verted U-or J-shaped relation between alcohol consumption and cognitive function (Dufouil et al., 1997; M.F. Elias et al., in press; P.K. Elias et al., in press; Launer et al., 1996). Drugs of abuse (e.g., opiates, cocaine) have been associated with poorer cognitive performance (Carlin and O'Malley, 1996; Strickland and Stein, 1995). In addition, several dietary insufficiencies, such as vitamin B6, vitamin B12, thiamine, folate, and zinc, have been related to cognitive difficulties (Lester and Fishbein, 1988; Riggs et al., 1996; Whitehouse et al., 1993). Greater caloric consumption in middle age has been shown to predict poorer mental status in old age (Fraser et al., 1996), and a proportionally greater intake of dietary refined carbohydrates has predicted lower IQ scores in children (Lester et al., 1982).
Health-enhancing behaviors have been associated with better cognitive functioning. For example, greater intake of vitamin C, an antioxidant, has been related to enhanced cognitive test performance and/or a lower prevalence of cognitive impairment (Gale et al., 1996; Jama et al., 1996; Paleologos et al., 1998). Greater levels of physical fitness (or physical activity) have also been associated with higher levels of cognitive functioning (Dustman et al., 1994). In addition, several investigations have revealed improvements in cognitive performance with aerobic exercise training (Emery and Blumenthal, 1991; Kramer et al., 1998).
Various hormonal factors have been associated with cognitive functioning. Again, direct biological effects on the brain are likely, in addition to indirect effects via promotion of systemic diseases. Relevant examples include poorer cognitive function in individuals with low levels of estrogen (Gordon et al., 1988; Erlanger et al., 1999), both high and low levels of various thyroid and pituitary hormones (Beckwith and Tucker, 1988; Gordon et al., 1988; Whitehouse et al., 1993; Erlanger et al., 1999), and either high basal levels of cortisol or greater stress-induced cortisol responses (Kirschbaum et al., 1996; Lupien and McEwen, 1997; McEwen and Sapolsky, 1995; Seeman et al., 1997; Erlanger et al., 1999). Beneficial effects of estrogen replacement therapy have also been noted (Haskell et al., 1997; Erlanger et al., 1999).
Genetic factors may predispose individuals to systemic diseases and influence numerous biological variables that can affect cognitive performance. One such factor that has received much recent attention is apolipoprotein E (APOE) polymorphism. Although most commonly examined in relation to dementias, the presence of one or two APOE-ε4 alleles has also been associated with poorer cognitive function, particularly on tests of learning, memory, and psychomotor speed, among nondemented individuals across a wide range of ages (Bondi et al., 1995; Carmelli et al., 1998; Flory et al., 1999; Yaffe et al., 1997). Genetic influences may also modify the impact of disease on cogni-