Cognitive engagement, cognitive plasticity, and health disparities
||November 7th and 14th at 9:00 11:30am
November 7th at ISERP room 801
November 14th at MSPH 9th Floor Conference Room
“Anna spoke wistfully. ‘It must make you happy, Jim, to have fine thoughts like that in your mind all the time, and to have words to put them in. I always wanted to go to school you know."'
-- My Ántonia
Does using your brain change it for the better? More specifically, does using your brain change it in either the long or short term in ways that delay or hasten damage – inducing pathological processes or make the brain resilient to damage once it is incurred?
If this were the case, it suggests that we might reduce cognitive decline in old age by promoting cognitive engagement in the elderly or earlier in life. To the extent that cognitive challenges and exposures differ across socioeconomic, ethnic/racial, or gender groups, this may also help understand and remediate disparities in cognitive aging and in health more broadly (cognitive impairment appears to strongly increase risk of physical morbidity and death).
Cognitive function in the elderly is a reflection of 3 distinct phenomena (figure): (1) where you started, (2) how much improvement/damage/decline you have experienced since you started (eg due to a disease process), and (3) how your brain has responded to that damage (e.g. reorganized to reduce functional consequences). Exposures that influence one of these phenomena may have no effect on the others (or even the opposite effect).
Here are some questions I’d be interested in discussing – if possible please respond to one of these before seminar and send me your thoughts:
- How likely is it that cognitive skills play an important role in explaining health disparities? What are the most important socially patterned environmental differences that influence cognitive outcomes? (i.e. are primary differences in the content of schooling or pre-school exposures or workplace exposures etc?)
- How convincing is the evidence for a beneficial effect of cognitive engagement on cognitive aging and what would more convincing evidence look like?
- What exactly might cognitive engagement do (i.e. structural changes versus how you “use” your brain? Neurogenesis, more synapses, or more “efficient” use of the networks you already have)?
- Which specific cognitive skills link to which specific health outcomes? Is it useful to disentangle specific cognitive skills from each other and from “IQ” and what’s the right framework for categorizing cognitive skills?
- Which cognitive skills can we change and how? What fosters cognitive engagement in beneficial ways? Are there specific ages for plasticity or is change possible throughout life?
Vaguely along these lines, I’ve suggested the following readings for seminar:
- Stern “Cognitive Reserve and Alzheimer Disease” Alz & Assoc Disorders, 2006.
- Willis et al “Long-term effects of cognitive training on everyday functional outcomes in older adults” JAMA 2006
- Fischer et al “Recovery of learning and memory is associated with chromatin remodeling” Nature 2007 - OR - Marilyn Albert (NEJM 2007) provides a very friendly 2 page summary of Fischer which you may prefer to read instead.
- Farah et al “Childhood poverty: specific associations with neurocognitive development” Brain Research, 2006
As further background, here are a few relevant research findings that support the hypothesis that cognitive engagement influences cognitive aging, and my opinions about them. Much of this is covered in Stern’s review paper:
- People with more education have later onset or less risk altogether of dementing illness.
- This is not fabulously informative because dementia diagnoses require crossing a threshold of impairment. Thus, if people with more education enter old age with higher functioning, they may have the same or faster rates of decline and no extra resilience to pathology and yet still have delayed dementia diagnosis. If people with high education enter old age with a higher level of cognitive function because the education did something beneficial for their brains, this is an interesting finding and perhaps suggests some useful interventions. If people with high education enter old age with a higher level of cognitive function because they were born with higher levels of cognitive functioning and consequently pursue more education, this is not really such an interesting finding.
- People with more education have slower rates of cognitive change across age.
- Although there are a lot of studies that purport to show this, I think most of them are biased by the way they set up the regression models, such that they are really just showing a cross-sectional relationship between cognitive function and education.
- People who engage in “cognitively stimulating activities”, such as crossword puzzles, experienced delayed onset of dementia or slower rates of cognitive decline.
- The critique here is that people may withdraw from activities that present mental challenges when they experience subtle, subclinical cognitive changes. Even with long delays between the activities assessment and the cognitive assessments, I am dubious about how to rule out cognitive changes. And possibly low cognitive skills (IQ?) are independent causes (or markers of an independent cause) of cognitive aging. Studies show cognitive differences in people ultimately diagnosed with dementia up to 20+ years prior to diagnosis (Framingham), at young adulthood (nun study) and at age 11 (Scottish study).
- The ACTIVE trial: initiated after promising but unconvincing results from small non-randomized studies of cognitive training, ACTIVE implemented a careful randomized design and successfully taught people to perform better on the exact tests upon which they were trained. At the two year follow-up, there was very little translation from the tests upon which people were trained to related cognitive skills or functional outcomes. By the 5 year follow-up, however, people who received the training appeared to be doing better on instrumental activities of daily living (this is one of the readings, Willis et al).
- Environmental enrichment or cognitive challenges and neurologic damage both seem to induce plasticity in animal models and, to the extent studied, in humans. The animal studies of environmental enrichment are broader, but difficult to know exactly how to translate to human findings. In animals, this research is fairly well developed and compelling (to my uncritical eye) – see for example the very cool design by Fischer et al. In humans, it’s harder to disentangle selection from causation, but we have studies showing correlations between hippocampal size and years of driving a cab or changes in gray matter induced by learning to juggle. After stroke there is prompt and substantial reorganization in the brain areas recruited to perform motor tasks. This reorganization continues to evolve for months (?) after the event, as performance of motor tasks improves, and is apparently promoted by motor challenges.
- I have largely ignored the literature on early childhood enrichment/deprivation and cognitive outcomes (hopefully someone else in seminar will feel more confident talking about this). But I think an important question is whether there are sensitive or critical periods in the lifecourse when cognitive engagement can make more of a difference. I am ambivalent about the Farah article here – is it really useful to show that SES correlates with cognitive skills in an unrepresentative sample of 60 kids? Is there something important from either a causal inference or a potential intervention perspective about linking SES to specific cognitive skills? Do the skills her results emphasize correspond with other sources of evidence and daily experience?
My view (absent a lot of empirical evidence) is that cognitive skills are extremely influential for health, including cognitive aging; cognitive skills are profoundly shaped by our environments; and that imposing environments that foster or deter cognitive development and reinforcing the impact of such environments by explicit rewards or punishment for exercise of cognitive skills is a primary means of maintaining many if not most social hierarchies.
Possibly Entirely Irrelevant Figure.
This figure illustrates three reasons it is difficult to interpret the association between education and cognitive outcomes: ceiling effects, test-taking skills, and distance to dementia. The figure contrasts two women’s scores on a brief cognitive test such as the mini-mental state exam (MMSE) compared to their hypothetical ‘true’ cognitive function, assuming such a measure were on the same scale as the MMSE. The MMSE has a maximum score of 30, and it is fairly common for individuals to score this maximum, i.e. to hit the ceiling. At age 66, the high education individual’s ‘true’ score is above the maximum score of the MMSE. Even after declines in true cognitive function, she achieves a perfect MMSE score. Only after experiencing a substantial decline does the woman’s true cognitive function fall below the maximum score for the MMSE, at which point the MMSE corresponds to her true cognitive function. The low education woman, on the other hand, may not be ‘good’ at taking the test and will score substantially below her ‘true’ cognitive function even when her score is not capped by the MMSE ceiling. Finally, because the high education woman starts higher, a large loss in cognitive function still does not reach the dementia threshold (indicated by the red line).