Determinant Factors for Memory Performance in Human Aging: A Study of Environmental, Person-Based, and Non-Cognitive Predictors Contributing to Memory Performance in old Adults

Sammanfattning: The aim of this thesis was to examine the importance of environmental, non-cognitive, and person-based variables for memory performance in old adults. Study I was conducted in cooperation with the Gerontology Research Centre and the Department of Psychology in Lund. The investigation included 69 people with an age-range from 55-94 years and examined if certain personality factors could explain the variance in memory functions in old adults. The participants’ personality factors were tested with Costa’s Big Five model of personality and were later compared to results on explicit and implicit memory tests. Results showed a relationship between specific personality factors like neuroticism, openness, agreeableness, and memory functions. A regression analysis showed how personality factors like openness and agreeableness were significant predictors for both memory and learning in old adults. Study II examined the importance of environmental, non-cognitive, and person-based markers for memory performance among centenarians (n=100). Selected as environmental, non-cognitive, and person-based variables were quality of life, life-habits, health, personality, and autonomy. These factors were later compared to memory performance in short-term memory, semantic memory, and episodic memory in centenarians. Using a structural equation model, results showed that performance in short-term memory was best predicted by life-habits and personality, while semantic and episodic memory performance were best predicted by autonomy and life-habits. Study III examined to what extent environmental, non-cognitive, and person-based variables may predict memory performance in 80-year-old individuals (n=181). The roles of these markers were examined by comparing them to various aspects of memory performance. As in study II, various aspects of quality of life, life-habits, health, personality, and autonomy were selected and later compared to performance in short-term memory, semantic memory and episodic memory. To support our hypothesis that these variables have predictive effects on memory processes, three structural equation models were used, one for each memory function. Results showed that performance in semantic memory and short-term memory were best predicted by life-habits and bad health, while performance in episodic memory was best predicted by bad health and personality. Life-habits and bad health were the two most substantial predictors for memory performance in 80-year old adults. In summary, the results from all three studies show a strong degree of association between changes in memory performance and changes in other latent environmental and person-based variables. Implementation of the results from a life-span perspective includes a closer monitoring of environmental markers in the future. In effect, this could preserve memory function and optimum health through old age, thus making interventions easier to realize.