a fear in our society of this inevitable course, it isn’t surprising that there is lots of incredible cognitive aging research being conducted. One of the more interesting articles recently published investigated not only memory ability in healthy older adults, but metamemory ability.

Metamemory is an individual’s ability to assess their own memory ability. For instance, if you were to test someone on a series of historical facts and then ask them how they did, their assessment of their own performance in comparison to their actual performance would represent their metamemory ability. The closer their assessment is to their actual performance, the better their metamemory is. In 2012, researchers at the University of Chicago hoped to assess whether there were any differences between the metamemory abilities of older and younger adults. Additionally, they hoped to examine whether recall ability, that is, the ability to recall previously learned information, is somehow connected with metamemory. Without this type of self-awareness, students cannot recognize when they know or don’t know test material. Similarly, in older adults with poor metamemory,  it can be extremely difficult for them to decide whether they’re beginning to experience memory issues or if their ability to properly assess their own memory is declining.

The researchers tested a variety of older and younger adults by exposing them to two types of images during a study phase. The first type was colored or greyscale picture and the second was words, colored in either red or blue. The researchers then tested participants’ recall ability by seeing whether they could remember and correctly identify the previously-learned words and pictures. After making each decision, participants were asked to rate their confidence in the accuracy of their answer. This confidence judgment was made using a percentage scale from 50% to 100%, with 50% representing pure chance and 100% representing certainty. The average confidence percentage was compared with the average accuracy of each participant in order to assess each participant’s metamemory ability.

Younger adults were better at accurately recalling objects compared to the older participants, which is to be expected, as slowed thought processes often occur with age. However, when the researchers tested metamemory, this created a potential bias because younger adults might be more confident in their test results than older adults, simply because the task was relatively easier for them. In order to minimize the deficits caused by age differences when assessing metamemory, the older adults were given a slightly easier version of the task, in which the test phase occurred more closely to the study phase than it did for the younger adults.  So, rather than studying both words and pictures at once like the younger adults, older adults were tested separately on the two categories.  Additionally, the study phase was made more difficult for younger adults by presenting them with extra information, designed to divide their attention and thereby decrease recall performance. After all of these precautions, what happened when researchers asked the participants to guess how well they felt they had performed during the task?

Older adults, in addition to their slightly lower performance on recall were also less accurate in evaluating that performance. Whereas younger adults could rather easily land in the right ballpark when assessing how well or how poorly they’d done, some older adults often thought they performed much worse than they actually had, and some thought they had done much better.

“Cool!” you might say, but what does this mean for my Nani, your father, or you? First, there is good news in that many healthy older adults are actually better at remembering things than they think or say they are. Sure, they might forget a name or misplace a note, but these little episodes aren’t as indicative of overall memory as our loved ones seem to think. Just as a couple of wrong answers in the experiment didn’t necessarily indicate a poor performance, a misplaced car key doesn’t necessarily mean your father is suffering from dementia. On the other hand, over estimation of memory ability in older adults could lead to underdiagnoses of memory deficits and diseases such as Alzheimer’s. An important point here for many family members and healthcare professionals is that it’s important to test an older adult’s memory rather than assuming their claims that they’re ‘slipping’ or ‘doing just fine’ are accurate. There are all sort of reasons we forget things at all ages, and it’s important to realize that as we grow older, we may be losing the ability not simply to remember, but to be aware of what it is we can and can’t remember. So, next time Nani tells me her memory isn’t what it used to be, I’ll tell her that may be true, but her memory is probably better than she thinks. Then again, if she tells me she’s doing just fine, I might recommend that she get a yearly test, just to be sure.

Citation:

Wong JT, Cramer SJ, Gallo DA. Age-related reduction of the confidence-accuracy relationship in episodic memory: Effects of recollection quality and retrieval monitoring. Psychology & Aging 2012 Dec;27(4):1053-65. doi: 10.1037/a0027686. http://www.ncbi.nlm.nih.gov/pubmed/22449027/.

Categories: Aging, Memory Tags: ,
  1. December 3rd, 2013 at 22:30 | #1

    It would be interesting if researchers were able to somehow compare older individuals with mild dementia of the Alzheimer’s type–a clinical dementia rating (CDR) of .5 to older adults without any memory impairments to see if the adults with Alzheimer’s compared in terms of metamemory to the healthy adults. I’m curious about this because with such a low CDR I wonder whether or not these individuals would be any less accurate than the healthy individuals.

    Additionally, when we talked about metacognition in class, we learned that students had the same prediction and postdiction results. That is to say, it didn’t matter whether students were asked to assess preparedness or performance before or after the exam–there metacognition was still the same. With this in mind, it would be interesting to see if the adults in the study exhibit the same effect with their metamemory. Would having the participants predict their performance on the task before completion be any more or less accurate than when they were asked to assess their performance after the fact. Taking this idea one step further, would this pre-test prediction have any effect on post-test assessment? In other words, would asking someone to think about how they would do cause them to be more aware of their performance during the actual test and in turn result in a more accurate post-test assessment?

  2. December 3rd, 2013 at 22:10 | #2

    I really enjoyed this post and it ended in a positive note too, which was a good feeling. I also think there is a societal understanding that with age, memory declines and hence there is a self-fulfilling prophecy that comes along with this kind of behavior. In addition, there is a practice of putting down elders and that might also explain the underestimation and poor metacognition. However, it is good to know that next time my grandma puts herself down, I can tell her that it is not as bad as she thinks!

  3. December 2nd, 2013 at 21:11 | #3

    I thought this was a really cool article. I think for future research it would be really interesting if they could test healthy older adults against older adults with dementia or some other age related cognitive impairment to see how a healthy adult compares. I wonder if their metamemory would be significantly better, or if because people with dementia know their memory is failing, their metamemory would be good? It might also be a cool study if they tested young adults with learning disabilities like ADHD or dyslexia, to see if they had a similar metamemory as their healthy counterparts.

  4. December 1st, 2013 at 22:14 | #4

    It is interesting to know that assessing how much information is stored in your memory (metacognition) is related to recalling the information you have stored in your memory. Could this also mean that recalling ability is influenced by metacognition? In other words, does poor metacognitive ability lead to decline in the ability to recall?
    Also, while reading this post, the term capacity of memory came to my mind. I wondered if poor metacognition in older adults represent reduced memory capacity because memory had declined with age. I would love to know about the correlation between capacity of memory and metacognition.
    On the other hand, it was certainly fascinating to learn that older adults remember better than what they say!

  5. December 1st, 2013 at 21:36 | #5

    This is very relatable as my older relatives always joke about how their memory is decreasing! Now I can assure them that it’s probably not as bad as they think. We talked about this in class in terms of metacognition, and how some students over estimated or underestimated their performance on a certain exam. I’d be curious to find out more about why older adults underestimated their results, and if that is somehow linked to their metacognition from when they were younger–like if they always underestimated themselves or if the underestimation was brought on by old age. How would a delay between the probing and recall effect performance? I also just wonder how the way the participants were probed affected their performance, in that if individuals were tested in a listening task, would the results change? Overall, I think this is very applicable and valid, and I enjoyed reading this!

You must be logged in to post a comment.