Abstract
Cognitive aging is the study of how mental abilities change across the adult lifespan — which decline, which are preserved, and why the two diverge. Fluid abilities such as processing speed, working memory, and reasoning fall gradually from early adulthood, while crystallized knowledge is maintained or grows into old age. This article states the fluid–crystallized distinction, the processing-speed theory that ties diverse declines to a single slowing, the inhibitory-deficit account, and the neurocognitive frameworks — compensation, scaffolding, and brain maintenance — that explain why some older adults preserve function. It distinguishes normal aging from disease and confronts the cross-sectional/longitudinal discrepancy that has shaped every estimate of decline. Three interactive demonstrations let the reader trace fluid and crystallized trajectories, watch processing speed cascade into higher-order performance, and see how cohort effects inflate cross-sectional estimates of age-related loss.
Keywords: cognitive aging, fluid intelligence, processing speed, brain maintenance, cognitive reserve
Cognitive aging names the ordinary, non-pathological changes in perception, memory, and thought that accompany getting older. Its central and most durable finding is not that the mind declines but that it differentiates: some abilities fall steadily from early adulthood while others hold firm or keep improving into the eighth decade. The abilities that decline are the ones that operate on novel material in the moment — the speed of processing, the amount held in mind at once, the on-the-fly reasoning that solves an unfamiliar problem. The abilities that are spared are the ones built from accumulated experience — vocabulary, general knowledge, the practiced routines of a lifetime (Horn & Cattell, 1967). A single portrait of the aging mind as uniform loss is therefore wrong; the real portrait is of two trajectories pulling apart.
The field's task is to explain both trajectories at once — why the fluid abilities erode, why the crystallized ones do not, and why individuals differ so enormously in how steeply the fluid decline runs. That last fact, the wide spread of outcomes, is what makes cognitive aging a science of maintenance as much as of decline: some older adults perform like people decades younger, and the frameworks developed since 2000 are largely attempts to say what those people are doing differently (Cabeza et al., 2018). Throughout, a boundary has to be kept in view: normal cognitive aging is not dementia. The gradual slowing described here is the healthy baseline against which disease is defined, not a mild form of it (Harada et al., 2013).
- Cognitive aging is differentiation, not uniform decline: fluid abilities (speed, working memory, reasoning) fall from early adulthood while crystallized knowledge is maintained or grows.
- The processing-speed theory holds that a single age-related slowing accounts for much of the decline across otherwise unrelated fluid tasks.
- The inhibitory-deficit hypothesis attributes memory and comprehension decline to weakened control over what enters and stays in attention, not to storage loss.
- Neurocognitive frameworks — the HAROLD compensation model, scaffolding theory, and brain maintenance — explain why some older adults preserve function despite structural brain change.
- Cross-sectional designs overstate decline because they confound aging with cohort differences; longitudinal designs show gentler, later declines and are the harder evidence.
What Cognitive Aging Is
Cognitive aging is defined by a pattern rather than a single number: the systematic, individually variable changes in cognition that occur across healthy adult development. The organizing distinction, inherited from the psychometric study of intelligence, is between fluid and crystallized abilities (Horn & Cattell, 1967). Fluid intelligence is the capacity to reason and solve problems in novel situations, independent of acquired knowledge — spotting the pattern in an unfamiliar sequence, holding several conditions in mind while manipulating them. Crystallized intelligence is the store of knowledge and skill built up through education and experience — vocabulary, facts, expertise. The two are measured by different tests and, crucially, they age differently: fluid measures peak in the twenties and decline more or less linearly thereafter, while crystallized measures rise through midlife and plateau, often not falling until very late in life if at all.
This divergence is the first thing any account of cognitive aging must capture, and it immediately rules out the folk model of the aging brain as a uniformly fading machine. What ages is a specific class of operations — those performed quickly, on new material, under load — and not the accumulated products of past cognition. MeSH files cognitive aging under the broader descriptor Aging (descriptor D000375, tree G07.345.124), and the placement is apt: cognitive aging is the cognitive slice of a whole-organism process, subject to the same distinction between what senesces and what is conserved.
A second defining feature is the sheer range of individual outcomes. Averaged curves hide the fact that some seventy-year-olds outperform the average forty-year-old on fluid tasks while others have declined steeply, and this variance grows with age rather than shrinking (Deary et al., 2009). Explaining that spread — what protects the high performers — has become the central preoccupation of the field, and the reason its modern theories are theories of preserved function, not only of loss.
Five accounts organize the modern literature, and the rest of this article takes them in turn. They are not rivals to be adjudicated so much as answers to different questions — one names the proximate mechanism of fluid decline, one names the control failure behind memory change, and three explain why the decline is so much gentler in some people than in others. Table 1 previews them.
| Account | Core claim | Chief proponent |
|---|---|---|
| Processing-speed theory | A single, general slowing of elementary operations accounts statistically for much of the decline across unrelated fluid tasks | Salthouse |
| Inhibitory-deficit hypothesis | Age differences in working memory and comprehension arise from weakened inhibitory control over the contents of attention, not from lost storage | Hasher & Zacks |
| HAROLD (compensation) | Reduced hemispheric asymmetry in older adults reflects additional neural recruitment that compensates for declining efficiency | Cabeza |
| Scaffolding theory (STAC) | The aging brain responds to decline by recruiting compensatory scaffolding, whose strength is shaped by a lifetime of enrichment | Park & Reuter-Lorenz |
| Brain maintenance | The sharpest older adults are those whose brains have accumulated the least age-related pathology in the first place | Nyberg |
The Processing-Speed Theory
The most influential single-mechanism account of cognitive aging is Timothy Salthouse's processing-speed theory (Salthouse, 1996). Its claim is economical and strong: much of the age-related decline observed across a wide variety of fluid tasks — memory, reasoning, spatial ability — is not a set of separate deficits but the downstream consequence of a single, general slowing of the speed at which elementary cognitive operations are executed. When older adults are slower, two mechanisms degrade performance. The limited-time mechanism means that later operations in a sequence get less time because earlier ones consume it; the simultaneity mechanism means that the products of early processing decay before later processing can use them. A slow system does not merely take longer — it arrives at worse answers.
The theory's empirical signature is statistical. When a measure of perceptual-comparison speed is entered first in an analysis of age differences on a complex cognitive task, it absorbs a large share of the age-related variance, often leaving little for age to explain independently (Salthouse, 1996). Speed behaves, in the statistics, like a common cause sitting between age and cognition. This does not prove that slowing causes the higher-order declines — a shared third factor could drive both — but the consistency of the mediation across dozens of studies made processing speed the default explanation to beat, and the benchmark every rival mechanism is measured against (Salthouse, 2019).
Figure 1
Fluid and Crystallized Trajectories Across the Adult Lifespan
The demonstration below makes the fluid–crystallized divergence concrete. The reader moves an age marker across the adult lifespan and watches the two ability curves separate — fluid falling from its early peak, crystallized holding and even rising — so that the same person is simultaneously getting worse at some things and better at others.
Fluid and crystallized trajectories
Drag the age marker across the adult lifespan. The same person is getting worse at fluid tasks and holding — even gaining — on crystallized ones.
At age 50, fluid ability is 49 and crystallized ability is 58. The gap between the two curves is the divergence any account of cognitive aging must explain: crystallized knowledge has overtaken fluid reasoning.
Figure. Fluid and crystallized T-scores as fixed functions of age. Curves are schematic, following the pattern of Horn & Cattell (1967) and Park et al. (2002). Nothing is sampled. Original illustration.
Memory, Inhibition, and Control
Not every decline reduces to speed. A second major mechanism concerns inhibition — the ability to keep irrelevant information out of working memory and to suppress no-longer-relevant material once it is in. Lynn Hasher and Rose Zacks proposed that many age differences in working memory and language comprehension arise not because storage capacity shrinks but because inhibitory control weakens, so that the older adult's working memory fills with off-goal thoughts, prior interpretations, and distracting detail that a younger adult would have filtered out (Hasher & Zacks, 1988). On this view the older mind is not a smaller container but a leakier one: its effective capacity is reduced because it is cluttered with what should have been excluded.
The inhibitory-deficit account reframes a range of findings. Older adults' apparent working-memory shortfall, their greater susceptibility to interference, their difficulty suppressing a habitual response — all become expressions of one weakened control process rather than separate losses. The account also predicts, unusually, occasional advantages: because older adults retain more incidental and off-goal information, they sometimes show better memory for material a younger adult discarded as irrelevant, a signature the pure-decline models cannot easily produce. Working memory itself — the limited system that holds and manipulates information in the service of ongoing cognition — is the arena in which speed and inhibition both do their damage, and lifespan studies show its measures among the earliest and steepest to decline (Park et al., 2002).
The demonstration below isolates the processing-speed mechanism. The reader sets the speed of a simulated system's elementary operations and watches performance on a downstream, multi-step task change — few operations completed slowly means later steps run short of time and early products decay, so a single speed parameter drives the higher-order score.
The processing-speed cascade
Set the speed of elementary operations. A single slowing hits a downstream, multi-step task twice: fewer steps finish, and the ones that do finish decay before later steps can use them.
At 100% speed, 10 of 10 operations finish and retained products fall to 100%, so the higher-order score is 100. A 0-point drop in speed costs 0 points downstream — the two mechanisms multiply, which is why a modest slowing produces a large cognitive deficit.
Figure. Higher-order score = (operations completed ÷ 10) × retention, both fixed functions of the speed knob. Nothing is sampled. Original illustration.
The Aging Brain: Compensation, Scaffolding, and Maintenance
Cognitive aging has a neural story, and since 2000 it has been dominated by three linked ideas that together explain preserved function. The first is compensation. Functional imaging revealed that older adults often recruit more brain tissue for a task than younger adults — most strikingly, they engage both prefrontal hemispheres where younger adults engage one. Roberto Cabeza named this the HAROLD model, for hemispheric asymmetry reduction in older adults, and argued that the extra recruitment is not noise or dysfunction but compensation: the aging brain enlists additional resources to counteract declining efficiency, and older adults who show the bilateral pattern often perform better than those who do not (Cabeza, 2002). Over-activation, on this reading, is the brain working to hold performance up, not a sign of it breaking down (Grady, 2012).
The second idea generalizes the first. The Scaffolding Theory of Aging and Cognition, developed by Denise Park and Patricia Reuter-Lorenz, casts the extra recruitment as scaffolding — the brain's normal, lifelong response to challenge, in which secondary circuits are erected to support a primary one that has become inefficient (Park & Reuter-Lorenz, 2009). Aging is one such challenge, met by the same scaffolding process that supports learning at any age. The theory's revision made the compensatory scaffold something that can itself be strengthened by education, exercise, and engagement, tying the neural account to the lifestyle factors that predict better cognitive outcomes (Reuter-Lorenz & Park, 2014).
The third idea is the most parsimonious. Brain maintenance, articulated by Lars Nyberg and colleagues, holds that the primary determinant of successful cognitive aging is not how well the brain compensates for damage but how little damage it accumulates in the first place (Nyberg et al., 2012). On this account the older adults who stay sharp are those whose brains show the least age-related pathology — the least atrophy, the fewest lesions, the best-preserved dopamine system — and maintenance, not compensation, is the leading edge of the difference (Nyberg & Pudas, 2019). A synthesis now treats the three as complementary levels of a single account: maintenance keeps the hardware intact, reserve buffers against the damage that does occur, and compensation is the active response when function is threatened (Cabeza et al., 2018). Longitudinal work linking the trajectory of brain structure to the trajectory of cognition is the empirical program these frameworks now share (Oschwald et al., 2020).
Worked Example
Consider why the design of an aging study changes the decline it finds — the single most important methodological fact in the field. A cross-sectional study tests people of different ages once and compares them; a longitudinal study tests the same people repeatedly as they age. Suppose a cross-sectional study finds that 25-year-olds score 55 on a reasoning test and 75-year-olds score 35. The apparent decline is 55 − 35 = 20 points over the 50-year span, a slope of 20 ÷ 50 = 0.40 points per year.
But the two groups differ by more than age. The older cohort was born 50 years earlier, with less schooling, poorer childhood nutrition, and no lifetime of test-taking — advantages that raise the younger cohort's score for reasons unrelated to aging. Suppose that cohort advantage is worth 8 points: the 75-year-olds would have scored only 47, not 55, at age 25. Then the part of the 20-point gap actually due to aging is 20 − 8 = 12 points, a true longitudinal slope of 12 ÷ 50 = 0.24 points per year. The cross-sectional design overstates the rate of aging by (0.40 − 0.24) ÷ 0.24 = 66.7%.
The arithmetic is the whole reason the cross-sectional and longitudinal literatures disagree, and it runs in one direction: because each successive generation has scored higher on fluid tests — the secular rise James Flynn documented across fourteen nations, now called the Flynn effect (Flynn, 1987) — cross-sectional comparisons confound aging with cohort and inflate decline, placing its onset earlier and its slope steeper than the longitudinal evidence supports (Schaie, 1994; Salthouse, 2019). It is why K. Warner Schaie's Seattle Longitudinal Study, following the same people for decades, found most mental abilities maintained into the sixties — a far gentler picture than the cross-sectional slope of 0.40 would predict.
Discussion
The deepest unresolved problem in cognitive aging is the cross-sectional/longitudinal discrepancy the worked example makes concrete. The two designs yield systematically different answers: cross-sectional data show fluid decline beginning in the twenties and running steeply, while longitudinal data show later onset and gentler slopes (Schaie, 1994). Neither is simply correct. Cross-sectional estimates are inflated by cohort effects — the older group was disadvantaged from the start — but longitudinal estimates are deflated by their own artifacts: practice effects from repeated testing, and the selective dropout of exactly the participants who are declining fastest, which flatters the surviving sample (Salthouse, 2019). The truth lies between two biased estimates, and pinning it down remains an active methodological fight.
A second open question is whether the single-mechanism ambition of the processing-speed theory can survive. Speed explains an impressive share of age-related variance, but statistical mediation is not causal mediation: a common upstream factor — declining white-matter integrity, say, or a shared neuromodulatory decline — could produce both the slowing and the higher-order deficits without either causing the other (Salthouse, 2019). The same common-cause logic underlies dedifferentiation, the observation that distinct cognitive abilities become more strongly correlated in old age, as though a single factor increasingly governs performance across domains — a pattern consistent with one shared mechanism but equally producible by several that decline in step. The neurocognitive frameworks face the mirror-image problem: compensation, scaffolding, reserve, and maintenance are richly descriptive, but disentangling them empirically is hard, because a brain that over-recruits, a brain with more reserve, and a brain that has simply stayed healthy can look similar at a single snapshot (Cabeza et al., 2018). This is why the field has turned decisively toward longitudinal, multi-modal designs that watch brain and behavior change together over years (Oschwald et al., 2020).
Current Directions
Three developments define the contemporary front. The first is the ascendancy of brain maintenance as the leading account of successful aging, and with it a shift in emphasis from repairing decline to preventing it. If the sharpest older adults are those whose brains have accumulated the least damage, then the research question becomes what preserves brain integrity across decades — genetic endowment, vascular health, education, physical activity — and the intervention target moves upstream, from cognitive training late in life to lifelong protection of the substrate (Nyberg & Pudas, 2019). This reframes cognitive aging as, in large part, a problem of brain health.
The second is the integration of the once-competing neural frameworks into a single, testable hierarchy. Rather than asking whether aging is explained by compensation or reserve or maintenance, current work treats them as distinct, measurable contributions — maintenance of the hardware, reserve as a buffer, compensation as an active response — and asks how much each contributes in a given person (Cabeza et al., 2018). The third is methodological and is the enabling condition for the other two: large longitudinal cohorts followed with repeated imaging and cognitive testing, designed specifically to estimate correlated change — whether the people whose brains decline fastest are the people whose cognition declines fastest — which is the evidence the cross-sectional era could never provide (Oschwald et al., 2020). The demonstration below shows why that evidence matters, letting the reader impose a cohort effect on a simulated dataset and watch the cross-sectional and longitudinal slopes pull apart.
Why designs disagree: cohort inflation
Add a cohort advantage — points the younger group scores for schooling and nutrition, not aging. The cross-sectional slope absorbs it; the longitudinal slope does not.
The cross-sectional design shows a 20-point drop — a slope of 0.40 pts/yr — but only 12 points are aging; the rest is cohort. The true longitudinal slope is 0.24 pts/yr, so the cross-sectional estimate overstates the rate of aging by 67%.
Figure. Cross-sectional slope = (true decline + cohort advantage) ÷ 50; longitudinal slope = true decline ÷ 50. Both are fixed functions of the cohort knob. Nothing is sampled. Original illustration.
Common Misconceptions
- Cognitive aging means across-the-board mental decline.
- It does not. Fluid abilities decline while crystallized knowledge is maintained or grows, so the aging mind gets worse at some things and better at others rather than uniformly fading (Horn & Cattell, 1967).
- Normal cognitive aging is early or mild dementia.
- Normal aging is the healthy baseline against which dementia is defined, not a mild version of it. The gradual slowing of healthy aging differs in kind and course from the pathological loss of Alzheimer's disease (Harada et al., 2013).
- More brain activity in older adults is a sign of damage.
- Often the reverse: the extra, bilateral recruitment older brains show is compensatory scaffolding, and the older adults who over-recruit frequently perform better than those who do not (Cabeza, 2002).
Glossary
- Brain maintenance.
- The account holding that successful cognitive aging results primarily from accumulating little age-related brain pathology, rather than from compensating for damage after it occurs.
- Cognitive reserve.
- The capacity, built by education, occupation, and engagement, to sustain cognitive performance despite brain damage or age-related change; a buffer between pathology and its expression.
- Cohort effect.
- A difference between age groups caused by their different years of birth — schooling, nutrition, test familiarity — rather than by aging; the chief confound of cross-sectional designs.
- Compensation.
- The recruitment of additional neural resources — such as the second prefrontal hemisphere in the HAROLD model — to counteract declining processing efficiency and hold performance up.
- Cross-sectional design.
- A study that tests people of different ages once and compares them; fast but confounded by cohort effects, which inflate estimates of age-related decline.
- Crystallized intelligence.
- The store of knowledge and skill acquired through education and experience — vocabulary, facts, expertise — which is maintained or grows through midlife and declines little with age.
- Dedifferentiation.
- The tendency for distinct cognitive abilities to become more strongly correlated in old age, as if a common factor increasingly governs performance across domains.
- Fluid intelligence.
- The capacity to reason and solve problems in novel situations independent of acquired knowledge; peaks in early adulthood and declines steadily thereafter.
- Flynn effect.
- The secular rise in average IQ-test performance across successive generations; because it lifts each younger cohort's scores, it is the engine of the cohort confound that inflates cross-sectional estimates of cognitive decline.
- HAROLD model.
- Hemispheric Asymmetry Reduction in Older Adults: the finding that older brains engage both prefrontal hemispheres where younger brains engage one, interpreted as compensation.
- Inhibitory-deficit hypothesis.
- The proposal that age differences in working memory and comprehension stem from weakened inhibitory control over the contents of attention, not from reduced storage capacity.
- Longitudinal design.
- A study that tests the same people repeatedly as they age; avoids cohort confounds but is subject to practice effects and selective dropout, which deflate estimates of decline.
- Processing speed.
- The rate at which elementary cognitive operations are executed; its age-related slowing is proposed to account statistically for much of the decline across fluid tasks.
- Scaffolding (STAC).
- In the Scaffolding Theory of Aging and Cognition, the brain's normal recruitment of secondary circuits to support an inefficient primary one; strengthened by education, exercise, and engagement.
- Terminal decline.
- The accelerated drop in cognitive performance observed in the years immediately preceding death, distinct from the gradual decline of normal aging.
- Working memory.
- The limited-capacity system that holds and manipulates information in the service of ongoing cognition; among the earliest and steepest of the fluid abilities to decline with age.
Key Researchers
Roberto Cabeza
(contemporary). Professor at Duke University; proposed the HAROLD model, reframing reduced prefrontal lateralization in older adults as functional compensation rather than decline. Google Scholar - Faculty Page
Lynn Hasher
(contemporary). Professor at the University of Toronto; with Rose Zacks developed the inhibitory-deficit hypothesis, locating age differences in weakened control over the contents of attention. Wikipedia - Google Scholar - Faculty Page
Lars Nyberg
(contemporary). Professor at Umeå University and PI of the Betula study; articulated brain maintenance as the leading account of successful memory aging. Google Scholar - Faculty Page
Denise C. Park
(1951-2026). Founder of the Center for Vital Longevity at the University of Texas at Dallas; co-developed the Scaffolding Theory of Aging and Cognition and led the Dallas Lifespan Brain Study. Wikipedia
Patricia A. Reuter-Lorenz
(contemporary). Professor at the University of Michigan; co-developed scaffolding theory and pioneered functional-imaging work linking age-related over-recruitment to compensation. Wikipedia - Google Scholar - Faculty Page
Timothy A. Salthouse
(contemporary). Professor at the University of Virginia; formulated the processing-speed theory of cognitive aging and directs the Virginia Cognitive Aging Project. Wikipedia - Faculty Page
K. Warner Schaie
(1928-2023). Professor at Pennsylvania State University; founded the Seattle Longitudinal Study, the definitive longitudinal evidence that most mental abilities are maintained into the sixties. Wikipedia
Frequently Asked Questions
What is cognitive aging in simple terms?
Cognitive aging is how mental abilities change as adults get older. It is not uniform decline: abilities that work on novel material in the moment, such as processing speed and reasoning, fall gradually, while accumulated knowledge like vocabulary is maintained or grows (Horn & Cattell, 1967).
Which mental abilities decline with age and which are preserved?
Fluid abilities (processing speed, working memory, and reasoning about unfamiliar problems) decline from early adulthood, whereas crystallized abilities (vocabulary, facts, and expertise) are maintained or improve through midlife and fall little until very late (Park et al., 2002).
Is cognitive aging the same as dementia?
No. Normal cognitive aging is the healthy baseline against which dementia is defined. Its gradual slowing differs in kind and course from the pathological memory loss of Alzheimer's disease and other dementias (Harada et al., 2013).
What is the processing-speed theory of aging?
It holds that a single, general slowing of elementary cognitive operations accounts statistically for much of the age-related decline seen across otherwise unrelated fluid tasks, because slower processing leaves later operations short of time and lets early products decay (Salthouse, 1996).
Why do older adults use more of their brain for the same task?
The extra, often bilateral recruitment is interpreted as compensation: the aging brain enlists additional resources to counteract declining efficiency, and older adults who show the pattern frequently perform better than those who do not (Cabeza, 2002).
What is brain maintenance?
Brain maintenance is the idea that the sharpest older adults are those whose brains have accumulated the least age-related damage in the first place, making prevention of pathology, rather than compensation for it, the primary route to successful aging (Nyberg et al., 2012).
Why do studies disagree about how fast cognition declines?
Cross-sectional studies compare different-aged people and overstate decline by confounding aging with cohort differences, while longitudinal studies follow the same people and understate it through practice effects and selective dropout; the truth lies between (Schaie, 1994).
Can anything protect cognition in old age?
The evidence points to lifelong factors that preserve brain integrity and build reserve, such as education, physical activity, and cognitive engagement, consistent with scaffolding and brain-maintenance accounts, though disentangling cause from correlation remains difficult (Reuter-Lorenz & Park, 2014).
References
Cabeza, R. (2002). Hemispheric asymmetry reduction in older adults: The HAROLD model. Psychology and Aging, 17(1), 85-100. https://doi.org/10.1037/0882-7974.17.1.85
Cabeza, R., Albert, M., Belleville, S., Craik, F. I. M., Duarte, A., Grady, C. L., Lindenberger, U., Nyberg, L., Park, D. C., Reuter-Lorenz, P. A., Rugg, M. D., Steffener, J., & Rajah, M. N. (2018). Maintenance, reserve and compensation: The cognitive neuroscience of healthy ageing. Nature Reviews Neuroscience, 19(11), 701-710. https://doi.org/10.1038/s41583-018-0068-2
Deary, I. J., Corley, J., Gow, A. J., Harris, S. E., Houlihan, L. M., Marioni, R. E., Penke, L., Rafnsson, S. B., & Starr, J. M. (2009). Age-associated cognitive decline. British Medical Bulletin, 92(1), 135-152. https://doi.org/10.1093/bmb/ldp033
Flynn, J. R. (1987). Massive IQ gains in 14 nations: What IQ tests really measure. Psychological Bulletin, 101(2), 171-191. https://doi.org/10.1037/0033-2909.101.2.171
Grady, C. (2012). The cognitive neuroscience of ageing. Nature Reviews Neuroscience, 13(7), 491-505. https://doi.org/10.1038/nrn3256
Harada, C. N., Natelson Love, M. C., & Triebel, K. L. (2013). Normal cognitive aging. Clinics in Geriatric Medicine, 29(4), 737-752. https://doi.org/10.1016/j.cger.2013.07.002
Hasher, L., & Zacks, R. T. (1988). Working memory, comprehension, and aging: A review and a new view. In G. H. Bower (Ed.), The psychology of learning and motivation (Vol. 22, pp. 193-225). Academic Press. https://doi.org/10.1016/S0079-7421(08)60041-9
Horn, J. L., & Cattell, R. B. (1967). Age differences in fluid and crystallized intelligence. Acta Psychologica, 26, 107-129. https://doi.org/10.1016/0001-6918(67)90011-X
Nyberg, L., Lövdén, M., Riklund, K., Lindenberger, U., & Bäckman, L. (2012). Memory aging and brain maintenance. Trends in Cognitive Sciences, 16(5), 292-305. https://doi.org/10.1016/j.tics.2012.04.005
Nyberg, L., & Pudas, S. (2019). Successful memory aging. Annual Review of Psychology, 70, 219-243. https://doi.org/10.1146/annurev-psych-010418-103052
Oschwald, J., Guye, S., Liem, F., Rast, P., Willis, S., Röcke, C., Jäncke, L., Martin, M., & Mérillat, S. (2020). Brain structure and cognitive ability in healthy aging: A review on longitudinal correlated change. Reviews in the Neurosciences, 31(1), 1-57. https://doi.org/10.1515/revneuro-2018-0096
Park, D. C., Lautenschlager, G., Hedden, T., Davidson, N. S., Smith, A. D., & Smith, P. K. (2002). Models of visuospatial and verbal memory across the adult life span. Psychology and Aging, 17(2), 299-320. https://doi.org/10.1037/0882-7974.17.2.299
Park, D. C., & Reuter-Lorenz, P. (2009). The adaptive brain: Aging and neurocognitive scaffolding. Annual Review of Psychology, 60, 173-196. https://doi.org/10.1146/annurev.psych.59.103006.093656
Reuter-Lorenz, P. A., & Park, D. C. (2014). How does it STAC up? Revisiting the scaffolding theory of aging and cognition. Neuropsychology Review, 24(3), 355-370. https://doi.org/10.1007/s11065-014-9270-9
Salthouse, T. A. (1996). The processing-speed theory of adult age differences in cognition. Psychological Review, 103(3), 403-428. https://doi.org/10.1037/0033-295X.103.3.403
Salthouse, T. A. (2019). Trajectories of normal cognitive aging. Psychology and Aging, 34(1), 17-24. https://doi.org/10.1037/pag0000288
Schaie, K. W. (1994). The course of adult intellectual development. American Psychologist, 49(4), 304-313. https://doi.org/10.1037/0003-066X.49.4.304