Introduction

Cross-sectional and longitudinal studies find robust declines in abilities such as encoding new memories of episodes or facts, working memory and information processing speed. By contrast, short-term memory, autobiographical memory, semantic knowledge and emotional processing remain relatively stable.

Behavioral research

Life-long declines. Processing speed, working memory and episodic memory showed linear life-long declines with little or no evidence for accelerated decline in the later decades. When performance is plotted as a function of time to mortality, there is an acceleration of cognitive decline that begins 3–6 years before death.

Late-life declines. Most of the adult lifespan is characterized by slight declines in well-practiced tasks or tasks that involve knowledge (also: vocabulary and semantic), with sharper declines observed after the age of 70. One possibility is that older adults use preserved knowledge and experience to form more efficient or effective strategies when performing tasks in which younger adults rely on processing ability.

Life-long stability. Autobiographical memory, emotional processing (also: attribution of mental states to other individuals 'theory of mind') and automatic memory processes seem to be unchanged throughout life. It has been found that automatic feelings of familiarity continue to be relied on even when effortful recollection fails with age.

Age-related neural changes

The brains of older adults tend to have lower volumes of grey matter than do the brains of younger adults, seemingly due to lower synaptic densities, which declines steadily over time. In particular the PFC and medial temporal structures are affected, while the occupational cortex remains relatively unaffected.

Normal and pathological aging. Normal aging involves changes in the frontostriatal system, with decreases in dopamine, noradrenaline and serotonin, and declines in the volume and function of the PFC. Pathological aging involves changes that occur primarily with pathology associated with Alzheimer’s disease, beginning with a loss of volume in the entorhinal cortex, an important relay between the hippocampus and association cortices, and progressively affecting the hippocampus proper.

PFC and striatal circuits. Structures of the PFC undergo the largest age- related volumetric changes in adulthood (decline about 5% per decade >20yr). Declines have also been found in the human striatum, an area that has extensive connections to the PFC and is responsible for a large proportion of dopamine production, and might therefore affect cognitive processes that are subserved by dopamine-dependent circuits (decline about 3% per decade).

Other declines with age: Dopamine concentration, transporter availability, dopamine D2 receptor density and serotonin receptor (5-HT2).

PET and fMRI studies show that older adults tend to exhibit less PFC activity during executive processing tasks than do younger adults

Hippocampus and medial temporal lobes (MTL). Important for declarative memory and relatively slight age-related changes in the absence of Alzheimer’s disease. Hippocampal volume declines are less apparent during normal aging, although declines in functional activations of the hippocampus and surrounding cortex have been observed in healthy older adults. By contrast, pathological processes, such as those that accompany Alzheimer’s disease, severely affect hippocampal regions. Another memory-related structure in the MTL, the amygdala, is less active in older adults than in younger adults in response to emotionally negative stimuli, but exhibits similar activity among age groups to emotionally positive stimuli.

Individual variability. Individual differences might include different life experiences, genetic influences, preferred strategies and susceptibility to neuropathology and variability within individuals across tasks might change with age. Previous longitudinal studies have found remarkable stability before the age of 60, with increased variability occurring only in later life.

Some elderly perform as well or better than younger people. Neuroimaging studies have found that elderly individuals often show greater functional activation of brain regions (usually in the PFC) that are less active in younger adults, and that such additional activations are often seen only in high- performing older adults. Therefore it seems that they show neural compensation, whereas their low-functioning counterparts, who experience failures of inhibition, show decreased activations or non-selective recruitment.

Potential predictors of accelerated cognitive decline in older adults:

1. Stay intellectually engaged

2. Stay physically active

3. Minimize chronic stressors

4. Maintain a brain-healthy diet (high in poly- and mono-unsaturated fatty acids, vitamin E and polyphenols and antioxidants).

Conclusion

Still a lot of questions are unanswered.

1. Are age-related declines due to normal or pathological processes?

2. Do normal age-related differences occur throughout adulthood, or only after some critical age?

3. To what extent does individual variability in behavioral, genetic and neurobiological markers of cognitive ageing reflect normal and pathological ageing?

4. What neural mechanisms do age-related differences in anatomy and functional activations represent?

5. To what extent are strategy changes in older adults responsible for, or a response to, neural changes?