This dissertation's primary aim was to clarify the emergence of smoking-cognitive associations within the first half of the lifespan. In Study 1, using data from the first assessment of the Colorado Adoption/Twin Study of Lifespan behavioral development and cognitive aging (CATSLife1), smoking behaviors were examined cross-sectionally to assess their association with cognition before midlife. In Study 2 we explored how smoking behavior influences cognitive development. Prospective data were available for CATSLife1 participants from two archival longitudinal projects of twins, siblings and adoptees to track cognitive development from early adolescence to midlife. Associations between smoking and cognition were assessed using mixed effects growth models fitted to the cognitive data, with (a) smoking consumption at year 16 and (b) the smoking consumption difference score.
Study 1 showed that current smoking was associated with lower cognitive performance across most domains, including episodic memory, processing speed, spatial, and verbal, apart from working memory/attention indexed by Digit Span. Associations remained after adjusting for cardiovascular health, educational attainment, and a state-level tobacco control score (TCS). TCS was not related to cognitive performance and did not moderate associations with smoking behaviors. Sensitivity analyses with year 12 IQ suggested that smoking-cognition findings were not due to differences in early life intellectual ability.
Study 2 results indicated adolescent smoking, and to a lesser extent adulthood smoking, have a small negative effect on cognitive performance and change from adolescence up to midlife. Higher year 16 smoking consumption was associated with lower average performance for nearly all tasks. When examining differential cognitive growth rates, results indicated that a more rapid decline in adulthood was associated with greater levels of adolescent smoking for the episodic memory task Names and Faces. For processing speed, reduced growth at greater levels of adolescent smoking at age 16 and gains in smoking into adulthood were observed for Digit Symbol. Moreover, the smoking difference score was associated with average cognitive performance at age 16 for Colorado Perceptual Speed and Block Design, indexing processing speed, and spatial domains, respectively, whereby lower age 16 performance was associated with smoking gains after age 16.
Overall, results from Study 1 and Study 2 suggest that the influence of smoking behavior on cognition emerges early in the lifespan. This dissertation's collective findings help elucidate the developmental windows of risk on cognitive development, but further work remains.