Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Smoking And Cognitive Functioning In Non-Demented Swedish Twins
(USC Thesis Other)
Smoking And Cognitive Functioning In Non-Demented Swedish Twins
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
INFORMATION TO USERS This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type o f computer printer. T he quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion. Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book. Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6” x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. UMI A Bell & Howell Information Company 300 North Zecb Road, Ann Arbor MI 48106-1346 USA 313/761-4700 800/521-0600 SMOKING AND COGNITIVE FUNCTIONING IN NON-DEMENTED SWEDISH TWINS By Julia Kasl-Godley A Thesis Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree Master of Arts (Psychology) May 1996 U M I Number: 1380470 Copyright 1996 by Kasl-Godley/ Julia E. All rights reserved. UMI Microform 1380470 Copyright 1996, by UMI Company. All rights reserved. This microform edition Is protected against unauthorized copying under Title 17, United States Code. UMI 300 North Zeeb Road Ann Arbor, MI 48103 UNIVERSITY O F SOUTHERN CALIFORNIA THE GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES. CALIFORNIA 0 0 0 0 7 This thesis, •written by JuIiavKasl-rCfjdley...................................... under the direction of her. Thesis Committee, and approved by all its members, has been pre sented to and accepted by the Dean of The Graduate School, in partial fulfillm ent of the requirements for the degree of J!das.ter..o£..Arts....____________________ L - U - V D tax May 1 0 , 1 9 9 6 ____ THESIS COMMITTEE O Chair&mJ ^ This work is dedicated with appreciation to Margaret Gatz, for her demand for critical inquiry, to Mark Godley, for his encouragement and support, and to Stanislav and Elizabeth Kasl, for providing a model of academic integrity, scholarship and dedication. T A B L E O F C O N T E N T S iii P A G E D E D I C A T I O N ii L I S T O F T A B L E S ;v A B S T R A C T v i I N T R O D U C T I O N 1 M E T H O D 12 R E S U L T S 2 2 D I S C U S S I O N 4 3 R E F E R E N C E S 4 9 LIST OF TABLES iv TABLE PAGE 1 . Pairwise and Singleton Sampling for SATSA and Present Sample 15 2. Sample Characteristics for Smokers and Non-Smokers 17 3. Raw Score Means and Standard Deviations for Dropouts and Continuing Participants 26 4. Test-retest Correlations 27 5. Results from Two Group Path Model 29 6a. Intercorrelations among Cognitive Measures at IPT1; Women Smokers 32 6b, Intercorrelations among Cognitive Measures at IPT2: Women Smokers 32 7a. Intcrcorrelations among Cognitive Measures at IPT1: Women Non-Smokers 32 7b. Intcrcorrelations among Cognitive Measures at IPT2: Women Non-Smokers 32 8a. Intcrcorrelations among Cognitive Measures at IPT1: Men Smokers 33 8b. Intercorrelations among Cognitive Measures at IPT2: Men Smokers 33 9a, Intcrcorrelations among Cognitive Measures at IPT1: Men Non-Smokers 33 9b. Intercorrclations among Cognitive Measures at IPT2: Men Non-Smokers 10. Correlations between IPT1 and IPT2 by Smoking Status and Sex 11. Raw Score Means and Standard Deviations on Cognitive Measures by Smoking Status and Sex 12. Standard Score Means and Standard Deviations on Cognitive Measures by Smoking Status and Sex 13. Concordance for Smoking Status among Co-Twin Pairs 14. Raw Mean Intrapair Difference and Standard Error for Female Discordant Pairs LIST OF FIGURE 1. Two group path model for independent group (continuous and dropout) regressions. ABSTRACT vi The present study addressed the question of whether non-impaired older adult smokers demonstrate better baseline cognitive performance than non-smokers and show less decrements in abilities expected to decline, normally, with age. Subjects were non demented, same-sex twin pairs fifty years of age and older who participated in the Swedish Adoption/Twin Study of Aging, a subsample of the population-based Swedish Twin Registry. Current smokers were defined as those who consistently reported being smokers during the time contemporaneous to and IS years prior to cognitive testing. Non-smokers were defined as those who consistently reported being non-smokers during this same interval. Five cognitive ability areas were assessed; attention, memory, perceptual speed/information processing, fluid abilities and acculturated knowledge. Individual level comparisons of current smokers to non-smokers indicated that changes in cognitive performance do not occur at different rates for smokers and non- smokers. However, smokers and non-smokers demonstrated mean level performance differences on specific cognitive ability tasks, a differential that was maintained over a three year interval. The direction and pattern of performance was contrary to expectation. Male smokers showed poorer spatial ability and perceptual speed/information processing than non-smokers and female smokers showed poorer spatial ability than non-smokers. Co-twin control analyses showed similar results. The study suggests that smoking may not attenuate non-pathological decline. These results contrast with previous literature showing smoking to be a protective factor for Alzheimer's disease. Julia Kasl-Godley vii Margaret Gatz, Ph.D. Smoking and Cognitive Functioning in Non-Demented Swedish Twins The present study tested whether smokers demonstrate better cognitive performance than non-smokers and show less decrements in abilities expected to decline, normally, with age. Subjects were noil-demented individuals, age fifty and older who participated in the Swedish Adoption/Twin Study of Aging. Smokers were defined as those who consistently reported being smokers during the time contemporaneous to and 15 years prior to cognitive testing. Non-smokers were defined as those who consistently reported being non-smokers during this same interval. Five cognitive ability areas were assessed: attention, memory, perceptual speed, fluid abilities and acculturated knowledge. Repeated measures ANOVAs indicated that changes in cognitive performance do not occur at different rates for smokers and non-smokers. However, ANOVAs showed mean level performance differences, with male smokers evidencing poorer spatial ability and perceptual speed than non-smokers and female smokers demonstrating poorer spatial ability than non-smokers. Co-twin control analyses showed similar results. The results suggest that smoking may not attenuate non-pathological decline which contrasts with previous literature showing smoking to be a protective factor for Alzheimer’s disease. 1 Healthy, cognitively intact older adult smokers have been found to perform poorer than non-smokers on measures of psychomotor speed (Hill, 1989), a finding which implies that smoking may compromise cognitive function perhaps through worsened pulmonary functioning or decreased cerebral blood flow (Yamashita, Kobayashi, Ynmaguchi, Kitani, & Tsunematsu, 1988). On the other hand, case-control studies suggest that smoking may have a neuroprotective effect on cognition, as suggested by the finding that smokers arc less likely than non-smokers to develop Alzheimer’s disease (Brenner et al., 1993; van Duijn & Hofinan, 1991; Hofinan & van Duijn, 1990; Ferini-Strambi, Smime, Garancini, Pinto, & Franceschi, 1990; Grossberg, Nakra, Woodward, & Russel, 1989), a disease marked by progressive, pervasive impairment of memory and other cognitive abilities, such as attention, reasoning and judgment. Nicotine, the active pharmacological agent in cigarettes, has been hypothesized to account for this apparent neuroprotective effect, although selection bias and selective survival have been offered as competing explanations (Riggs, 1993). Given these alternative explanations and the inconsistent findings about the effect of smoking on normal and pathological cognitive function in older adults, nicotine’s effect on cognitive functioning merits further exploration. If mechanisms exist by which nicotine use over time could counteract the process of cognitive impairment seen in Alzheimer’s disease, then it also can be asked whether nicotine could be protective against non- pathological, age-associated declines in cognitive functioning. One way to test this question is to examine the performance of older, non-impaired individuals with a history of smoking. Inferences about nicotine’s effects arc extrapolated from smokers because examination of the effects of long-term nicotine use on cognitive performance through direct administration, via injection, pill or gum, would be difficult to perform with humans. If nicotine protects against non-pathological cognitive decline, then smokers should demonstrate better cognitive performance than non-smokers and should show less decline than non-smokers in abilities expected to decline, normally, with age. The present study evaluated this hypothesis using a population of older adult Swedish twins. It extended the existing research on the relationship between normal cognitive function and smoking in older adults (Hill study) in five ways: a) use of individuals with a prospectively documented history of smoking, b) use of a three year follow-up period, c) exploration of possible sex differences through stratification by sex d) examination of selection bias and e) incorporation of a matched-twin case control design. A matched twin design enables additional control of variables that could influence the association between smoking and cognitive function, e.g., whether smokers who survive have hardier genes and therefore reduced genetic susceptibility to cognitive decline than non-smokers. This latter issue can be controlled for by examining twins who are discordant for smoking status, that is, when one twin smokes and the other does not, given that twins share genetic make-up (either fifty or one-hundred percent depending on zygosity). Mechanisms The hypotheses for this study were based on two putative mechanisms by which nicotine could be protective against non-pathological, age-associated declines in cognitive performance: enhanced attention and information processing and increased cholinergic system efficiency. Enhanced attention and information processing Hie first mechanism by which nicotine could influence cognitive performance is through its action as a stimulant. Stimulants tend to heighten central nervous system activity, thereby enhancing mental clarity, increasing vigilance and allaying fatigue all of which increase the ability to attend (Wenk, 1994), As proposed by Luria (1973), arousal not only permits alertness to incoming information but influences registration of information, the subsequent integration, analysis and organization of information and expression of a response. Tlius, increased attention influences the efficiency with which information is organized and encoded (Rusted & Warburton, 1992; Warburton, Rusted & Fowler, 1992). Increased attention may have secondary effects on memory function through more efficient encoding and retrieval. Enhanced attention reduces interference of off-task information which clutters working memory (Dempster, 1994; Ynntis, 1992) and impinges information processing. Furthermore, information that is better organized is better retrieved fiom memory (Horn et. al., 1981; Horn, 1991). Thus, enhanced attention via nicotine’s stimulating effects on CNS function could facilitate cognitive performance, particularly attention, information processing and working memory. Tentative evidence for this assertion comes from findings that speed and accuracy of information processing and short term memory improve when smokers are given nicotine (see Warburton, 1992b; Wesnes & Warburton, 1984). These changes appear to persist with sustained nicotine use (see Levin, 1992 for review). For example, in a study of 13 smokers who were tested on three occasions over a three week period, subjects were administered nicotine or placebo gum every hour for three hours. Improvements in psychomotor performance followed the repeated doses of nicotine (Sherwood, Kerr & Hindmarch, 1992). Nicotine’s enhancing effect appears to be circumscribed, with the degree of improvement in performance inversely mediated by dosage level, that is, individuals smoking lower level nicotine cigarettes tending to show greater improvement than medium or high nicotine containing cigarettes (Colrain et al., 1992; Hindmarch, Kerr & Sherwood, 1990). This dose-response pattern may occur if nicotine is too arousing at higher levels because it would interfere with performance by inhibiting the ability to attend. This hypothesis is consistent with the finding of a study of 20 young adult smokers that word list recall was enhanced after post-learning nicotine administration hut was not maintained when a distracter task was introduced during the post-trial period in which subjects smoked (Rusted and Warburton, 1992). The authors propose that nicotine improves performance when it can increase the attentional resources allotted to the current task. In summary, nicotine may improve cognitive performance through enhanced attention. Enhanced attention may have secondary effects on information processing and memory function which could impact other higher order abilities. Increased cholinergic system efficiency Another mechanism by which nicotine may enhance cognitive performance is through nicotine-induced changes in cholinergic activity (Benwell, Balfour & Anderson, 1988), a neurotransmitter system which is believed to be involved in learning and memory. Nicotine has been found to stimulate the release of acetylcholine, a prcsynaptic cholinergic marker (Americ, 1991; Johnston & Silverstein, 1992) which regulates the receptor activity in localized regions of the cerebral cortex and the hippocampus (Americ, 1991). This effect on synaptic receptor activity may promote information processing, given that most neuronal communication and thus, information processing, occurs at synaptic junctions (Ivy, MacLeod, Petit & Markus, 1992). If nicotine enhances the efficacy with which neurons communicate by stimulating receptor activity, it is likely to improve certain aspects of cognitive function. Furthermore, if receptor efficiency can be maintained, then it is possible that age-related declines in cognitive performance could be attenuated. In summary, nicotine may help maintain cholinergic receptor efficiency which could attenuate age-associated declines in cognitive performance. Although this hypothesis is plausible, it has been examined only in individuals with Alzheimer’s Disease who may be too far along in the progression of the disease to benefit from nicotine’s putative neuroprotective effects on cognitive performance. Interaction o f the two mechanisms o f action The degree to which nicotine may alter the trajectory of decline in cognitive performance is likely to be a function of the degree to which the cognitive ability being measured is dependent on biologically based processes, such as speed of information processing (Brody, 1992), efficiency of inhibitory processes (Dempster, 1992; Kliegl & Lindenberger, 1993) and efficiency of working memory (Lindenberger & Baltes, 1994; Baddely, 1992), given that these processes decline with age and are influenced by CNS function. Abilities thought to be most vulnerable are attention, working memory, processing speed and visual processing, and to a lesser extent, non-verbal reasoning (Horn, 1994). If nicotine can affect CNS function through enhanced arousal and cholinergic system efficiency, it is possible that the trajectory of decline in cognitive abilities most dependent on CNS function could be altered. However, as mentioned earlier, little is know about effects of long-term direct administration of nicotine in humans and thus, inferences are made from comparisons of smokers and non-smokers. In order to explicate more fully, the way in which smoking may interact with age differences in abilities widely viewed as dependent on biologically based processes, a more expanded discussion of the pattern of normal age-associated changes in cognitive function may be helpful. Therefore, a brief review of age-associated changes in specif c cognitive abilities is provided. Patterns of age-associated changes in specific cognitive abilities Attention Attention can be defined as a cognitive mechanism that selects relevant perceptual information from the environment, maintains the information for processing until it is delivered to working memory and controls performance of concurrent cognitive operations in support of an individual’s goals (Yantis, 1992; Hartley, 1992). Age differences in attention are noted, particularly in the ability to focus, divide or maintain attention in the presence of irrelevant stimuli and in the ability to select information relevant to the task at hand. With aging, there is an increased susceptibility to interference, that is, in the ability to inhibit or suppress irrelevant information (Davies, Jones & Taylor, 1984). Inhibitory processes, as mentioned earlier, are believed to be biologically based. 7 Memory Memory can be viewed as successive stages of processing in which information is held in conscious awareness within a specific sensory modality, is selected by attentional processes and passed on to short-term memory where it is integrated with previously learned information and enters conscious awareness. Working memory directs the flow of information as it is rehearsed and learned, passing from primary or sensory memory to secondary memory, and is transferred to long-term memory where it is dropped from conscious awareness. Working memory encodes and maintains verbal and perceptual information while simultaneously accessing long term memory (Craik & Jennings, 1992) to solve problems, make decisions and leant new concepts (Engle, 1994; Craik & Jennings, 1992). Age differences in primary memory are slight or non-existent but differences increase as tasks involve a larger working memory component (Craik & Jennings, 1992). Age differences in working memory are believed to occur in both storage and processing of information (Babcock & Salthouse, 1990). Changes may occur in the types of operations carried out on events to be remembered, in the number of representations able to be maintained in an active state, in the ability to maintain processing while attending to new information, in the extent to which context influences specificity of encoding and in the ability to recapitulate original processing operations at the time of retrieval (Craik & Jennings, 1992). As mentioned earlier, working memory is believed to be strongly linked to biological processes and therefore, is particularly susceptible to changes in CNS function. 8 Changes in working memory are likely to contribute to age differences in a number of higher order abilities, such as non-verbal reasoning or problem solving. These differences are more pronounced when the demands on processing are increased (Mayr & Kliegl, 1993; Craik & Jennings, 1992). For example, if a reasoning task requires that old information be used to interpret new information, a diminished capacity to maintain information in working memory would affect ability to perform the task. Thus, limits on the amount of information that can be maintained in working memory can place constraints on higher order abilities dependent on those functions (Hasher & Zachs, 1988), such as fluid abilities. Fluid abilities Fluid abilities, which entail the capacity for reasoning and problem solving, include abstract thinking, non-verbal reasoning and spatial ability. Abstraction, however, may be a less pure fluid ability given that it is more culturally linked (Kyllonen, 1994). All three abilities, reasoning and spatial ability in particular, involve a large working memory component and therefore can be expected to evidence age-associated changes. The literature on fluid abilities supports this assertion. Abstraction can be defined as the processes by which individuals separate objects from their immediate context and categorize things into higher order classes, a necessary prerequisite for seeing relations among disparate objects (Kyllonen, 1994). Abstraction, as measured by tasks requiring individuals to find the similarity or category membership of increasingly dissimilar objects or to compute the similarity of the relationship, tends to decline with age (Salthouse, 1992). However, these tasks have a large verbal component in that they tend to be somewhat specific in the knowledge they tap and require verbalization of the relationships between objects. Reasoning ability, an intermediate level of processing between memory and complex thinking, involves the transformation or manipulation of information that is being maintained in working memory. Reasoning shows greater declines when assessed using tasks not dependent on the amount or type of knowledge possessed such as measures of syllogistic reasoning or other tasks of integrative reasoning, series completion and concept identification (Salthouse, 1992). Overall, age differences in reasoning ability indicate small decrements up through age 60 and more rapid declines afterwards, though the extent of declines varies across individuals, with some showing no decline through age 70 (Kyllonen, 1994). Spatial ability refers to the ability to generate, retain, retrieve and transform visual images (Lohman, 1994). Age appears to be associated with slower rotation processes although it is difficult to isolate other component processes involved in spatial tasks. Attempts to analyze component processes of a visuo-construction task, i.e., block design, have found age differences at the level of segmentation of the target pattern into cells corresponding to the individual blocks and manipulation of the block to match the pattern on the target cell of the to-be-reproduced pattern (Royer, Gilmore & Gruhn, 1984; Salthouse, 1987). Overall, age has been associated with changes in fluid abilities; namely, with a decreased ability to preserve relevant information from earlier premises during presentation of later information, a deterioration in the integration of information and processes concerned with detecting spatial patterns and abstract relations, and a reduction 10 in the rapid generation of alternative formulations or conceptualizations. Declines may be due to changes in the ability to generate and retain mental representations, in the amount of visual-spatial information that can be maintained in an active state, in the flexibility of available problem-solving strategies and in the speed of performing transformations (Salthouse, 1992), each of which implicate working memory function. The larger demands placed on working memory by tasks requiring manipulation of non-verbal information may make them vulnerable to age-associated changes in biologically based processes (Salthouse & Skovronek, 1992). A cculturated knowledge In contrast to age-associated changes in attention, memory and fluid abilities, acculturated knowledge, that is, the acquisition of practical experience and accumulated knowledge, has been found to remain stable or to improve over much of late adulthood (Horn, 1976). This relative resiliency is likely to be due to diminished sensitivity to central nervous system damage because of overdetermination, i.e., when information is based in more highly interconnected networks that contain many possible ways to access the information. In addition, acculturated knowledge is not very dependent on working memory. In summary, there appear to be multiple age gradients in cognitive abilities, with some abilities remaining stable and others deteriorating at various rates (Lindenberger & Bakes, 1994). This variability is believed to be a function of the degree to which abilities are dependent on biological processes (Brody, 1992) like information processing, and therefore are vulnerable to changes in central nervous system function, with those abilities most vulnerable showing the largest age-related decline. However, this same vulnerability may make these abilities most malleable by substances that enhance central nervous system functioning, i.e., nicotine, such that their trajectory of decline is altered. Tlius, those who use nicotine, as measured by tobacco smoking, may be expected to show different patterns of age-associated changes in cognitive function than those who do not smoke. The present study examined whether smokers demonstrate better cognitive performance than non-smokers and whether smokers show less decline than non-smokers in abilities expected to decline, normally, with age. Both group comparisons and discordant twin pair comparisons were conducted. Specific hypotheses were made about which abilities would be the most likely to show baseline differences and different patterns of change between smokers and non-smokers. Cognitive abilities were grouped into five areas: attention, perceptual speed/information processing, memory, fluid abilities and acculturated knowledge. It was hypothesized that smokers would show better baseline performance and less decline than non-smokers 011 tasks measuring those cognitive abilities most closely linked to nicotine's mechanism of action, that is improved attention, information processing and memory through enhanced arousal and cholinergic system activity. On the other hand, smokers and non-smokers were expected to show similar baseline performance on tasks measuring abilities relatively independent of arousal and cholinergic system activity. Specifically, 1) Smokers would show better attention, perceptual speed/information processing, memory and fluid abilities than non-smokers, a mean level difference that would be maintained at the three year follow-up. 2) Attention, perceptual speed/information processing, memory and fluid abilities would sliow age-associated declines in non-smokers but would decline at a slower rate or remain stable in smokers. 3) Smokers and non-smokers would show no mean level difference in performance in acculturated knowledge, a similarity that would remain across time. 4) Acculturated knowledge would improve or remain stable for both smokers and non- smokers. Hypothesis 1 and 2 were also examined in twins. Specifically, 5) Among twin pairs discordant for smoking, smokers would demonstrate better cognitive functioning than their non-smoking partners, particularly with respect to attention, perceptual speed/information processing, memory and fluid abilities. METHOD Study Population Participants were from the Swedish Adoption/Twin Study of Aging (SATSA), a sample o f same sex twin pairs reared apart and a matched sample of twins reared together drawn from the larger population-based Swedish Twin Registry (Cederlof & Lorich, 1978). As described in Cederlof and Lorich (1978) and Medlund, Cederlof Floderus- Myrhed, Friberg, and Sorensen (1976), the Swedish Twin Registry represents a population-based sample of over 25,000 same-sex twin pairs. There are two cohorts: the "old" cohort, which includes twin pairs bom from 1886 to 1925 of whom both members were alive in 1961 and, the "new" cohort, which includes all twin pairs bom from 1926 to 1958 of whom both members of the pair were alive in 1972. In 1963 and then again in 1967 or 1970, questionnaires were sent to the old cohort asking about health and 13 occupational history, rearing environment, diet, exercise, stress, and smoking and alcohol consumption. A similar questionnaire asking about smoking, alcohol consumption, health, diet, exercise, occupation and personality was sent out in 1973 to the new cohort. As described in Pedersen et al. (1991), the SATSA subsample was created in 1984 and was comprised of all pairs of twins in the registry who had been separated before the age of 11 and reared apart (TRA) and a sample of twins reared together (TRT) matched on gender and date and county of birth (see Pedersen, Friberg, Floderus-Myrhed, McCleam & Plomin, 1984). Both members of 591 TRA and 627 TRT pairs were alive and, one member of 221 TRA pairs and of 197 TRT pairs was alive, yielding a total of 2,854 individuals. Age at separation was highly skewed, with fifty-two percent of TRA twins separated before their first birthday, 69% by their second birthday and 82% by age 5 (Pedersen, Plomin, Nesselroade, &McCIeam, 1992). In October 1984, a two-part questionnaire assessing physical and mental health, personality, activities of daily living, health-related behaviors such as smoking and alcohol use, early rearing environment, zygosity and occupational, familial and social environments was sent out. 2,018 individuals (including 758 complete pairs) responded to the first part of the questionnaire (Ql-Red) yielding a response rate of 71%. These individuals received the second part (Ql-Blue) of the questionnaire about one week later. The first follow-up began in October 1987, when a second questionnaire (Q2) was mailed to all surviving individuals of the original SATSA subsample who had not declined further participation (2,581 individuals). A total of 1,637 individuals responded (548 complete pairs), 1,544 of whom had answered Ql. One hundred and five individuals had died between Q1 and Q2. 14 From November 1985 to November 1988, all but 12 of the 548 pairs aged 50 and older of whom both members had responded to Q 1-Red, were contacted to participate in an in-person testing evaluation (IPT1) which involved an interview, cognitive testing and a health assessment. One or both members of 201 pairs were lost either to death (68 pairs), illness (7 pairs), blindness (7 pairs) or refusal to participate (119 pairs). In another 32 pairs, only one member was tested. 291 pairs aged 50 or older and 12 pairs under age 50 (645 individuals, 606 of whom were from complete pairs) participated in the in-person testing (IPT1). Participants were tested individually in one session that lasted an average of four hours. A second in-person testing (IPT2) occurred from January 1989 to November 1991 (593 individuals, 536 of whom were complete pairs), which included 93 new individuals who had reached age 50 (90 of whom were from complete pairs). A total of 502 individuals (448 of whom were from complete pairs) participated in both IPT1 and IPT2. For the present study, the eligible sample consisted of those non-impaired individuals aged 50 and over who had participated in both of the SATSA in-person cognitive assessments (IPT1 and IPT2). From the available 502 individuals who participated, 30 were excluded because they were under age 50 (21 individuals, 16 of whom were complete pairs) or demented (9 individuals, including one complete pair.) Another pair was dropped because of undetermined zygosity. Pairwise breakdowns of TRT and TRA by zygosity for the SATSA sample and the sample used in the present study are reported in Table 1 . Zygosity was first determined on the basis of physical similarity and was confirmed by serological analyses. 15 Table 1 Pairwise end Singleton Sampling for SATSA and Present Sample SATSA TRA MZ DZ Intact pairs IB L MZ OZ TRA MZ Singletons DZ I B I MZ DZ Q1-Red 99 233 166 221 62 188 53 124 Q2 72 178 127 155 72 197 66 131 IPT1 46 100 67 89 9 16 6 6 IPT2 37 93 61 76 10 18 10 18 Present Sample 7 34 20 25 15 35 25 31 Smoking groups The sample used for analyses included all smokers (n=48) and non-smokers (n=230) in the eligible sample of 470 individuals. Smokers were defined as those who consistently reported being smokers in the questionnaires from the original twin registry (old or new cohort) and in both SATSA surveys (Ql-Red and Q2). Non-smokers were defined as those who consistently reported not being smokers in the registry and in both SATSA surveys. The remaining individuals (n=191) were classified as former smokers or had missing smoking data and thus, were excluded from analyses. Exclusion of former smokers from the analyses may be overly restrictive but it was decided that inclusion of only those individuals who consistently reported being smokers would most clearly address the question of whether smoking has an effect on cognitive functioning. However, dropping the former smokers may have introduced selection biases. This possibility was examined by creating a dichotomous variable which represented those in the sample used for analyses (selected sample) and those who were dropped (unselectcd sample). There were significantly more women represented among the selected sample than among the unselected sample (71.1% vs. 28.9%) and slightly more men among the unselected than selected sample (41.6% and 58.4%, respectively). The samples did not differ on educational level. At Q 1-Red, there were 110 differences between the samples in the number of illnesses (as measured by indices for respiratory, metabolic or cardiovascular disorders), in the incidence of coronary heart disease or cancer or on self-reports of whether their health had declined over the last five years. However, at Q2, significantly fewer of the selected sample had cardiovascular disorders (38.7%) compared to those in the unselected sample (50.0%). A non-significant trend in the opposite direction was obtained for the incidence of cancer (4.40% vs. 1.2%). Taken together, the selected sample were no more educated than the unselected sample, had fewer cardiovascular disorders but a slightly higher incidence of cancer and tended to rate their health as better. These findings may reflect that possibility that former smokers may have quit smoking because of health concerns and suggest that smokers who continue smoking may be healthier than those who quit. This interpretation may pertain to the men in particular, as indicated by the disproportionate number of men among the smokers and non-smokers in comparison to the general SATSA sample. Whereas men represent, on average, 40% of the SATSA sample at each questionnaire and assessment point, they made up over half (56%) of the smokers but only one-fifth (22%) of the non-smokers in the present sample. This variability suggests that male smokers may be a hardier group than the non-smokers. Characteristics o f the selected sample Within sex comparisons of smokers and non-smokers were made on potential confounding variables, i.e., factors thought to influence cognitive functioning: namely, age, educational attainment, physical health status and heavy alcohol consumption. Heavy consumption was defined as drinking more than 5 bottles of beer, 1 bottle of wine or 1/2 bottle of liquor at one occasion. A seven point Likert scale was used to assess frequency: 17 0 indicated never, 1 indicated 1-3 times a year, 2 indicated 4-6 times a year, 3 indicated once a month, 4 indicated a couple times a month, 5 indicated once a week, 6 indicated a couple times a week and 7 indicated nearly every day. Hormone replacement in women also was examined, given the possibility that estrogen may be protective against cognitive decline (Henderson, Paganini-Hill, Emanuel, Dunn & Buckwalter, 1994). Between sex comparisons also were conducted within the smoking group. Demographic and descriptive information for smokers and non-smokers, shown separately for men and women, is summarized in Table 2. Table 2 Samp/e Characteristics for Smokers and Non-Smokers Women Men Smokers Non-Smokers Smokers Non-Smokers Variable (n=21) (n=179) <n=27) (n=51) Mean age at 01-Red (SD) 60(6.7) 64 (8.0) 62 (5.0) 62 (5.6) Education mean (SD) 1.3 (0.5) 1.4 (0.7) 1.8 (1.1) 1.8 (1.1) No. of physical Illness Q1-Red mean (SD) 2.3 (1.5) 2.1 (1.7) 1.6 (1.0) 1.4 (1.2) Q 2m ean (SD) 3.2 (1.9) 3.0 (2.2) 2 2 (26) 1.9 (1.5) Heavy alcohol use Q1-Red mean (SD) 0.3 (1.1) 0.0 (0.2) 0.8 (1.3) 0.3 (1.0) Q 2m ean (SD) 0.3 (1.0) 0.0(0.11 0.6(1.2) 0.3 (0.8) Mata Educdlen It pros anted on a icola tam 1 (damtmtay tcfiool) to 4 (un/VWJIy or hlghd); d c c M u jsu poiartod a t a tc d a trom 0 (nnrtr) to 7 fnaor V wary ttay) Men Smokers and non-smokers did not differ, significantly, on age, educational attainment, self-rated health or in the total number and type of physical illnesses (at either Q 1 or Q2). The types of physical illnesses examined were: heart disease and other cardiovascular problems (angina pectoris, heart infarct, thrombosis, heart insufficiency), stroke, cancer, respiratory disorders (chronic bronchitis, tuberculosis, emphysema, asthma, prolonged cough), gastro-intcstinal disorders (ulcer, stomach or intestinal disorder, gall bladder problems, liver disease), urologic disorders (kidney disease, urinary or prostrate problems) and disorders of the metabolic type (diabetes, goiter, gout, anemia). Smokers and non- 18 smokers differed on self-reported alcohol consumption, with 100% of the smokers at Q1 reporting that they drink compared to 72% of the non-smokers X2 (2, N = 77) = 9.24, p < .002. However, the low frequency of heavy alcohol consumption makes this difference of small concent with respect to harmful effects on cognitive function. 85.7% of the non- smokers never drank heavily, 10.2% drank heavily only 1-3 times a year or less and the remaining 4.1% drank heavily about once a week. 63% of the smokers never drank heavily, 22.2% drank heavily between 1 and 6 times a year and the remaining 14.8% drank heavily once a month to a couple times a month. These proportions remained roughly similar at Q2, with an overall movement towards even less heavy drinking. Women Smokers and non-smokers did not differ, significantly, on age, educational attainment, self-rated health or in the total number of physical illnesses at either Q] or Q2. However, they differed significantly on two types of physical illnesses at Q I— intestinal disorders X2 (I, N = 195) = 7.8, p < .005 and metabolic disorders X2 (I, N = 199 ) = 3.78, p < .05, the latter of which can affect cognitive performance if left untreated. Smokers had more intestinal problems but less metabolic disorders than non-smokers, although only the difference in intestinal disorders X" (1, N = 187 ) = 7.46, p < .006, was maintained at Q2, suggesting that smokers arc catching up to non-smokers. Alcohol consumption among women was extremely low, with 23.8% of smokers and 38.1% of non-smokers being complete abstainers. O f those who drank, 96.5% of the non-smokers and 90.4% of the smokers never drank heavily. Smokers and non-smokers showed no differences on hormone intake. Between sex comparisons indicated that men and women differed in the duration of smoking, as well as the age at which an individual started smoking. Men started smoking at an earlier age, with an average age of 18 (5.0) in comparison to a average age of 28 (9.7) for women and, they smoked longer, with an average of 42.7 (7.2) years in comparison with 31 (7.6) years for women. A difference in average daily cigarette consumption also was observed, with men smoking an average of 17.4 (10.2) cigarettes and women smoking an average of 12.4 (5.6) cigarettes. Measures Smoking Smoking data were available in the registry (old and new cohort) and SATSA (Ql- Red and Q2) surveys, collected from the subjects when they were aged 16 through 60, and then 9-16 years later at the first SATSA survey (Ql). (The exact time interval varied depending on whether an individual was from the old or new registry cohort). Smoking data were collected again three years later (Q2). Information about lifetime prevalence (ever/never smoked), duration of smoking, current smoking status and consumption, type of tobacco product used (cigar, cigarette, pipe) and brand o f cigarette was gathered. Information on current smoking status .(non-smoker or current smoker) was used as the basis for creating smoking groups. Cognitive functioning Cognitive data were collected on two occasions across a three-year interval. The cognitive battery consisted of tests designed to assess general and specific cognitive abilities. The measures were given in a standard order of administration which was Folstein Mini-Mental State Examination, Information, Digit Span, Figure Logic, Figure 20 Identification, Names and Faces-Immediate, [Grip strength test], Names and Faces- Delayed, Synonyms, Picture Memory, Block Design, [coffee break], Analogies, Symbol- Digit and Card Rotations. A description of the 11 tests used for analyses and the cognitive ability for which each is designed to measure (attention, perceptual speed/information processing, memory, fluid ability or acculturatcd knowledge) follows. The Folstein Mini-Mental State Examination was not used in the present study. Information (CVB-modified WAIS, Jonsson & Molander, 1964). Participants are asked to respond to questions that tap general knowledge. This test measures acculturated knowledge. Synonyms (Dureman-Salde Batteiy, Dureman, Kebbon & Osterberg, 1971). Participants are given a word and asked to choose from a list of five alternatives which one has the meaning most similar to the target. This test measures acculturated knowledge, although it requires some conceptual reasoning. Analogies (W1T-1II, Westrin, 1969). Participants are asked to identify the logical relationships between words. This test measures reasoning ability but is thought to have n moderate verbal component. Figtire Logic (Dureman-Salde Batteiy, Dureman, Kebbon & Osterberg, 1971). Participants are asked to identify from a list of five figures the one that does not belong. This test measures spatial reasoning. Kohs Block Design (Point Scale of Performance, Arthur, 1947). Participants are given blocks and then, are presented a picture of a design. They are required to assemble the blocks such that their finished product matches the picture. This test measures visuo- construction and spatial ability. Card Rotations (Educational Testing Service Reference Kit, Ekstrom, French & Hannan, 1976), Participants are asked to compare a test figure to a sample figure and to decide whether or not the test figure has been rotated or reversed. This test measures spatial ability. Symbol-Digit (CVB-modified WAIS, Jonsson & Molander, 1964). Participants are shown a printed key that pairs numbers with symbols. Then, they are given a matrix of boxes that have symbols in the top portion and blank spaces in the bottom portion. Participants are asked to use the key as a guide to fill in the blank spaces under the symbols with the appropriate number. This test measures perceptual speed and attention. Figttre Identification (Dureman-Salde Batteiy, Dureman, Kebbon & Osterberg, 1971). Participants are shown a picture and then are asked to decide which one of five figures matches the target figure. This test measures perceptual speed. Digit Span (CVB-modified WAIS, Jonsson & Molander, 1964). Participants are asked to listen to a series of numbers and then to repeat the numbers in forward (Digit Span Forward) and backward (Digit Span Backward) sequences. Digit Span Forward measures attention and to a lesser extent, memory. Digit Span Backwards measures working memory. Thurstone’ s Picture Memory (Thurstone's Primary Mental Abilities, Thurstone, 1938). Participants are shown several pictures. Then, they are shown a line of pictures in which the target is embedded and are asked to identify, respectively, the target picture. This test measures visual memory. Names and Faces (Colorado Adoption Project, DeFries, Plomin, Vandenberg & Kuse, 1981). Participants are shown several faces with assigned names and then, when shown 22 the faces alone, are asked to recall the names. Names and Faces-lmmediate measures immediate visual memory and Names and Faces-Delayed measures delayed visual memory. RESULTS Attrition Before analyses examining cognitive functioning among smokers and non-smokers were performed, the possibility of selective, or non-random attrition, was investigated. Selective attrition refers to the fact that individuals who remnin in a sample are likely to be different from those who drop out, i.e., they may evidence better cognitive functioning then dropouts and thus, if analyses arc performed on only those left in the sample, conclusions are likely to be biased toward high predictive continuity, thereby providing underestimates of change or variability. This possibility could minimize the likelihood of finding differences in rate o f cognitive change between non-smokers and smokers. To examine how attrition may influence measurements o f cognitive change, two groups of subjects were created: dropouts and continuing participants. Dropouts were those individuals who participated in the first cognitive testing occasion (IPTI) but did not participate in the second occasion (IPT2). There were 143 dropouts, but 17 were excluded because they were demented (n=9) or under the age of 50 (n=8), Of the remaining 126 individuals (38 men and 88 women), 116 had complete data, i.e., information on every test, and were used for analyses. (Complete data are necessary because analyses used structural equation modeling techniques which require a full correlation matrix.) Continuing participants were those who participated iu both IPTI and IPT2 (28 had died). There were 500 continuing participants, 30 of whom were excluded 23 because they were demented or under the age of 50, Of the remaining 470 people (190 men and 280 women), 367 had complete data and were used for analyses. Because attrition analyses were performed only on those with complete data, the sample used for analyses may still be somewhat selective. This rigorous definition of attrition would overestimate the effects of attrition. Inclusion of dropouts in data analyses allows for the examination of the potential threats of selective attrition on measures of cognitive performance obtained at IPT2. If attrition is selective, one would expect that measures based on the continuing participants scores would be different from those obtained from estimates based on the entire sample at the first time of measurement (inclusion of those who later dropped out at IPT2). Although estimates can be obtained by simple linear regression, structural equation modeling techniques were used because they make no assumption of equal within cell variance (as in ANOVA) or equal error terms. Thus, these models permit a test of the equality of the entire regression function, i.e. whether or not the same regression line can be used to characterize the data when the data include and omit the dropouts, while allowing the spread around the regression line to vary between the groups. A two group path model carried out this determination by constraining the regression parameters for each respective cognitive ability measure to be equal for the continuing participants and the dropouts. The estimates reflect the regression function that would have been obtained had retest information been available for everyone. LISREL 8 (Joreskog & Sorbom, 1993) was used to obtain estimates by using the dropout group’s observed scores at IPTI and latent variables for scores at IPT2. In doing so, the path between the first and second time of measurement can be constrained to be 24 invariant— to have an equivalent relationship across groups. If this invariant model fits the data, then it suggests that the same regression function describes both groups, which would indicate that the EPT2 cognitive performance scores obtained from the continuing participants are not different from those that would have been obtained if the dropouts had remained in the sample. LISREL 8 obtained estimates by using maximum likelihood procedures to get parameters that best reproduce the data based 011 a function involving the difference between an observed matrix of product moments and the expected or estimated matrix derived from the model parameters. Moments matrices, a combination of summaiy statistics that include the means, standard deviations and correlations, were used for this study because the models place joint constraints on summary statistics. The likelihood function is distributed as a Chi-square value with the degrees o f freedom equal to the number of observed statistics minus the number of estimated parameters. Hie Chi-square provides the relative magnitude of the difference between the original data matrix and the estimated moment matrix. Each model constraint generated one degree of freedom for testing the fit of the constrained model (against a less restricted alternative model, in this case, a model of independent groups). Model parameters are presented in unstandardized form because they are easier to interpret. For example, standardized results may distort between group comparisons if standardization is done within groups. An index of goodness of fit is calculated by dividing the Chi-square by the degrees of freedom: generally, if the Chi-square value is more than I 1/2 to 2 times larger than the degrees o f freedom, the model does not fit well. However, because this criterion is a rough indicator of goodness o f fit, it was supplemented by another fit index, the Akaike’s Information Criterion (AIC: Akaike, 1987). The AIC takes both the Chi-square and the degrees of freedom into account, thereby providing a measure of model fit and parsimony. Furthermore, the AIC’s of various models can be compared (given that they have the same degrees of freedom), thereby providing a relative index of fit. This comparison is permissible because the Chi-square is not a function of the scale of measurement. The mean raw scores and the standard deviations for all cognitive measures at the two occasions of measurement are shown separately by attrition group in Table 3. 26 TABLE 3______________________________________________________________________________ Raw Sccre Means and Standard Deviations for Dropouts and Continuing Participants _____________________________________________________________________________ GROUPS MEASURE DROP OUTS (N-116) CONTINUING PARTICIPANTS (n=367) ___________________________________________ IPTI_____________IPT2___________________________________IPTI_____________IP72 Information M 30.1S 32.01 32.67 SD (0.59) (7.61) (7.24) Synonym* M 16.29 19.13 1920 SD (8.02) (5.63) (5.39) Analogic* M 1328 14.74 14.92 SD (3.74) (3.90) (3 67) F lgue Logic M 16.90 1020 17.91 SD (4.19) (3.70) (3.96) Block Deilgn M 16.59 __ 19.72 19.04 SD (7.48) (7.40) (6.74) Card Rotation* M 43.31 5125 50.26 SD (16.33) (19.15) (16.31) Digit Span Forward M 5.47 5.71 5.60 SD (1.22) (1.16) (1.18) Digit Span Backward M 3.97 4,19 4.14 SD (1.08) (1.20) (1.24) T huttona Picture Memory M 20.09 21.03 2067 SD (4-22) (4.54) (4.53) Name* and Facea-lmmedate M 1.70 — 1.06 2.20 SD (1.84) (1.68) (1.83) Name* and FacevD dayed M 1.25 2.06 1.74 SD (1.46) (1.64) (1.78) Digit Symbol M 34.86 __ 40.62 3722 SD (11.91) (10.65) (10.75) Flgire Identification M 26.90 3029 29.78 SD (7.69) (7.16) (7.78) The raw score means indicate that, on average, the dropout group had slightly lower scores at IPTI than the continuing participants. Measures of spatial ability and perceptual speed/information processing showed the largest differences. Test-retest reliabilites reflecting the observed correlations in the continuing group and the estimated correlations from inclusion of the dropout group (based 011 assuming the same regression function for both groups) are shown in Table 4, 27 Table 4 Test-retest Correlations GROUPS MEASURE OBSERVED: DROPOUTS OBSERVED: CONTINUING PARTICIPANTS ESTIMATED: CONTINUING PARTICIPANTS AND DROPOUTS IPT1 Information — .88 .90 Synonyms — .86 .88 Analogies — .66 .66 Figure Logic — .55 .57 Block Design — .82 .83 Card Rotations — .73 .73 Digit Span Forward — .54 .54 Digit Span Backward — .56 .55 Thurstone Picture Memory — .67 .66 Names and Faces-lmm edlate — .59 .60 Names and Faces-Delayed — .64 .64 Digit Symbol — .81 .86 Figure Identification .73 .76 The finding that these estimated correlations were nearly identical to those obtained from the continuing participant group suggests that test reliability was not much affected by attrition. Thus, estimates of change in cognitive function could be considered reliable. Attrition analyses were performed using a two group path model which is conceptualized in terms of slope estimates, mean cognitive performance scores, standard deviations and random error. Circles represent latent or unobserved variables, squares represent observed variables and triangles represent a constant. The path diagram in Figure I seen below illustrates the variables analyzed in this study. 28 & k X Y B X B t s , O KJ u 8 f s . © KJ ts . © KJ Figure 1 . Two group path model for independent group (continuous and dropout) regressions. In the dropout group, the Time 2 variable is latent. In this figure, X refers to the respective cognitive performance score at the first time of measurement. Y refers to respective cognitive performance score at the second time of measurement, e refers to random error that can represent errors in measurement which are expected to change randomly over time for any individual. 1 refers to a constant vector of ones. The directed arrow going from 1 to X reflects the average mean for the first time of measurement. The directed arrow going from the 1 to Y reflects the intercept and the directed arrow going from X to Y reflects the slope. The arrows going from e to X and e to Y represent the standard deviation and the error variance, respectively. As mentioned earlier, this model constrained the path between IPTI and IPT2 to be invariant, i.e., to have an equivalent relationship across groups. A model fit indicates that there is no difference between the two groups in the regression function. This model can be compared both to a saturated model, which is what is expected if the dropout group matched the continuing participants exactly, and to an independence model, which tests the hypothesis that the relationship between the first and second time of measurement is 29 different for dropouts and continuing participants. The model with the lowest AIC provides the best description o f the data, which in this case, describes whether the respective observed IPT2 score for the continuing participants is different from that which would have been obtained from the whole sample (predicted IPT2 scare based on inclusion o f dropouts’ IPTI score). As indicated in Table 5, the study model fit most of the cognitive measures fairly well. Jtam T w o Q fO tp P * h k k x tti _ PAMMETEHS MEASURES_______________________________________________________________ TllltrtMB UMiailD MMMIIO noim iiocr w r i n i m r m i nctvxa m ci* n c i ^ ttvr rw iti _ iirciwinoa nioiTMi m to a in iooie m w i i w t w ro tw ia w tn w a mwhw w u m n t w (Biinnouwi EQUAL REGRESSION BETWEEN GROUPS {INVARIANT) Chl-Squir* (DF-J) 6.33 2.04 12.56 12.76 15.35 15.42 4.16 5-21 4.77 3.43 7.52 25.84 10.56 AIC Model 20.33 10.04 20.56 20,78 20.35 20.42 16.18 1021 16.77 17.43 21.52 30.84 33.56 Cetureltd Model 24,00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24.00 24,00 24.00 24.00 24.00 Independence Model 3025.67 2050.65 2566.34 2772.12 2167.53 2055.60 2646.01 2316.73 2660.45 860.66 747.80 260727 2670.71 PARA METER ESTIMA TES SLOPE - ► B > 0,61 0.63 0.02 0.56 0.74 0.00 0.55 0.56 0.67 0.50 0.68 0.62 0.70 INTERCEPT - ► 0, e.ei 3.42 5.70 7.30 4.30 14.70 2.68 1.72 6.60 0.00 0.66 3.60 5.75 MEAN TIME 1 —► M, 31.57 16.03 14.30 17.60 18.06 40.35 5.65 4.14 20.60 1.08 1.60 3024 20.48 STANDARD DEVIATION TIME 1 - ► 0. 6.04 5.74 3.02 3.67 7.54 1020 1.18 1.18 4.54 1.65 1.66 1123 7.43 ERROR VARIANCE TIME 2 - ► S. 3.45 2.75 2.70 3.32 3.60 12.50 1.00 1.03 3.37 1.47 1.37 6.80 5.31 In the case of Information, Synonyms, Digit Span Forward and Digit Span Backward, Thurstone Picture Memory, and both Names and Faces-Immediate and Names and Faces- Delayed, the study model best described the data, suggesting that there is no difference between the observed IPT2 scores and the predicted IPT2 scores, i.e., those that would have been obtained had the dropouts remained in the sample. This conclusion also applies for Analogies and Figure Logic given that the study model and the saturated model were nearly equivalent. However, for Block Design, Card Rotations, Figure Identification and Digit Symbol, the discrepancy between the study model and the saturated model was larger, intimating that there may be selective attrition effects on spatial ability (Block Design and Card Rotations) and perceptual speed (Digit Symbol and Figure Identification), i.e., that the observed IPT2 score is likely higher than would be expected had the dropouts remained in the sample. This finding suggests that any indicator of change in these abilities may be an underestimate. The selection effects observed may have been due to differences between the continuing and dropout groups on variables such as health or education. However, group comparisons indicated that dropouts and continuing participants did not differ significantly on self-reported educational attainment or on several indicators of physical health status (metabolic conditions, cancer, cardiovascular disease and respiratory conditions.) However, the dropouts were older, with a mean age at Q 1 of 67 (standard deviation of 8 years) at Q l, in comparison the continuing participants’ mean age of 63 (standard deviation of 7 years). In summary, to examine the possibility of non-random attrition of subjects over a three year interval, all available information from the dropouts and continuing participants was included. The invariant models fit the data relatively well for most abilities, with the exception of perceptual speed and spatial ability. Cognitive functioning Intercorrelations between the measures were examined in order to confirm the five areas of cognitive function. These correlations are reported, by smoking status, sex, and time of measurement (IPTI and IPT2) in Tables 6a - 9b. In general, measures assessing acculturated knowledge are highly correlated as are those assessing perceptual speed. Among fluid ability measures, analogies is more correlated with acculturated knowledge tasks than with the other fluid tasks, particularly in the women smokers. Measures of spatial ability and visual-construction are moderately correlated. Visual memory and verbal memory measures tend not to be highly correlated, among women in particular. Most measures show a moderate correlation with perceptual speed. These associations vary across time, particularly for the women smokers, which may reflect the effects of interindividual differences in intraindividual change. 32 Table 6a IntOKQtmlatlons among Cognltho Moasums at IPH: Woman Smokors MEASURE 1 2 3 4 S 6 7 6 9 10 11 12 13 1. INFORMATION — .63 .06 .03 .27 .05 .04 .16 .13 .12 .08 .39 .19 2. SYNONYMS — .31 .03 .43 .43 .33 .06 .36 .23 .40 .57 .50 3. ANALOGIES — -.29 .16 -.09 .33 .09 .08 .33 .46 .03 .01 4. FIGURE LOGIC — .36 .56 .10 .29 .14 -.11 -.07 .43 .59 S. BLOCK DESIGN — .47 .29 .57 .42 .16 .34 .66 .49 6. CARD ROTATIONS — .07 .17 .32 .10 .19 .70 .77 7. DIGIT SPAN FORWARD — ,31 .01 .20 .51 .07 .10 S. DIGIT SPAN BACKWARD — -.06 .11 .09 .41 .09 9. PICTURE MEMORY — .27 .27 .47 .59 10. NAMES AND FACES - IMMED. — .79 .19 -.11 11. NAMES AND FACES-DELAYED — .12 .06 12. DIGIT SYMBOL — .76 13. FIGURE IDENTIRCATION — Table 6b Intonxmilatkins among Cognltivo Moasums a t IPT2: Womon Smokors MEASURE 1 2 3 4 5 6 7 6 9 10 11 12 13 1. INFORMATION — ,6B .34 ‘ .26 .36 .13 .19 .21 .40 .03 .18 .43 .45 2. SYNONYMS — .24 .11 .24 .09 .40 .31 .32 .40 .50 .34 .18 3. ANALOGIES — .16 -.05 .12 -.03 .20 -.03 .26 .17 .15 .26 4. FIGURE LOGIC — .72 .44 .10 .17 .37 .38 .44 .43 .50 5. BLOCK DESIGN — .33 .08 .29 .41 .28 .39 .48 .41 6. CARD ROTATIONS — -.30 .32 -.28 .26 .26 .06 .25 7. DIGIT SPAN FORWARD — .36 .32 .00 .07 ,27 .01 6. DIGIT SPAN BACKWARD — .03 -.30 -.08 .26 .13 9. PICTURE MEMORY — .33 .31 .52 .45 10. NAMES AND FACES - IMMED. — .85 .33 .37 11. NAMES AND PACES-DELAY — .32 .33 12. DIGIT SYMBOL — .73 13. FIGURE IDENTIFICATION — TaUe7a Iriazcm ltth n s among Cogntlvn Manures *1 IPTI: lAbmen Non-smokers MEASURE 1 2 3 4 5 6 7 8 a 10 11 12 13 1. INFORMATION — ,70 .52 .41 .49 .17 .20 .40 .40 .28 .34 .51 .33 2. SYNONYMS — .60 .40 .52 .14 .21 .32 .33 .31 .36 .53 .34 3. ANALOGIES — .40 .48 .23 .21 .35 .32 .36 .39 .52 .36 4. FIGURE LOGIC — .54 .37 .13 .29 .34 .24 .21 .47 .46 5. BLOCK DESIGN — .52 .20 .41 .41 .30 .32 .66 .58 6 CARD ROTATIONS .06 .15 .24 .21 .14 .44 .56 7. DIGIT SPAN FORWARD — .44 .17 .04 .O S ,17 .10 6. DIGIT SPAN BACKWARD — .24 .23 .29 .40 .23 9. PICTURE MEMORY — .34 .37 .45 .32 10. NAMES AND FACES - IMMEDIATE — .64 .42 .27 11. NAMES AND FACES-DELAYED — .44 .27 12. DIGIT SYMBOL — .70 13. FIGURE IDENTIFICATION — TiU eTb IrtercomJatkms among Coonitve Manures at IPT2: Wtxnon Non-smokers MEASURE 1 2 3 4 5 6 7 8 9 10 11 12 13 1. INFORMATION — .71 .48 .32 .46 .19 .20 .34 .42 .37 .34 .45 .28 2. SYNONYMS — .62 .4t .47 .20 .29 .32 .44 .40 .35 .52 .33 3. ANALOGIES — .39 .39 .20 .29 .30 .40 .34 .29 .50 .18 4. FIGURE LOGIC — .47 .38 .17 .16 .37 .42 .36 .50 .45 5. BLOCK DESIGN — .58 .35 .39 .42 .44 .42 .70 .59 6. CARD ROTATIONS — .14 .20 .25 .30 .31 .62 .59 7. DIGIT SPAN FORWARD — .47 .25 .29 .26 .29 .19 6. DIGIT SPAN BACKWARD — .33 .22 .30 .33 .21 9. PICTURE MEMORY — .43 .41 .50 .31 10. NAMES AND FACES-IMMEDIATE — .83 .51 .36 11. NAMES AND FACES -DELAYED — .48 .34 12. DIGIT SYMBOL — .65 13. FIGURE IDENTIFICATION — 33 Table 8a InlareomUalfans among Coon M m Moasums at IPT1: Mon Smokors MEASURE 1 2 3 4 S B 7 8 B 10 11 12 13 1. INFORMATION — .73 .72 .41 .50 .44 .40 .43 .18 .20 .25 .52 .48 2. SYNONYMS — .77 .14 .33 .30 .52 .37 .33 .42 .14 .64 .24 3. ANALOGIES — .15 .37 .32 .42 •4t .33 .32 .13 .48 .33 4. FIGURE LOGIC — .49 .24 .25 .39 .01 .32 .31 .35 .4B 5. BLOCK DESIGN — .73 .36 .52 .41 .37 .26 .65 .46 6. CARO ROTATIONS — .50 .24 .38 .25 .17 .66 .64 7. DIGIT SPAN FORWARD — .58 .19 .28 .25 .65 .58 8. DIGIT SPAN BACKWARD — .29 .42 .45 .47 .30 B. PICTUREMEMORY — .35 .34 .40 .14 10. NAMES AND FACES - IMMEDIATE — .45 .42 .18 11. NAMES AND FACES-DELAYED — .24 .13 12. DIGIT SYMBOL — .58 13. FIGURE IDENTIFICATION — TABLE 8b Intorvonolotlon s amona Cognltlm Moasiwtwol IPTZ Mon Smokors MEASURE 1 2 3 4 5 6 7 8 a 10 11 12 13 1. INFORMATION — .73 .70 .35 .57 .36 .41 .37 .41 .50 .31 .72 .58 2. SYNONYMS — .72 .31 .43 .40 .61 .48 .33 .50 .41 .67 .68 3. ANALOGIES — .41 .60 .46 .64 .41 .46 .68 .46 .65 .79 4. FIGURE LOGIC — .52 .28 .32 .41 .55 .28 .22 .30 .44 5. BLOCK DESIGN — .55 .42 .32 .30 .27 .30 .60 .69 6. CARD ROTATIONS — .25 .33 .46 .27 .16 .47 .71 7. DIGIT SPAN FORWARD — .52 .14 .31 .06 .48 .49 8. DIGIT SPAN BACKWARD — .62 .41 .36 .38 .31 9. PICTUREMEMORY — .45 .34 .40 .40 10. NAMES AND FACES-IMMEDIATE — ,72 .41 .50 11. NAMES AND FACES-DELAYED — .11 .43 12 DIGIT SYMBOL — .76 13. FIGURE IDENTIFICATION — Table 9a Intaocm /atkm t among Cognilvo M s a s tm at IPTI: Afcn N oryim okm MEASURE 1 2 3 4 5 6 7 a B 1D 11 12 13 1. INFORMATION — .84 .57 .38 .49 .33 .46 .47 .37 .29 .38 .74 .43 2. SYNONYMS — .74 .44 .56 .37 .58 .57 .52 .46 .51 .73 .53 3. ANALOGIES — .33 .45 .33 .43 .54 .45 .23 ,32 .64 .37 4. FIGURE LOGIC — .45 .63 .43 .28 .50 .24 .29 .57 .61 5. BLOCK DESIGN — .64 .33 .24 .56 .34 .46 .72 .66 6. CARD ROTATIONS — .33 .14 .32 .22 .28 .51 .70 7. DIGIT SPAN FORWARD — .40 .39 .43 .37 .45 .25 8 DIGIT SPAN BACKWARD — .30 .24 .23 .45 .20 9. PICTUREMEMORY — .24 .21 .53 ,43 10. NAMES AND FACES-IMMEDIATE — .77 .39 .32 11. NAMES AND FACES- DELAYED — .46 .44 12 DIGIT SYMBOL — .67 13. FIGURE IDENTIFICATION — Table 9b Irterzomlathns among Cognltivo M a u tn t at IPT2: Msn Non-smokeri MEASURE 1 2 3 4 5 6 7 8 9 13 11 12 13 1. INFORMATION — .85 .61 .44 .54 .17 .50 .51 .51 .29 .27 .68 .34 2 SYNONYMS — .65 .48 .57 .19 .51 .61 .59 .41 .32 .63 .34 3. ANALOGIES — .56 .70 .39 .33 .45 .35 .16 .18 .63 .55 4. FIGURE LOGIC — .67 .43 .24 .29 .26 .31 .34 ,57 .44 5. BLOCK DESIGN _ .60 .24 .43 .48 .22 .24 .80 .72 6. CARD ROTATIONS — .13 .11 .23 -.10 -.04 .52 .64 7. DIGIT SPAN FORWARD — .51 .20 .40 .32 .34 .09 8 DIGIT SPAN BACKWARD — .57 .37 .25 .44 .34 B . PICTUREMEMORY — .27 .22 .44 .26 10. NAMES AND FACES-IMMEDIATE .86 .27 .04 11. NAMES AND FACES -DELAYED — .22 .18 12. DIGIT SYMBOL — .65 13. FIGURE IDENTIFICATION 34 To examine intraindividual change and whether it was similar for smokers and for non- smokers, correlations between IPTI and IPT2 scores were calculated for eaclr cognitive measure. These arc shown in Table 10 for smokers and for non-smokers, by gender. Table 10 Correlations betw een IPT1 an d IPT2 b y Smoking Status and Sox__________________________________________ SMOKERS NON-SMOKERS _____________________________________________________ MEN WOMEN_____________ MEN WOMEN INFORMATION .63 .86 .92 .87 SYNONYMS ,85 .89 .93 .84 ANALOGIES .71 .43 .64 .66 FIGURE LOGIC .34 .65 .59 .57 BLOCK DESIGN .83 .87 .89 .78 CARD ROTATIONS .67 .47 .76 .69 DIGIT SPAN FORWARD .69 .74 .51 .51 DIGIT SPAN BACKWARD .63 .28 .51 .52 PICTURE MEMORY .50 .80 .69 .65 NAMES AND FACES - IMMEDIATE .43 .66 .63 .53 NAMES AND FACES - DELAYED .25 .75 .69 .59 DIGIT SYMBOL .68 .76 .90 .81 FIGURE IDENTIFICATION .57 .72 .70 .78 There was no particular evidence of differences between groups in correlations between the two times of measurement, allowing for the differences in size of the different subsamples. Mean raw scores and standard deviations for all measures at IPTI and IPT2 are shown by smoking status (non-smokers or current smokers) and sex in Table 11. 35 Tafal* 11 K«w Scara Maatu mnJ S ttn d rd Pavlalfare on CaanlMi MwiifM by Stnokhg S titu i and Sctt SMOKERS NON-SMOKERS IPTI IPT2 IPTI IH E MEASURE MEN |n-27) WOMEN (n-21) MEN (nK 27> WOMEN (rt-211 MEN In-51) WOMEN In-179) MEN tn-51) WOMEN (rt-179) INFORMATION M 34.69 28.57 35.15 30.80 35.50 29.78 35.55 30.19 SO SCO 7.07 5.54 7.05 860 7.76 8,54 7.61 SYNONYMS M 10.89 20.10 10.02 10.75 19.70 16.54 10.75 10.86 SO 5 57 5.21 5.07 4.62 7.02 5.20 6.63 5.27 ANALOGIES M 15.33 13.61 18.42 14.05 13.35 14.07 15.79 14.36 SO 4.24 3.22 4.35 2.52 4.51 3.54 3.02 3.43 FIGURE LOGIC M 10.75 15.10 1523 17.65 15.71 17.53 19.54 17.40 SO 3 00 2.00 3.25 3.09 4.04 4.05 4.12 3.08 BLOCK DESIGN M 15.53 15.57 17.77 15.75 21.04 15.00 21.40 18.58 SO 7.25 5.00 5.72 5.78 7.28 7.57 8.71 7.17 CARD ROTATIONS M 40.00 48 20 47.60 44.53 6222 45.31 60.93 45.75 CO 17.05 2082 16.07 12.45 20.09 17.45 10.43 16.72 DIGIT SYMBOL M 35.00 42.52 33.63 37.80 4250 30.36 40.49 36.37 CD 10.42 5.11 t127 9 29 13.12 11.44 12.66 11.66 FIGURE IDENTIFICATION M 27.52 31.20 25.01 28.50 29.67 30.01 30.38 20.63 SO 5 87 7.33 0.05 8.48 7.63 7.01 8.04 8.28 DIGITCPAN FORWARD M 3,78 5.57 6.00 6.00 5.70 5.66 5.58 5.74 SD 1.31 1.40 1.40 1.30 1.24 1.11 1.13 1.14 DIGIT SPAN BACKWARD M 4.44 4.05 4.25 4.10 4.35 4.04 4.29 298 SO 1.05 1.20 0.08 1.30 1.16 1.16 1.36 1.32 PICTUREMEMORY M 15.51 21.71 19.00 20.05 20.43 2t.35 20.04 21.28 SD 4.22 4.20 4.50 4.55 4.40 4.35 4.34 4.44 NAMESANO FACES. IMMEDIATE M 1.35 1.75 1.60 2.05 1.02 236 1.74 2.47 SO 1.35 1.55 1.45 1.76 1.67 1.01 1.69 1.78 NAMES ANO FACES - DELAYED M 0.65 1.52 1.45 1.65 1.55 1.50 1.35 203 SD 0,00 1.60 1.42 1.42 1.46 1.78 1.61 1.81 On average, male non-smokers showed better performance than male smokers, especially 011 fluid measures, whereas, on average, female non-smokers show only some performance advantages over female smokers. However, raw scores do not reflect change in functioning relative to baseline performance (IPTI). In order to obtain estimates of change relative to baseline performance, raw scores for each cognitive test were converted to Zrscores: IPTI scores were standard normal and IPT2 scores were standardized using the raw means and standard deviations from IPTI. The standard scores for all measures at IPTI and IPT2 are shown by smoking status (non-smokers or current smokers) and sex in Table 12. 36 Tabffl 12 Siand# d Scora M w ii And Sl&ndard PqvMtvii an Cognj Va trad Ay $moi w g S t* us M d Star 6M0KER8 NON-SMOKERS IPTI 1EJ2 IPTI [PT2 M EN WOMEN M EN WOMEN M EN WOMEN M EN WOMEN MEASURE (n-27) (HF71) ln-27) (nrtl) (n-5l| rnvt?9) (n«51) <n-17B) INFORM ATION M *0.00 •0.14 0.11 0.19 0.03 0.02 0.17 0,07 80 0.75 0.00 0.60 0.61 1.12 1,00 1.21 0.67 SYNONYMS M 0.01 0.22 0.10 0.10 -0,01 -0.03 0,13 0.02 6D 0,65 1.00 0.67 0.69 1.06 1.00 1.06 1.01 ANALOGIES M 0,00 -0TJ7 020 -0.03 Q .Q 0 041 O.tl 0.00 SO 0.07 0.02 1.03 0,71 1.03 1,01 0.93 0.07 FIGURE LOGIC M 021 •0.19 -0.14 0.05 -0.11 0.02 022 •0.01 80 0.81 0.75 0.89 1.06 1.06 1,03 1.13 1.00 BLOCK DESIGN M •4.30 •0.41 •0.31 -0.36 0.1Q 0.05 0.17 0.01 SO 4.06 0.70 0.66 0.60 0.68 1.01 0.86 1,00 CARD ROTATIONS M -0.40 0.19 -0.44 •0.34 0 2 \ -0,02 027 0.05 8D 0.66 1.10 0.69 0.60 1.00 0.S6 0.66 0.66 DIGIT SYMBOL M •0.24 0.25 -0.93 -0.19 0,13 •0.03 0.00 -0.26 80 0.64 0.73 1.00 0.85 1.00 1.03 1.13 1.00 FIGURE IDENTIFICATION M -0.20 0.19 •0.30 -0.17 0.12 -0.02 D .29 0 02 80 0.61 0.63 0.69 1.10 1,06 1,01 1.16 1.08 DIGIT SPAN FORWARD M 001 •0.07 026 024 0.00 0.01 0,16 0.12 SD 1.04 1.23 129 t.to 0.09 0.67 0.67 1-01 DIGIT SPAN BACKW ARD M 0.05 0.00 *0.03 0.00 ■0,03 0.00 0.00 -0.10 SO 0.04 1.02 0.67 1.11 1.04 1,00 1*20 1,13 PICTURE M EMORY M ■024 0,06 •024 •0.31 0.13 •0.01 •0.01 -0.03 SD 0.9G 0.67 1.01 1.10 1.01 1,01 0.65 f.01 NAMES AND FACES - IM M EDIATE M •0.24 *0,28 -0.02 -4.09 0,13 0.03 0.00 0.16 SD 0.60 0.62 0,61 0.97 1.00 1.02 0.04 0.66 NAMES ANO FACES - DELAYED M -0.32 -0,66 0.07 0.00 0.17 0 01 0.01 022 80 0.74 0.61 0.92 0.63 1.06 1.01 1.04 1.00 Standardized scores enabled relative comparisons across cognitive ability measures. In addition, by standardizing IPT2 scores based on IPTI scores, standardization also enables relative comparisons of rates of change. Standardization was done within sex because males and females have been found to evidence different patterns of performance on certain cognitive abilities (Halpem, 1994; Wiederholt et al., 1993). In addition, the proportion of smokers to non-smokers differed significantly for men and women. Tlius, all analyses were done separately by sex. Analyses were conducted at an individual level, which included all individuals regardless of twin status such that some are partners to one another, and at a co-twin control level, which included only twins that are partners to one another. 37 Individual Data Analyses Three types of analyses were performed at the individual level. Analyses of variance (ANOVAs) were performed to examine whether smokers showed higher mean level cognitive performance than non-smokers at IPTI and IPT2. Next, analyses of covariance (ANCOVAs) were performed, using age as a covariate, to test whether mean level performance differences observed between smokers and non-smokers at both IPTI and IPT2 remained above and beyond the effects of age. It is possible that despite being similar ages, smokers and non-smokers may evidence varying effects of intraindividual change, i.e., effects of disease states on function, for which age would serve as a proxy. Age explained some, but not all of the differences in cognitive performance between smokers and non-smokers at both times of measurement and thus, age was partialled out of the cognitive function scores used for the third analyses: a 2 (smoking group) by 2 (time of measurement) repeated measures ANOVA. Residualized scores were used so that any differences observed in rates of change in cognitive performance between smokers and non-smokers could be explained over and above age differences. Although specific hypotheses were made about which abilities would show differences among smokers and non-smokers, hypotheses were not made for the specific psychometric tests measuring the five areas of cognitive ability, i.e., whether among the tests measuring spatial ability (a fluid ability), Card Rotations would show greater performance differences than Block Design. From this standpoint, the latter analyses were considered exploratory and, accordingly, a Bonferroni procedure was used to correct for a family-wise Type I error rate of .20 (acceptable standard for exploratory analyses). A p- value of p <.008 was obtained. Both the adjusted and original p-values are reported. 38 Analysis of variance (ANOVA1 Hypotheses 1 and 3 were examined using a one-way analysis of variance by standardized cognitive performance scores. Men A significant group effect was found for spatial ability and visual memory but the direction o f the difference was contrary to expectation. For men, at IPTI, smokers showed significantly lower mean levels on Card Rotations, F(l,73) = 6.86, p = .012 and Names and Faces-Delayed, F(l,73) = 4.34, p = .041. Although differences were not significant at the adjusted alpha level, the smokers tended to have worse spatial ability and visual memory at IPTI than non-smokers. At IPT2, smokers showed lower mean levels on Block Design, F( 1,74) = 5.0, p = .028; Card Rotations, F(l,68 = 9.04, p = .004; Digit Symbol, F(l,69) = 4.71, p = .033 and Figure Identification, F(l,72) = 4.34, p = .041. Although differences were not significant at the adjusted alpha level except for one measure of spatial ability, smokers tended to continue to show worse spatial ability than non-smokers when tested three years later. Differences emerged at reassessment on perceptual speed/information processing, with smokers tending to perform worse than non-smokers: in comparison to their IPTI scores, smokers showed some slight decrements in performance whereas non-smokers showed some slight improvements. Women At IPTI, smokers showed significantly lower mean levels on Block Design, F( 1,197) = 3.99, p = .047. These findings suggest that smokers tended to show poorer spatial ability than non-smokers. 39 At IPT2, no significant group effects were observed. Block Design no longer showed significant group differences: at the three year reassessment, smokers appeared to show minor improvements. Analyses of Covariance (ANCOVAl Hypothesis I and 3 were re-examined by an analyses of covariance (ANCOVA), using age as a covariate, to test whether mean level performance differences observed between smokers and non-smokers at both IPTI and IPT2 remained above and beyond the effects of age differences. The ANCOVA revealed that differences in cognitive performance between smokers and non-smokers at both times of measurement were not attributed to age alone, but to the group effect on spatial ability, visual memory and psychomotor speed/information processing. Men At IPTI, Block Design, F(2,72) = 5.40, p = .023; Card Rotations, F(2,72) = 10.94, p = .002 and Names and Faces-Delayed, F(2,72) = 5.17, p = .026 showed group differences. At IPT2, Block Design, F(2,73) = 7.52, p = .008; Card Rotations, F(2,67) = 12.13, p = .001; Digit Symbol, F(2,68) = 5.95, p = .017 and Figure Identification, F(2,71) = 6.21, p = .015 showed group effects. Although the differences were not significant at the adjusted alpha level, in each case the smokers performed worse than the non-smokers. Differences between smokers and non-smokers on spatial ability, visual memory and perceptual speed/information processing tended not to be explained by the effects of age alone. Women At IPTI, Block Design, F(2,196) = 7.95, p = .005 showed a group effect. At IPT2, Block Design, F(2,190) = 7.54, p = .007 and Card Rotations, F(2,176) = 6.35, p = .013 40 showed group effects. Differences remained significant at the adjusted alpha level for Block Design. Smokers performed worse than non-smokers on spatial ability, a difference that was not explained by the effects of age alone. Repeated measures analyses of variance (ANOVAI Hypotheses 2 and 4 were examined using a 2 by 2 repeated measures ANOVA on residualized cognitive performance scores, partialled for age effects. Age was partialled out of the cognitive ability scores because it explained some, but not all of the differences in cognitive performance between smokers and non-smokers at both times of measurement. Repeated measures ANOVA computes a main effect for smoking group, a main effect for time of measurement and an interaction between smoking group and time of measurement. The main effect for group is the same for the one-way ANOVAs but it has the effect of age taken out. Findings were unexpected for both men and women with respect to anticipated changes in performance on fluid ability, perceptual speed and memory measures. Findings for acculturated knowledge were in the expected direction. Men There was a significant group main effect for Block Design, F(l,72) = 6.15, p = .016; Card Rotations, F(l,66) = 13.58, p = .001; Digit Symbol, F( 1,68) = 5.52, p = .022 and Figure Identification, F( 1,71) = 6.37, p = .014, indicating maintenance of mean level differences from IPT1 to IPT2. Smokers continued to demonstrate worse spatial ability and perceptual speed/information processing, although only differences in one measure of spatial ability remained significant at the adjusted alpha level. No subsequent interaction terms were significant, indicating that there were no differences between smokers and non-smokers in the rate of change in cognitive performance. 41 Women There was a significant main effect for smoking group for Block Design, F( 1,191) = 9.43, p = .002 and Names and Faces-Immediate, F( 1,190) = 3.93, p = .049 indicating maintenance of mean level differences at 1PT1 and IPT2. Smokers continued to demonstrate worse spatial ability and showed a tendency towards worse verbal memory. There was a significant time and interaction effect for Card Rotations, F( 1,176) = 5.46, p = 021 and F( 1,176) = 6,89, p = .009, respectively, indicating that the change from the first to second time of measurement was different for the two groups: smokers showed a slight decline in performance compared to a slight improvement in performance for the non-smokers. Taken together, the findings suggest that, over and above the effects of age, smokers and non-smokers show mean level differences in cognitive performance that arc maintained at a three year reassessment. The specific abilities that showed mean level differences reported by sex were: among men, smokers showed poorer spatial ability and perceptual speed than non-smokers whereas among women, smokers showed poorer spatial ability than non-smokers. Performance tended not to change at different rates for smokers and non-smokers, with the possible exception o f spatial ability hi women. Co-Twin Control Analyses Hypotheses about mean level differences explored in the individual level ANOVAs were examined in twins, with smoking and non-smoking groups represented by twin pairs in which one member was a smoker and the partner was a non-smoker. In the sample of 86 pairs, there was a very high concordance rate for smoking status: only 12 pairs (11 female and 1 male) were discordant (refer to Table 13). 42 Table 13 Concordance for StnoUng Status among Co-Twin Pars Wbmen Men Number of pairsconoordart for smoking 1 5 Number of pairs conoordart for not smoking 55 13 tim ber of dscordart pairs 1 1 1 For the discordant pairs, difference scores were calculated by subtracting the non- smoker's cognitive performance score from the smoker’s score. A positive difference score indicated that the smoker showed better cognitive performance than the non smoking partner whereas a negative difference score indicated that the non-smoker demonstrated better cognitive performance than the smoking partner. Paired t-test were used to compare the smokers and non-smokers on all raw cognitive test scores at IPT1 and IPT2. Table 14 shows results for women only, given that there was only one male discordant pair. Table 14 Raw Mean Intmpalr Diffemnca and Standard Error for Fema/e Discordant Pairs IPT1 IPT2 INFORMATION -5.54 23 9 p< .04 -4.36 233 ns SYNONYMS •0.55 1.67 ns -1.30 1.94 ns ANALOGIES -1.S4 0.87 ns -1.10 1.33 ns FIGURE LOGIC -1.82 DBS ns •0.62 0.87 ns BLOCK DESIGN -3.36 1.89 ns -4.18 2 2 7 ns CARD ROTATIONS -4.64 7,B O ns ■4.44 4.38 ns DIGIT SPAN FORWARD 0.27 0,41 ns 0.09 0.39 ns DIGIT SPAN BACKWARD 0.00 0.19 ns 0.55 0.34 ns PICTURE MEMORY -0.09 1.25 ns -200 1.03 ns NAMES AND FACES - IMMEDIATE -1.18 0.44 p< .02 -1,09 0.37 p<.01 NAMES AND FACES - DELAYED 43.01 0.37 p< .03 -1.00 0.57 ns DIGIT SYMBOL -2.18 2.39 ns -5.20 271 ns FIGURE IDENTIFICATION -4.00 1.65 D< .03 -7.38 1.84 D< .003 A trend favoring the non-smoker was found at IPT1 for Information (p<.043), Names and Faces-Immediate (p<.023), Names and Faces-Delayed (p<.033) and Figure Identification (p<.032). Non-smokers showed better visual memory, perceptual * speed/information processing and acculturated knowledge than smokers. At IPT2, a significant difference was found in Figure Identification at IPT2 (p<.003) and Names and 43 Faces-Immediate (p<.0l4), indicating that non-smokers continued to show better visual memory and perceptual speed/information processing than smokers. DISCUSSION The purpose of this study was to examine the relationship between smoking and cognitive performance in non-impaired older adults. It extended a previous investigation by Hill (1989) in five ways: a) use of individuals with a prospectively documented history of smoking, b) use of a three year reassessment of cognitive performance, c) separate analysis of men and women d) examination of selection bias and e) incorporation of a matched-twin case control design. The matched twin design enabled additional control of variables that could influence the association between smoking and cognitive function, e.g. the effects of differential survival, that is, whether smokers who survive have hardier genes than smokers who do not survive. Differences in survival can bias comparisons of smokers and non-smokers. Tins latter issue was controlled by examining twins who were discordant for smoking status, that is, one twin smoked and the other did not. Because twins share the same genetic makeup, any observed differences on cognitive performance would be due, likely, to whether or not the individual smokes. This study examined whether smokers demonstrate better cognitive performance than non-smokers and show less decrements than non-smokers in abilities expected to decline, normally, with age. The magnitude of decline was expected to vary across smoking groups for those abilities most closely linked to nicotine’s putative mechanism of action, that is improved attention, information processing and memory through enhanced arousal and cholinergic system activity. Specifically, 1) attention, perceptual speed/information processing and fluid abilities (reasoning and spatial ability) were expected to decline in 4 4 non-smokers but to decline at a slower rate or to remain stable in smokers. Acculturated knowledge was expected to remain stable or improve and to show no differences between the two groups. Results indicate a pattern of performance differences that was contrary to expectation and varied slightly by sex. Over and above the effects of age differences in cognitive performance, smokers and non-smokers demonstrated mean level performance differences on specific cognitive ability tasks, a differential that was maintained over a three year period. Male smokers tended to show poorer spatial ability, visual memory and perceptual speed/information processing than non-smokers, whereas female smokers tended to show poorer spatial ability than non-smokers. Co-twin control results showed similar findings, suggesting that differences between smokers and non-smokers were not due to effects of differential survival. Despite mean level differences, changes in cognitive performance did not occur at different rates for smokers and non-smokers. Overall, results support the hypothesis that smoking compromises cognitive performance in older adults, namely perceptual speed/information processing and spatial ability. The findings are consistent with a previous investigation by Hill (1989) and may support concerns raised elsewhere (see Riggs, 1993) that findings from case-control studies of smoking as a protective factor may be artifactual. Riggs (1993) argues that smokers who survive to the age of risk for cognitive impairment may have hardier, superior aging genes than those who do not survive. Smokers would therefore be a more select group than non-smokers, a selection effect that could be controlled for only by matching on genes, not by age or sex as in case-control studies. Co-twin control analyses 45 provide a way to explore Riggs’s argument but in a sample of non-impaired smokers and non-smokers. Results from the co-twin control analyses suggest that selective survival is not operative, i.e., that smokers have hardier genes than non-smoker. However, it must be noted that findings are based on a small number of discordant female pairs and did not address whether other variables such as health status explain performance differences. Post hoc paired t-test analyses showed no differences between the smoking and non smoking twins on the number of physical illnesses, cardiovascular disease, metabolic conditions, cancer, emphysema or bronchitis. Overall, smokers do not evidence better health than non-smokers. However, the discordant pairs may still represent a select group in that both members of the pair are required to be alive, such that both the non-smoker and the smoker may be hardy. This group may be different from those in which only one member survives whether it is the smoker or non-smoker. Possible selective attrition for physical health may have affected individual level analyses. Post hoc analyses show that smokers who survived to IPT1 tended not to exhibit more health problems than their non-smoking counterparts. If, in general, smoking is associated with poorer health and iucreased morbidity (Colsher et al., 1990), then smokers should show more health problems than non-smokers. Given that this finding was not observed in the current sample, the smokers could be a select group who are less prone to illness and disease that can compromise cognitive performance. Thus, they may be less likely to exhibit change in cognitive functioning. Although this effect may have . occurred in the present sample, it would be likely to have less o f an influence on estimates of cognitive change than in previous research (Hill, 1989), because participants in the 46 present study were not excluded on the basis of physical health problems. Despite differences in selection on health, both the present study and the Hill study found minimal change between the times of measurement. Minimal change limits the possibility of testing the hypothesis that there would be group differences in patterns of change. The small amount of change may be due to the age of the sample. The sample was relatively young, with a mean age of 60 years and 64 years for women smokers and non-smokers, respectively and, 62 year for both men smokers and non-smokers. Given that age-related declines may not occur in many abilities until age 70 and beyond, the ability to find potential differences in patterns of change may have been reduced. This speculation is consistent with the fact that the relative magnitude of observed performance changes was small. However, as seen in the attrition analyses, changes in perceptual speed and spatial ability are likely to underestimate decline, given that individuals who dropped out of the study differed from continuing participants on these measures. Therefore, the observed differences in spatial ability may be particularly noteworthy. Hie likelihood of observing differences in patterns of change also may have been reduced if the amount of change that occurs over a three year interval is small. Although the interval of time between assessments is important to detecting change, other factors should be considered. One factor is the type of ability examined and the covariates that are considered (Hultch, Hertzog, Small, McDonald-Miszczak & Dizon, 1992). A second factor is the degree of practice effects (Matarazzo, Carmody & Jacobs, 1980). Practice effects are likely to be of particular concern for fluid abilities tests given that the task is dependent on novelty. With practice, novel tasks become more familiar and therefore 47 repeated use of the fluid ability task may not detect much change (Kaufman, 1994). However, the use of a three year time interval between testing may minimize these effects. On the basis of these results and those of Hill (1989), a number of directions for future research can be recommended. First, several cognitive assessments periods should be included. Second, direct measures of the biologically based component processes of memory and fluid abilities may be more sensitive to the types of changes that were expected to occur. It is possible that because the measures did not assess biologically based processes directly, tests allowed individuals room to compensate for change using other more ovcrdetermined abilities. More discrete measures of cognitive abilities, i.e., those that do not correlate with other ability measures, also may be more sensitive to change. Third, examination of individual variability in cognitive change is warranted. Mean level differences obscure individual variability and increases in intraindividual variability, possibly due to increasingly dysfunctional physiological mechanisms or increasing disparate intraindividual change due to the accumulated effects of different life histories (Nesselroade, 1991) may be expected. Interindividual differences in intraindividual change may have interfered with examination of group differences. A fourth recommendation is that the effects of direct administration of nicotine be examined. Attempts to ascertain the relationship between nicotine and cognitive change should not be abandoned given the neuropathology of Alzheimer’s disease. Individuals with Alzheimer’s disease show a decrease in acetylcholine metabolism (Neary et al., 1986; Sims et al., 1983) due to a loss of cholinergic cell bodies, larger neurons and degeneration of neuronal connections (Johnston, 1992). When given nicotine, acetylcholine metabolism is stimulated and some nicotinic cholinergic receptor activity is restored (Nordberd et al., 1993). By restoring activity to some of the remaining cholinergic receptors in AD individuals, cognitive impairment could be attenuated. However, if individuals are too far along in the progression of the disease they may lose the capacity for compensatory responses (Dekosky & Scheff, 1990). Given that acetylcholine metabolism has been found to correlate with cognitive decline and impairment (DeKosky & Scheff, 1990; Johnston, 1992), it remains possible that prolonged nicotine use early on could maintain receptor efficiency and retard the disease process, thereby actually delaying the manifestation of cognitive decline. It remains possible that there are positive effects of nicotine and negative effects of smoking, namely changes in cognitive function that are the result of a disease process such as impaired pulmonary function (Emery, Huppert & Schein, in press). Using smoking as a measure of the effects of nicotine is likely not to be a sufficient way to examine the relationship between nicotine and cognitive function. As suggested by this study and others examining physical performance measures, smoking seems likely to compromise cognitive performance, by lowered pulmonary function. 49 REFERENCES Akaike, H. (1987). Factor analysis and AIC. Psvchometrika. 52, 317-332. Americ, S. P. (1991). New nicotinic agonists and cerebral blood flow. In R. Becker & E. Giacobini (Eds.), Cholinergic basis for Alzheimer's therapy (pp. 386-394). Arthur, G. (1947). A point scale of performance tests. Revised format II. New York: Psychological Corporation. Babcock, R. L. & Salthouse, T. A. (1990). Effects of increased processing demands on age differences in working memory. Psychology and Aging. 5, 421-428. Baddeley, A. D. (1992). Working memory, Science. 255. 556-559. Benwell, M. E. M., Balfour, D. J. K., & Anderson, J. M. (1988). Evidence that tobacco smoking increases the density of(-)-[;>H]nicotiiie binding sites in human brain. Journal of N euro chemistry, 50. 1243-1247. Brenner, D. E., Kukull, W. A., van Belle, G., Bowen, J. D., McCormick, W. C., Tcri, L., & Larson, E. B. (1993). Relationship between cigarette smoking and Alzheimefs disease in a population-based case-control study. Neurology. 43. 293-299. Brody, N. (1992), Intelligence. New York: Academic Press. CDC. Smoking-attributable mortality and years of potential life lost— United States, 1988. (1991). Morbidity and Mortality Weekly Report, 40(4), 62-71. Campito, J. S. (1994). Verbal ability. In R. J. Sternberg (Ed.). Encyclopedia of Human Intelligence. Macmillan Publishing Company: New York. Ceci, S. J. (1994). Bioecological theory of intelligence. In R. J. Sternberg (Ed.), Encyclopedia of Human Intelligence. Macmillan Publishing Company: New York. Cederlof, R. & Lorich. U. (1978). The Swedish twin registry. In W. E. Nance (Ed.), Twin Research. Part B. Biology and Epidemiology fpp. 189-196). Alan R. Liss: New York. Colrain, I. M., Mangan, G. I., Pellett, O. L., & Bates, T. C. (1992). Effects of post- earning smoking on memory consolidation. Psvchophnrmacology. 108.448-451. Colsher, P. L., Wallace, R. B., Pomrehn, P. R., LaCroix, A. Z., Comoni-Huntley, J., Blazer, D.( Scherr, P. A., Berkman, L & Hcnnekens, C. H. (1990). Demographic and health characteristics of elderly smokers: results from Established Populations for Epidemiologic Studies of the Elderly. Ameiicnn Journal of Preventive Medicine. 6(2), 61-70. 50 Craik, F. I. M. & Jennings, J. M. (1992). Human memory. In F. I. M. Craik & T. A. Salthouse (Eds,), The handbook of aging and cognition (pp. 51-110). Hillsdnle, NJ: Erlbaum. Davies, D. R,, Jones, D. M. & Taylor, A. (1984). Selective and sustained-attention tasks: Individual and group differences. In R. A. Parasuraman & D. R. Davies (Eds.), Varieties of Attention (pp. 395-447). New York: Academic Press. DeFries, J. C., Plomin, R., Vandenberg, S. G., & Kuse, A. R. (1981). Parent-offspring resemblance for cognitive abilities in the Colorado Adoption Project: biological, adoptive, and control parents and one-year old children. Intelligence. 5, 145-277. Dekosky, S. T. & Scheff, S. W. (1990). Synapse loss in frontal cortex biopsies in Alzheimer's disease: correlation with cognitive severity. Annals of Neurology. 27, 457-464. Dempster, F. N. (1994) Interference. In R. J, Sternberg (Ed.), Encyclopedia of Human Intelligence. Macmillan Publishing Company: New York. Dempster, F. N. (1992). The rise and fall of the inhibitory mechanism: Toward a unified theory of cognitive development and aging. Development Review. J2, 45-75. Dureman, I., Kebbon, L., & Osterberg, E. (1971). Manual till DS-Batterret [Manual of the DS-Battery]. Stockholm: Psykologi Forlaget. Ekstrom, R. B., French, J. W., Hannan, H. H. & Derman, D. (1976). Kit of Factor- Referenced Cognitive Tests. Princeton, NJ: Educational Testing Service. Emery, C. E., Huppert, F. A. & Schein, R. L. (in press). Do pulmonaiy function and smoking behavior predict cognitive function? Findings from a British sample. Psychology and Health. Engle, R. W. (1994). Memory. In R. J. Sternberg (Ed.), Encyclopedia of Human Intelligence. Macmillan Publishing Company: New York. Ferini-Strambi, L., Smime, S., Garancini, P., Pinto, P. & Franceschi, M. (1990). Clinical and epidemiological aspects of Alzheimer's disease with presenile onset: a case-control study. Neuroepidemiology. 9. 39-49. Grossberg, G. T., Nakra, R,, Woodward, V., & Russell, T. (1989), Smoking as a risk factor for Alzheimer's Disease. Journal of the American Geriatric Society. 37. 819- 820. Halpem, D. F. (1994). Gender differences in intellectual abilities. In R. J. Sternberg (Ed.), Encyclopedia o f Human Intelligence. Macmillan Publishing Company: New York. 51 Hartley, A. A. (1992). In F. I. M. Craik & T. A. Saltliouse (Eds.), The handbook of aging and cognition (pp. 1-49). Hillsdale, NJ: Erlbaum. Hasher, L. & Zacks, R. T. (1988), Working memory, comprehension, and aging: A review and a new view. In G. H. Bower (Ed.), The psychology of learning and motivation (Vol. 22. pp. 193-224). New York: Academic Press. Henderson, Paganini-Hill, Emanuel, Dunn & Buckwalter. (1994). Estrogen replacement therapy in older women: Comparisons between Alzheimer's Disease cases and nondemented control subjects. Archives of Neurology. 51, 896-900. Hill, R. D. (1989). Residual effects of cigarette smoking on cognitive performance in normal aging. Psychology and Aging. 4(2), 251-254. Hindmarch, I,, Kerr, J. S., & Sherwood, N. (1990). Effects of nicotine gum on psychomotor performance in smokers and non-smokers. Psycliopharmacologv, 100. 535-541. Holman, A. & van Duijn, C. M. (1990). Alzheimefs disease, Parkinson's disease and smoking. Neurobiologv of Aging. JJ., 295. Horn, J. L. (1994). Theory of fluid and crystallized intelligence. In R. J. Sternberg (Ed.), Encyclopedia of Human Intelligence. Macmillan Publishing Company: New York. Horn, J. L. (1991). In McGraw, K. S., Werder, J. K. & Woodcock, R. W, (Eds.), Woodcock-Johnson Technical Manual: A Reference on Theory and Current Research (pp. 197-246). Allen, Texas: DLM Teaching Resources. Horn, J. L. (1976). Human abilities: a review of research theory in the early 1970s.. Annual Review of Psychology. 27. 437-485. Horn, J. L., Donaldson, G., & Engstrom, R. (1981). Apprehension, memoiy, and fluid intelligence decline in adulthood. Research on Aping. 3(1), 33-84. Hultsch, D. F., Hertzog, C., Small, B. J., McDonald-Miszczak, L. & Dixon, R, A. (1992). Short-term longitudinal change in cognitive performance in later life. Psychology and Aging, 7, 571-584. Ivy, G. O., MacLeod, C. M., Petit, T. L. & Markus, E. J. (1992). In F. I. M. Craik & T. A. Saltliouse (Eds.), The handbook of aging and cognition (pp. 273-314). Hillsdale, NJ: Erlbaum. James, J. R. & Nordberg, A. (1995), Genetic and environmental aspects of the role of nicotinic receptors in neurodegeuerative disorders: emphasis on Alzheimer's disease and Parkinson's disease. Behavior Genetics, 25(2), 149-159. 52 Johnston, M. V. & Silverstein, S. (1992). Fundamentals of drug therapy in neurology. In M. V. Johnston, R. L., MacDonald & A, B. Young (Eds.), Principles of Drug Therapy in Neurology (pp. 1-49), Philadelphia: F. A. Davis Company. Jonsson, C. O. & Molander, L. (1964). Manual till CVB-Skalan [Manual of the CVB- Scales]. Stockholm: Psykologi Forlagot. Joreskog, K. G. & Sorbom, D. (1993). LISREL 8: User’s reference guide. Chicago: Scientific Software. Kliegl, R & Lindenberger, U. (1993). Modeling intrusions and correct recall in episodic memory: Adult age differences in encoding list context. Journal of Experimental Psychology: Learning. Memory, and Cognition. 19. 617-637. Kyllonen, P. C. (1994) Abstraction. InR. J. Sternberg (Ed.). Encyclopedia of Human Intelligence. Macmillan Publishing Company: New York. Levin, E. D. (1992). Nicotinic systems and cognitive function. Psvchopharmacologv. 108,417- 431. Lindenberger, U. & Baltes, P. B. (1994). Aging and intelligence. In R. J. Sternberg (Ed.), Encyclopedia of Human Intelligence. Macmillan Publishing Company: New York. Lindenberger, U., Mayr, U. & Kliegl, R. (1993). Speed and intelligence in old age. Psychology and Aging. 8, 207-220. Lohman, D. F. (1994) Spatial ability. In R. J. Sternberg (Ed.), Encyclopedia of Human Intelligence. Macmillan Publishing Company: New York. Matarazzo, J. D., Carmody, T. P., & Jacobs, L. D. (1980). Test-retest reliability and stability of the WAIS: A literature review with implications for clinical practice. Journal of Clinical Neuropsychology. 2, 89-105. Mayr, U. & Kliegl, R. (1993). Sequential and coordinative complexity: Age-based processing limitations in figural transformations. Journal of Experimental Psychology: Learning. Memory, and Cognition. 19. 1297-1320. Medlund, Cederlof, Floderus-Myrhed, Friberg & Sorensen. (1976). A new Swedish Twin Registry. Acta Medica Scand. Suppl. no. 600. Morrow, D., Leirer, V. Alteri, P. & Fitzsimmons, C. (1994). When expertise reduces age differences in performance. Psychology and Aging. 9f 1). 134-148. Neary, D., Snowden, J. S., Mann, D. M. Bowen, D. M., Sims, N. R., Northen, B., Yates, P. O. & Davison, M. N. (1986). Alzheimer's disease: a correlative study. Journal of Neurology. Neurosurgery. & Psychiatry. 49. 229-237. 53 Nesselroade, J. R. (1991). Interindividual differences in intraindividual change. In L. M. Collins & J. L. Horn (Eds,), Best Methods for the Analysis o f Change: Recent Advances. Unanswered Questions Future Directions. Washington, D.C.: American Psychological Association. Nordberg, A., Lilja, A., Lundqvist, H., Hartvig, P., Amberla, K., Viitanen, M., Warpman, U., Johnnnsson, M., Hellstrom-Lindahl, E., Bjurling, P., Fasth, K., Langstrom, B,, & Winblad, B. (1992). Tacrine restores cholinergic nicotinic receptors and glucose metabolism in Alzheimer patients as visualized by Position Emission Tomography. Neurobiology of Aging. 13. 747-758. Pedersen, N. L., Plomin, R., Nesselroade, J. R., & McCleam, G. E. (1992). A quantitative genetic analysis of cognitive abilities during the second half of the life span. Psychological Science. 3(6), 346-353. Pedersen, N. L., McCleam, G. E., Plomin, R., Nesselroade, J. R., Berg, S. & DeFairc, U. (1991). The Swedish adoption twin study of aging: an update. Acta Geneticae Medicae et Gemellologiae. 40, 7-20. Pedersen, N. L,, Friberg, L., Floderus-Myrhed, B., McCleam, G. E. & Plomin, R.. (1984). Swedish early separated twins: identification and characterization. Acta Geneticae Medicae et Gemellologiae. 33, 243-250. Riggs, J. E. (1993). Smoking and Alzheimer’ s disease: protective effect or differential survival bias? The Lancet. 342, 793-794. Rosecrans, J. A. (1995). The psychopharmacologicnl basis of nicotine's differential effects on behavior: individual subject variability in the rat. Behavior Genetics. 25, 187-195. Royer, F. L., Gilmore, G. C. & Grulin, J. J. (1984). Stimulus parameters that produce age differences in block design performance. Journal of Clinical Psychology. 40, 1474-1484. Rusted, J. M. & Warburton, D. M. (1992). Facilitation of memory by post-trial administration of nicotine: evidence for an attcntional explanation. Psvchopharmacologv. 108. 452-455. Saltliouse, T. A. & Skovronek (1992). Within-context assessment of age differences in working mcmoiy. Journals of Gerontology: Psychological Science. Saltliouse, T. A. (1992). Reasoning and spatial abilities. In F. I. M. Craik & T. A. Saltliouse IEds.1. The handbook of aging and cognition fpp. 167-211). Hillsdale, NJ: Erlbaum Saltliouse, T. A. (1991). Mediation of adult age differences in cognition by reductions in working memoir and speed of processing. Psychological Science. 2, 345-371. 54 Saltliouse, T. A, (1987). Sources of age-related individual differences in block design tasks. Intelligence. 11. 245-262. Sherwood, N., Kerr., J. S., & Hindmarch, I. (1992). Psycliomotor performance in smokers following single and repeated doses of nicotine gum. Psvchopharmacology. 108.432-436. Sims, N. R., Bowen, D. M., Allen, S. J, Smith, C. C., Neary, D., Thomas, D. J. & Davison, A. N. (1983). Presynaptic cholinergic dysfunction in patients with dementia. Journal of Neurochemistrv. 40. 503-509. Thurstone, L. L. (1938). Primary mental abilities. Chicago: University of Chicago Press. van Duijn, C. M. & Hofman, A. (1991). Relation between nicotine intake and Alzheimer’ s disease. British Medical Journal. 302. 1491-1494. Warburton, D. M. (1992a). Nicotine as a cognitive enhancer. Progress in Neuro- Psvchophnrmacolopv & Biol-Psvchiatry. 16. 181-191. Warburton, D. M. (1992b). Nicotine issues. Psvchopharmacologv. 108. 393-396. Warburton, D. M., Rusted, J. M., & Fowler, J. (1992). A comparison of the attentional and consolidation hypotheses for the facilitation of memoiy by nicotine. Psvchopharmacologv. 108. 443-447. Wechsler, D. A. (1972). Wechsler Adult Intelligence Scale. Manual. New York: Psychological Corporation. Wenk, G. L. (1994). Drugs and intelligence. In R. J. Sternberg fEd.I. Encyclopedia of Human Intelligence. Macmillan Publishing Company: New York. Wesnes, K. & Warburton, D. M. (1984). The effects of cigarettes of varying yield on rapid information processing performance. Psvchophnrmncologv. 82, 338-342. Westerin, P. A. (1969). WIT III Manual. Stockholm: Scandinaviska Test Forlaget. Wiederholt, W. C., Calm, D., Butters, N. M., Salmon, D. P., Kritz-Silvcrstein, D. & Barrett-Connor, E. (1993). Journal of the American Geriatrics Society. 41. 639-647. Yamashita, K, Kobayashi, S., Yamaguchi, S., Kitani, M., & Tsunematsu, T. (1988). Effect of smoking on regional cerebral blood flow in the normal aged volunteers. Gerontology. 34. 199-204. Yantis, S. (1994). Attention. In R. J. Sternberg (Ed.). Encyclopedia of Human Intelligence. Macmillan Publishing Company: New York.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Cognitive functioning and dementia following cancer: A Swedish twin study
PDF
Sexuality, Quality Of Life, And Emotional Distress Following Radical Prostatectomy For Carcinoma Of The Prostate
PDF
Experimental evaluation of a new intervention designed to promote dementia caregivers' acceptance and empathic responding towards care -recipients
PDF
Head injury and dementia: A co-twin control study of Swedish twins
PDF
Associations and mechanisms among attention deficit hyperactivity disorder symptoms, cognitive functioning, and drinking habits
PDF
History of depression, antidepressant treatment, and other psychiatric illness as risk factors for Alzheimer's disease in a twin sample
PDF
Alcohol expectancies and consumption: Age and sex differences
PDF
Mood Induction As An Analog To Depression In Older Adults: Aspects Of Depression And Cognitive Effects
PDF
Cognitive Assessment Of Reactance Using The Articulated Thoughts In Simulated Situations Paradigm
PDF
Articulated thoughts regarding cognitions toward older adults
PDF
Fine motor skills of two- to three-year-old drug exposed children
PDF
Violent environments and their effects on children
PDF
Coping Strategies Of Three Adolescents With Disabilities
PDF
Adolescents' social attitudes: Genes and culture?
PDF
Depression as a risk factor for dementia
PDF
Dementia And The Perception Of Experimental Pain
PDF
Predictors of pretend play in Korean-American and Anglo-American preschool children
PDF
Conflict And Generalized Conflict In Verbal And Motor Responses
PDF
The Emotional, Physiological, And Cognitive Reactions Of Boys And Girls From High-Conflict And Low-Conflict Homes To Simulated Marital Conflict
PDF
An examination of affective modulation, psychopathy, and negative schizotypy in college and community samples
Asset Metadata
Creator
Kasl-Godley, Julia E.
(author)
Core Title
Smoking And Cognitive Functioning In Non-Demented Swedish Twins
Degree
Master of Arts
Degree Program
Psychology
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
Health Sciences, Pharmacology,OAI-PMH Harvest,Psychology, clinical,psychology, developmental,psychology, physiological
Language
English
Contributor
Digitized by ProQuest
(provenance)
Advisor
Gatz, Margaret (
committee chair
), Meyerowitz, Beth E. (
committee member
), Van Horn, John D. (
committee member
)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c18-13193
Unique identifier
UC11357687
Identifier
1380470.pdf (filename),usctheses-c18-13193 (legacy record id)
Legacy Identifier
1380470-0.pdf
Dmrecord
13193
Document Type
Thesis
Rights
Kasl-Godley, Julia E.
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA
Tags
Health Sciences, Pharmacology
psychology, developmental
psychology, physiological