TY - JOUR
T1 - Aging and late-life termina decline in perceptual speed
T2 - A comparison of alternative modeling approaches
AU - Thorvaldsson, Valgeir
AU - Hofer, Scott M.
AU - Johansson, Boo
PY - 2006/8/31
Y1 - 2006/8/31
N2 - Individual changes in perceptual speed were modeled as a conditional function of age and time-to-death. Alternative time-structured models were evaluated in a Swedish population-based, age-homogeneous sample (Gothenburg H70; N = 764) of individuals assessed at ages 70, 75, 79, 85, 88, 90, 92, 95, 97, and 99. Modeling time as proximity to death accounted better for the heterogeneity of individual changes than an age-based time structure. Time-to-death was a significant predictor of individual differences in rates of change in the age-based model but age did not significantly predict individual differences in rates of change in the model structured by proximity to death. In both the age-based and death-based time-structured models, accelerated changes prior to time of death were observed and provide support for the terminal-decline hypothesis. Identification of health-related factors and other sources of causal heterogeneity of aging-related change can make productive use of alternative time specifications that emphasize congruency between within-person change processes and aggregate population change.
AB - Individual changes in perceptual speed were modeled as a conditional function of age and time-to-death. Alternative time-structured models were evaluated in a Swedish population-based, age-homogeneous sample (Gothenburg H70; N = 764) of individuals assessed at ages 70, 75, 79, 85, 88, 90, 92, 95, 97, and 99. Modeling time as proximity to death accounted better for the heterogeneity of individual changes than an age-based time structure. Time-to-death was a significant predictor of individual differences in rates of change in the age-based model but age did not significantly predict individual differences in rates of change in the model structured by proximity to death. In both the age-based and death-based time-structured models, accelerated changes prior to time of death were observed and provide support for the terminal-decline hypothesis. Identification of health-related factors and other sources of causal heterogeneity of aging-related change can make productive use of alternative time specifications that emphasize congruency between within-person change processes and aggregate population change.
KW - Aging
KW - Cognition
KW - Longitudinal
KW - Perceptual speed
KW - Terminal decline
UR - http://www.scopus.com/inward/record.url?scp=33747825473&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=33747825473&partnerID=8YFLogxK
U2 - 10.1027/1016-9040.11.3.196
DO - 10.1027/1016-9040.11.3.196
M3 - Review article
AN - SCOPUS:33747825473
VL - 11
SP - 196
EP - 203
JO - European Psychologist
JF - European Psychologist
SN - 1016-9040
IS - 3
ER -