This Article Abstract Full Text (PDF) Services Download to citation manager Citing Articles Citing Articles via HighWire PubMed PubMed Citation

Effects of Time of Day on Age Differences in Working Memory

^a Rotman Research Institute of Baycrest Centre and University of Toronto, Canada

Robert West, Department of Psychology, University of Notre Dame, Notre Dame, IN 46556-0399 E-mail: west.19{at}nd.edu.

Decision Editor: Toni C. Antonucci, PhD

	Abstract

TOP Abstract Methods Results Discussion Appendix ENDIX References

 
This study investigated the hypothesis that the influence of time of day on the efficiency of working memory is greater for older than younger adults. Groups of younger and older adults performed a working memory task on 4 consecutive days, twice in the morning and twice in the evening. Objective (body temperature) and subjective (alertness ratings) measures of arousal were taken during each session. Temperature increased across the day equally for younger and older adults, whereas alertness ratings were higher in the morning for older adults and in the evening for younger adults. The efficiency of the access and deletion functions paralleled the subjective alertness rating for younger and older adults, and age-related differences in these functions were greater when individuals were tested at nonoptimal times of day. The efficiency of the response inhibition function was similar for younger and older adults and paralleled changes in body temperature.

SINCE the earliest days of experimental psychology, it has been known that time of day (TOD) can dramatically influence the efficiency of cognitive processing. For instance, Ebbinghaus 1964 reported that the number of trials required for participants to learn a series of consonant-vowel-consonant trigrams increased dramatically from late morning to early evening. More recently, other investigators have reported that TOD influences the efficiency of short-term memory (Baddeley, Hatter, Scott, and Snashall 1970), sustained attention (Blake 1967), inhibitory processing (May and Hasher 1998), and semantic activation (Anderson, Petros, Beckwith, Mitchell, and Fritz 1991). The influence of TOD on cognitive function has also been found to interact with developmental (Folkard, Monk, Bradbury, and Rosenthall 1977) and individual (Horne and Ostberg 1977) differences. For example, increases in cognitive efficiency across the day are positively related to variations in measures of physiological arousal (i.e., increases in body temperature) in individuals who report their optimal TOD to be in the evening (Horne, Brass, and Pettitt 1980), whereas for individuals who report their optimal TOD to be in the morning there is an equally strong negative relationship between increases in body temperature and decreases in cognitive efficiency across the day.

Renewed interest in the relationship between cognitive efficiency and TOD has recently surfaced in the cognitive aging literature. The impetus for this research stems from the finding that across the adult life span there is a shift in the self-reported time of peak arousal on the Morningness-Eveningness Questionnaire (MEQ; Horne and Ostberg 1976). This shift reflects a tendency for the optimal TOD to become earlier in the day with advancing age (May, Hasher, and Stoltzfus 1993; Mecacci, Zani, Rocchetti, and Lucioli 1986).

Recent work by May and Hasher 1998 has explored the hypothesis that age-related differences in cognitive function are modulated by TOD within the context of an inhibition-based model of working memory proposed and extensively developed by Hasher, Zacks, and colleagues (Hasher and Zacks 1988; Stoltzfus, Hasher, and Zacks 1996). In this model, inhibitory processes are proposed to support efficient working memory in the following ways: by limiting the access of irrelevant information into working memory, by deleting no longer relevant information from working memory, and by inhibiting prepotent responses (response inhibition). A primary tenet of this model is that the efficiency of inhibitory processes declines as part of the normal aging process, a proposal supported by the findings of several empirical reports over the past decade (for an alternative view, see McDowd 1997). Numerous studies have demonstrated age-related declines in the efficiency of inhibitory processes at each level of the working memory system, that is, access (Connelly, Hasher, and Zacks 1991; Tipper 1991), deletion (Hamm and Hasher 1992; Hartman and Hasher 1991), and response inhibition (Kramer, Humphrey, Larish, Logan, and Strayer 1994; West 1999a).

In two experiments, May and Hasher 1998 found that TOD modulates the magnitude of age-related declines in inhibitory processes supporting the deletion and response inhibition functions of working memory. In one experiment they used the stop-signal paradigm, a task where individuals are required to inhibit an activated response on trials when a stopping signal is presented (for a review, see Logan 1994), to explore the sensitivity of response inhibition to TOD. Age-related differences in stopping efficiency and stopping latency were greater when testing occurred at nonoptimal times (i.e., a 20% difference in stopping efficiency and a 150-ms difference in stopping time between younger and older adults) than when testing occurred at optimal times (i.e., an 11% difference in stopping efficiency and a 112-ms difference in stopping time between younger and older adults; May and Hasher 1998, Experiment 2). In another experiment the sensitivity of the deletion function to TOD was explored. In this experiment older adults demonstrated a priming effect for disconfirmed sentence completions that was greater when testing occurred at nonoptimal times relative to when testing occurred at optimal times. In comparison, the magnitude of priming for target completions was similar for younger and older adults tested at optimal times and greater for younger than older adults tested at nonoptimal times (May and Hasher 1998, Experiment 1). These findings are consistent with the hypothesis that age-related deficits in inhibitory processes supporting the deletion and response inhibition functions of working memory are magnified at nonoptimal times of day (May and Hasher 1998).

In contrast to data indicating that age-related declines in the deletion and response inhibition functions of working memory are modulated by TOD, other data indicate that age and TOD do not interact with respect to the access function of working memory. For example, one study revealed that the magnitude of negative priming was similar for younger adults (M = 17 ms) and older adults (M = 14 ms) tested at their optimal TOD (Intons-Peterson, Rocchi, West, McLellan, and Hackney 1998). In comparison, for those individuals tested at their nonoptimal TOD negative priming was markedly attenuated in the older adults and positive priming was observed in the younger adults. Also, a study of the efficiency of inhibitory processes supporting speeded reading in the face of distraction revealed that the interference effect resulting from the presence of distracting information was independent of TOD (Li, Hasher, Jonas, Rahhal, and May 1998). This finding led Li and colleagues 1998 to propose that inhibitory processes supporting the deletion and response inhibition functions of working memory, but not the access function, are sensitive to TOD.

This explanation seems to account for the positive findings of May and Hasher 1998, who reported an age-related inhibitory deficit in the deletion and response inhibition functions of working memory that was modulated by TOD. The differential effects of TOD on working memory can also account for the negative findings of Li and colleagues 1998 related to the access function of working memory, where age did not interact with TOD in the reading with distraction task. However, this proposal is not completely satisfactory because it does not account for the results of Intons-Peterson and colleagues 1998, who found that the magnitude of the negative priming effect was greater for those individuals tested at optimal than nonoptimal times of day but did not vary between younger and older adults. If one assumes that the inhibitory processes giving rise to negative priming operate at the level of access into working memory, this finding would indicate that access processes are sensitive to TOD, but not age, a proposal inconsistent with the data and conclusions of Li and colleagues 1998.

The impact of TOD on the efficiency of cognitive processes has been interpreted as resulting from shifts in arousal over the course of the day (for a review, see Folkard 1982). This interpretation stems from the finding that body temperature, often taken as a marker of physiological arousal, increases gradually from early morning to the evening hours, peaking at around 20:00 hr, and then decreases at a somewhat faster rate into the early morning hours (Kleitman 1963). The rise in body temperature, from early morning on, closely parallels gains and losses in cognitive efficiency observed over the course of the day across a broad range of tasks (Blake 1967). However, what is sometimes presented as a straightforward relationship between arousal, body temperature, and cognitive performance is complicated by the findings of a number of studies. Some research has revealed a dissociation between circadian variation in body temperature and performance. For example, Horne and colleagues 1980 reported that the increase in body temperature across the day was approximately equal for self-reported morning- and evening-type individuals, even though performance on a simulated production line task declined across the day for morning-type individuals and increased across the day for evening-type individuals. Further, reports of a tendency for older adults to perform better in the morning and younger adults to perform better in the evening (May et al. 1993) can be contrasted with the finding that the circadian pattern of body temperature across the day is virtually identical in younger adults and healthy older adults (Collins, Abdel-Rahman, Goodwin, and McTiffin 1995). On the basis of these findings, it seems that the conclusion of recent work that the shift toward a morning preference in older adults "may echo physiological circadian rhythms" (Intons-Peterson et al. 1998, p. 371) requires further investigation before it can be accepted, because the age-related shift toward a morning preference seems to occur in the absence of changes in the pattern of physiological arousal reflected in measures of body temperature.

The goals of the current study were threefold. Because age-related differences in the impact of TOD on the access, deletion, and response inhibition functions of working memory were observed across a variety of tasks in earlier studies, we sought to explore interactions between age and TOD on these aspects of working memory within a single task. The use of a single task should reduce potential ambiguities that arise when tasks vary in their structural characteristics and operational demands. Also, in light of the apparent dissociation between the age-related shift in morningness-eveningness preference and the age-related stability in the circadian pattern of body temperature, we obtained both subjective (alertness ratings) and objective (body temperature) indices of arousal during test performance from younger and older adults tested in the morning and in the evening on two occasions. These measures facilitated a comparison of self-reported and physiological measures of arousal obtained during the testing sessions and self-reported morningness-eveningness preference measured on the MEQ. Finally, we sought to examine TOD effects in a population-based (with reference to morningness-eveningness preference) sample of younger and older adults. This can be contrasted with the strategy adopted in recent studies of synchrony effects in younger and older adults (Li et al. 1998; May and Hasher 1998). In these studies a high proportion of definite or moderate evening-type younger adults were tested, and a high proportion of moderate or definite morning-type older adults were tested. The potential biasing effect of this sampling procedure becomes apparent when one considers that the majority of younger adults are typically classified as neutral types, whereas the majority of older adults are typically classified as morning types. Therefore, the use of such a selection strategy means that in recent studies the samples of younger adults probably did not accurately reflect the younger adult population, whereas the samples of older adults were more representative of the older adult population.

In the present study, groups of younger and older adults performed a choice reaction time task that varied the demands placed upon working memory by requiring individuals to identify and respond to the spatial location of a target (immediate response) or the location of the target presented in the previous display (1-back response; West 1999b). For half the blocks of trials, a to-be-ignored distractor was presented with the target. From the 1-back conditions of the task one can obtain measures of the efficiency of the access, deletion, and response inhibition functions of working memory. A measure of the efficiency of the access function can be obtained by considering the impact of the distractor in the 1-back condition on the number of nonintrusion errors (where the response represents neither the location of the current nor previous target) because there should be no effect of the distractor when inhibitory processes limiting access into working memory perfectly filter irrelevant information. If access processes are sensitive to TOD, the effect of the distractor in the 1-back conditions should be greater when testing occurs at nonoptimal times than when testing occurs at optimal times. A measure of the deletion function can be obtained by examining performance in the 1-back condition when the target is presented in isolation, because the 1-back condition requires the continuous deletion of no longer relevant information from working memory in support of task performance. If the deletion function is sensitive to TOD, performance in the 1-back condition when the target is presented in isolation should be reduced at nonoptimal times of day. Finally, an index of the response inhibition function can be obtained by considering intrusion errors where the individual responds with the location of the current target instead of the location of the target 1-back in time, presumably failing to inhibit a prepotent tendency to respond with the location of the current target (West 1999b). If the response inhibition function is sensitive to TOD, the number of intrusion errors should increase from optimal to nonoptimal times of day. In all cases these effects should be greater for older adults than younger adults if sensitivity to TOD increases as part of the aging process.

	Methods

TOP Abstract Methods Results Discussion Appendix ENDIX References

 
Participants
Twenty younger adults and 20 older adults participated in the study. Individuals who began testing in the morning (10 younger and 10 older adults) or evening (10 younger and 10 older adults) were matched for years of age (see Table 1 , Fs < 1). Younger and older adults had attained similar levels of formal education (F < 1), and there was no effect of TOD on the level of educational attainment, F(1, 36) = 1.25, p > .25, MSE = 5.77. The older adults scored higher on the vocabulary subtest of the Wechsler Adult Intelligence Scale-Revised than the younger adults, F(1, 36) = 4.59, p < .04, MSE = 28.26. Vocabulary performance was not influenced by TOD, F(1, 36) = 1.56, p > .20, MSE = 28.26, and the effects of age and TOD did not interact (F < 1). Older adults scored higher than younger adults on the MEQ, F(1, 36) = 22.96, p < .001, MSE = 92.83, indicating an age-related shift toward a morning preference. There was no effect of TOD or interaction of TOD and age on the MEQ (Fs < 1).

Materials
Four-box task.
Four conditions were represented in the task. In the target-isolated immediate response condition, a target (a smiling face icon) appeared in one of four boxes displayed on a computer screen, and the individual was instructed to press a key on a response box mapped to the four display locations as quickly and accurately as possible. In the target plus distractor immediate response condition, the target appeared in one of the four boxes, and a distractor (a star-like icon) appeared in the box to the left or right of the one occupied by the target. For this condition the individual was instructed to ignore the distractor and respond with the location of the target. In the target-isolated 1-back condition, a target appeared in one of the four boxes and the individual was instructed to remember the location of the current target and respond with the location of the previous target. Finally, in the target plus distractor 1-back condition, a target appeared in one of the four boxes and a distractor in one of the adjacent boxes. The individual was instructed to remember the location of the current target, ignore the distractor, and respond with the location of the previous target.

These four conditions were presented in a constant order of increasing difficulty (target-isolated immediate response, target plus distractor immediate response, target-isolated 1-back, target plus distractor 1-back) in four blocks of trials so that participants would have an opportunity to adapt to the increasing demands of the task in a gradual manner. Each block consisted of 50 trials presented in a quasi-random order where targets and distractors could not appear in the same location(s) on consecutive trials. In the 1-back conditions the initial target or target and distractor were presented in one or two of the four boxes for 2 s and no response was required. This served to establish a response for the second display in the 1-back condition. Before beginning each block of trials, the individual received eight practice trials with the stimulus display and task requirements for that block. During practice the computer beeped if an error occurred. Following each response the current target or target and distractor were erased from the screen and four empty boxes were presented for 200 ms. After this interval the target or target and distractor for the next trial appeared in one or two of the boxes.

The display consisted of four unfilled boxes measuring 22 mm x 18 mm outlined by two white rectangles presented on a black background. These boxes were separated by 4 mm, making the entire display 100 mm x 18 mm. The boxes were mapped to four keys of a response box. The middle and index fingers rested upon one of the four keys. On each trial a target (i.e., a smiling face icon; ASCHII character 001) measuring 3 mm x 5 mm appeared inside one of the boxes. In the distractor conditions, a distractor (i.e., a star-like figure; ASCHII character 015) appeared in the box to the right or left of the target. Individuals were allowed to adjust viewing distance to establish optimal viewing conditions of the target.

Alertness rating scale.
A measure of subjective alertness was taken at the beginning of each testing session and then repeated at 30-min intervals over the course of the 2-hr session. For these ratings participants were given an 8 1/2 x 11-in sheet of paper folded in half. In the middle of the folded sheet a 100-mm line was presented bound by the words VERY SLEEPY or VERY ALERT presented 5 mm below the ends of the line. Above the line instructions appeared: "Please rate your level of alertness at this moment in time by placing a mark on the line below." After a judgment of alertness was made the experimenter turned the sheet over so that the participant could not see the placement of the mark when making the next judgment of alertness. We derived rating of alertness by measuring the distance from the left edge of the line (i.e., very sleepy) to the mark provided by the participant. We then derived the mean and standard deviation of these ratings over the course of the testing session to provide an index of the average level of alertness during that day's testing session and the degree of variability of alertness over the course of the session.

Oral temperature.
Oral temperature was taken at the beginning and end of each testing session with a digital thermometer (FILAC F-1010; Sherwood Medical, St. Louis, MO). We then averaged these measurements to provide a mean temperature level for the morning and evening testing sessions.

	Results

TOP Abstract Methods Results Discussion Appendix ENDIX References

 
Age, TOD, and Arousal
The proportion of younger and older adults classified as definite evening, moderate evening, neutral, moderate morning, or definite morning in the current sample is presented in Table 2 . As in previous research, the majority of younger adults were classified as neutral types, and the majority of older adults were classified as moderate morning types (May and Hasher 1998). A chi-square test of this distribution against the expected distribution for a group of this size based upon the normative data of May and Hasher 1998 was not significant, ² (4, n = 40) = 7.16, p > .12, indicating that the distribution of the current sample did not differ from published data including a large number of younger and older adults.

Age, TOD, and Working Memory
The total number of errors committed by younger and older adults in the morning and evening testing sessions for the four conditions of the four-box task are presented in Fig. 1. The log transforms of these data were submitted to a 2 (age) x 2 (TOD) x 2 (distraction: target isolated or distractor present) x 2 (response: immediate or 1-back response) analysis of variance (ANOVA; see Appendix, Note 1). As in previous research (West 1999b), more errors were made when a distractor was present in the display (M = 6.68) than when the target was presented in isolation (M = 6.56), F(1, 38) = 14.41, p < .001, MSE = .334, and errors were more frequent in the 1-back condition (M = 6.86) than in the immediate response condition (M = 6.38), F(1, 38) = 148.47, p < .001, MSE = 1.27. Also, the effect of the distractor was greater in the 1-back condition (isolated M = 6.78, distractor M = 6.94) than in the immediate response condition (isolated M = 6.33, distractor M = 6.42), F(1, 38) = 5.61, p < .03, MSE = .380. The only other effect to approach significance was the Age x TOD x Distraction interaction, F(1, 38) = 3.68, p < .07, MSE = .202, with the effect of the distractor magnified in older adults when they were tested in the evening.

View larger version (21K): [in this window] [in a new window]   Figure 1. Average number of errors for younger and older adults as a function of time of day, the presence or absence of a distractor, and an immediate or 1-back response in the four-box task.

Age-related declines in the access and deletion functions of working memory were assessed in a 2 (age) x 2 (distraction) x 2 (TOD) ANOVA where the number of nonintrusion errors was included as the dependent variable. Nonintrusion errors were more frequent in older adults (M = 1.86) than in younger adults (M = 1.30), F(1, 38) = 4.29, p < .05, MSE = 2.86, and when a distractor was present in the display (M = 1.84) relative to when the target was presented in isolation (M = 1.32), F(1, 38) = 30.10, p < .001, MSE = .35. More importantly for the purposes of the current study, the Age x TOD interaction was significant, F(1, 38) = 5.11, p < .03, MSE = .292, with the number of nonintrusion errors increasing for older adults from the morning (M = 1.72) to the evening (M = 2.00) and decreasing for younger adults from the morning (M = 1.36) to the evening (M = 1.25). The TOD x Distractor was significant (Table 4 ), F(1, 38) = 4.69, p < .04, MSE = .223. The decomposition of this interaction revealed that the effect of age was not significant and did not interact with TOD when the target was persented in isolation (Fs < 1), whereas the number of errors increased from the morning to the evening for older adults and decreased from the morning to the evening for younger adults when a distractor was present in the display, F(1, 38) = 11.65, p < .001, MSE = .223.

Age-related decline in the response inhibition function of working memory was assessed in a 2 (Age) x 2 (Distraction) x 2 (TOD) ANOVA where the number of intrusion errors was entered as the dependent measure. The number of intrusion errors increased from the morning (M = .79) to the evening (M = 1.10), F(1, 38) = 6.28, p < .02, MSE = .621. The Age x TOD interaction was not significant (F < 1), with younger adults (morning M = .77, evening M = .98) and older adults (morning M = .80, evening M = 1.22) both demonstrating an increase in the number of intrusion errors from the morning to the evening. The main effects of distraction and related interactions were not significant (Fs < 1).

TOD effects were not observed in the response latency data for younger or older adults because no main effects or interactions involving this factor reached significance (all Fs < 1.33, ps > .25). Response latency was slower for older adults (M = 1047 ms) than younger adults (M = 632 ms), F(1, 38) = 33.58, p < .001, MSE = 410410; was slower in the 1-back condition (M = 1072 ms) than the immediate response condition (M = 608), F(1, 38) = 64.67, p < .001, MSE = 2514202.07; and was slower when a distractor was present (M = 898 ms) than when the target was presented in isolation (M = 782 ms), F(1, 38) = 52.30, p < .001, MSE = 20440.06. The Age x Distraction x Response interaction (see Table 5 ), F(1, 38) = 5.98, p < .02, MSE = 15587.72, and three of the subordinate interactions were significant. For Age x Response, F(1, 38) = 9.43, p < .005, MSE = 266651.77; for Age x Distraction, F(1, 38) = 8.38, p < .007, MSE = 20440.11; and for Distraction x Response, F(1, 38) = 19.37, p < .001, MSE = 15587.72. This interaction reflected the tendency for the age-related increase in response latency to be greater in the 1-back than immediate response condition and for this difference to be enhanced when a distractor was present in the display.

	Discussion

TOP Abstract Methods Results Discussion Appendix ENDIX References

Our primary goal in the present study was to examine the hypothesis that TOD has a greater impact upon the cognitive efficiency of older adults than that of younger adults (May and Hasher 1998). More specifically, we sought to determine whether age-related declines in the efficiency of inhibitory processes supporting the access, deletion, and response inhibition functions of working memory were perturbed when older adults were tested at nonoptimal times of day.

We obtained an index of the efficiency of the access function of working memory by considering the impact of the distractor in the 1-back condition of the choice reaction time task. On the basis of previous research (West 1999b), we had predicted that if aging is associated with a decline in the efficiency of inhibitory processes supporting the access function, the impact of the distractor would be greater for older than younger adults. Furthermore, age-related differences in the influence of a distractor were expected to be greater when testing occurred at nonoptimal times of day—defined by scores on the MEQ and alertness rating—than when testing occurred at optimal times of day. These hypotheses were supported because the effect of the distractor was greater for older adults than younger adults and the magnitude of this difference was greater at nonoptimal than optimal times of day. This finding can be contrasted with those of recent studies where the effects of age and TOD did not interact with regard to the efficiency of inhibitory processes supporting the access function of working memory (Intons-Peterson et al. 1998; Li et al. 1998). One possible reason for the divergence of the current data from previous findings may be the increased demands placed upon executive control processes supporting working memory in the current study that may not have been present in the studies of Li and colleagues and Intons-Peterson and colleagues. In the current study individuals were required to simultaneously inhibit the distractor, encode the identity of the target, and execute a motor response for the location of the previous target. In comparison, in the negative priming task used by Intons-Peterson and colleagues 1998 individuals were required only to ignore a distractor and respond with the identity of the target. Some evidence in support of this proposal may be the fact that age, TOD, and the presence of a distractor did not interact in the immediate response condition in the present study, where there was no need to maintain the identity of a target or a response to a target from the previous display.

When the full data set was considered, the number of nonintrusion errors made when the target was presented in isolation did not differ between younger and older adults and was not influenced by TOD. This finding is surprising given previous work demonstrating that older adults were more susceptible to nonintrusion errors in the target-isolated 1-back condition (West 1999b). Further analysis of these data for only Days 1 and 3 revealed that on Day 1 the number of nonintrusion errors was greater for older adults than younger adults and this difference was greater when individuals were tested at nonoptimal times of day. In comparison, on Day 3 of testing, younger adults and older adults committed similar numbers of nonintrusion errors and there was no effect of TOD. The findings from Day 1 are consistent with the work of May and colleagues (May et al. 1993; May and Hasher 1998) indicating that deficits in the deletion function of working memory are modulated by TOD in older adults. However, the data from Day 3 indicate that age-related deficits in the deletion function of working memory are attenuated with a relatively limited amount of practice, consistent with the data of West 1999b. The failure of the Age x TOD x Day interaction to remain significant following log transformation of the data, which serves to reduce the degree of positive skewing of the distribution, may indicate that this effect was primarily driven by a limited number of older adults who performed rather poorly in the evening on Day 1 of testing and who regularized their performance by Day 3 of testing. This proposal is consistent with other recent work indicating that age-related differences in diversity, or interindividual variability, are eliminated after limited amounts of practice in a task placing similar demands upon working memory as that used in the current study (West, Murphy, Armilio, Craik, and Stuss in press).

The number of intrusion errors was similar and increased uniformly across the day, paralleling the body temperature data, for older and younger adults suggesting that the response inhibition function of working memory was generally unaffected by the aging process in this sample of individuals. This finding is consistent with West 1999b, where older and younger adults made similar numbers of intrusion errors in the 1-back condition of this task when possible effects of task order were controlled. However, the lack of an effect of aging on response inhibition in the current and previous studies is inconsistent with the findings of other work. May and Hasher 1998 found that older adults were less likely than younger adults to stop an initiated response in the Stop Signal Paradigm and that this difference was modulated by TOD. West 1999a reported an age-related increase in the number of intrusion errors in the Stroop task across three experiments, but in a later study (West and Alain 2000) using the same paradigm as that used in Experiment 3 of West 1999a did not find age-related differences in the number of intrusion errors. On the basis of inconsistencies observed across these studies, more evidence seems necessary before a definitive conclusion regarding the effect of age on the response inhibition functions of working memory can be made.

The findings of the current study indicate that the efficiency of the access, deletion, and response inhibition functions of working memory was modulated by TOD and that for the access and deletion functions this effect was greater for older adults than younger adults (May et al. 1993; May and Hasher 1998). The present study extends the dissociation between subjective (MEQ and alertness ratings) and objective (body temperature) measures of arousal to include younger and older adults. Older adults were more likely to be classified as morning types than evening types and to report being more alert in the morning than evening; younger adults were typically classified as neutral types and reported being less alert in the morning than in the evening. In comparison, there was a similar rise in body temperature from the morning to the evening for both younger adults and older adults. The efficiency of the access and deletion functions of working memory paralleled reports of alertness in younger and older adults, whereas the efficiency of response inhibition paralleled changes in body temperature. Together these findings indicate that care must be taken when investigators are considering the effects of arousal on cognitive performance in younger and older adults, because subjective and objective measures of arousal appear to index physiological mechanisms that are differentially coupled to the efficiency of the access, deletion, and response inhibition functions of working memory.

	Acknowledgments

 
This research was funded by grants from the James S. McDonnell Foundation awarded to F. I. M. Craik (96-53) and D. T. Stuss and colleagues (98-31) and Grant AG13845-01 from the National Institute of Aging awarded to L. L. Jacoby and F. I. M. Craik.

Received for publication July 6, 1999. Accepted for publication October 2, 2000.

	Appendix ENDIX

TOP Abstract Methods Results Discussion Appendix ENDIX References

 
Notes

1. Inspection of the error data revealed that the distributions for younger and older adults were positively skewed, a condition that could serve to inflate the error variance in the ANOVAs and obscure relevant main effects and interactions. In light of this, all analyses (except where noted) were performed on the log transform of the number of errors, and mean conditional values are reported in log units.
   
2. Data from Days 1 and 3 were included in this analysis so that the same individuals would be tested in the morning and in the evening on both days.
   

	References

TOP Abstract Methods Results Discussion Appendix ENDIX References

 

• Anderson M. J., Petros T. V., Beckwith B. E., Mitchell W. W., Fritz S., 1991. Individual differences in the effect of time of day on long-term memory access. American Journal of Psychology 104:241-255.
• Baddeley A. D., Hatter J. E., Scott D., Snashall A., 1970. Memory and time of day. Quarterly Journal of Experimental Psychology 22:605-609.
• Blake M. J. F., 1967. Time of day effects on performance in a range of tasks. Psychonomic Science 9:349-350.
• Collins K. J., Abdel-Rahman A., Goodwin J., McTiffin L., 1995. Circadian body temperature and the effects of a cold stress in elderly and young subjects. Age and Ageing 24:485-489. [Abstract/Free Full Text]
• Connelly S. L., Hasher L., Zacks R. T., 1991. Age and reading: The impact of distraction. Psychology and Aging 6:533-539. [Medline]
• Ebbinghaus H., 1964. Memory: A contribution to experimental psychology (H. A. Rugen & C. E. Bussenius, Trans.) Dover. (Original work published 1885), New York.
• Folkard S., 1982. Circadian rhythms and human memory. Brown F. M., Graeber R. C., , ed.Rhythmic aspects of behavior 241-272. Erlbaum, Hillsdale, NJ.
• Folkard S., Monk T. H., Bradbury R., Rosenthall J., 1977. Time of day effects in school children's immediate and delayed recall of meaningful material. British Journal of Psychology 68:45-50.
• Hamm V. P., Hasher L., 1992. Age and the availability of inferences. Psychology and Aging 7:56-64. [Medline]
• Hartman M., Hasher L., 1991. Aging and suppression: Memory for previously relevant information. Psychology and Aging 6:587-594. [Medline]
• Hasher L., Zacks R. T., 1988. Working memory, comprehension, and aging: A review and a new view. Psychology of Learning and Motivation 22:122-149.
• Horne J. A., Brass C. G., Pettitt A. N., 1980. Circadian performance differences between morning and evening "types.". Ergonomics 23:29-36. [Medline]
• Horne J. A., Östberg O., 1976. A self-assessment questionnaire to determine morningness-eveningness in human circadian rhythms. International Journal of Chronobiology 4:97-110. [Medline]
• Horne J. A., Östberg O., 1977. Individual differences in human circadian rhythms. Biological Psychology 5:179-190. [Medline]
• Intons-Peterson M. J., Rocchi P., West T., McLellan K., Hackney A., 1998. Aging, optimal testing times, and negative priming. Journal of Experimental Psychology: Learning, Memory, and Cognition 24:362-376.
• Kleitman N., 1963. Sleep and wakefulness University of Chicago Press, Chicago, IL.
• Kramer A. F., Humphrey D. G., Larish J. F., Logan G. D., Strayer D. L., 1994. Aging and inhibition: Beyond a unitary view of inhibitory processing in attention. Psychology and Aging 9:491-512. [Medline]
• Li K. Z., Hasher L., Jonas D., Rahhal T. A., May C. P., 1998. Distractibility, circadian arousal, and aging: A boundary condition?. Psychology and Aging 13:574-583. [Medline]
• Logan G. D., 1994. On the ability to inhibit thought and action: A user's guide to the stop signal paradigm. Dagenbach D., Carr T. H., , ed.Inhibitory processes in attention, memory, and language 189-240. Academic Press, San Diego, CA.
• May C. P., Hasher L., 1998. Synchrony effects in inhibitory control over thought and action. Journal of Experimental Psychology: Human Perception and Performance 24:363-379.
• May C. P., Hasher L., Stoltzfus E. R., 1993. Optimal time of day and the magnitude of age differences in memory. Psychological Science 4:326-330.
• McDowd J. M., 1997. Inhibition in attention and aging. Journal of Gerontology: Psychological Sciences 52B:P265-P273. [Abstract]
• Mecacci L., Zani A., Rocchetti G., Lucioli R., 1986. The relationship between morningness-eveningness, aging, and personality. Personality and Individual Differences 7:911-913.
• Stoltzfus E. R., Hasher L., Zacks R. T., 1996. Working memory and aging: Current status of the inhibition view. Richardson J. T. E., , ed.Working memory and human cognition 66-77. Oxford University Press, New York.
• Tipper S. P., 1991. Less attentional selectivity as a result of declining inhibition in older adults. Bulletin of the Psychonomic Society 29:45-47.
• West R., 1999a. Age differences in lapses of intention in the Stroop task. Journal of Gerontology: Psychological Sciences 54B:P34-P43. a. [Abstract]
• West R., 1999b. Visual distraction disrupts working memory processes and gives rise to goal neglect in younger and older adults. Memory and Cognition 27:1064-1072. b. [Medline]
• West R., Alain C., 2000. Age-related decline in inhibitory control contributes to the increased Stroop effect observed in older adults. Psychophysiology 37:179-189. [Medline]
• West, R., Murphy, K. J., Armilio, M. L., Craik, F. I. M., & Stuss, D. T. (in press). Lapses of intention and performance variability reveal age-related increases in fluctuations of executive control. Brain and Cognition.

This article has been cited by other articles:

				K. Pillemer and J. J. Suitor Making choices: a within-family study of caregiver selection. Gerontologist, August 1, 2006; 46(4): 439 - 448. [Abstract] [Full Text] [PDF]

				B. R. Levy Mind Matters: Cognitive and Physical Effects of Aging Self-Stereotypes J. Gerontol. B. Psychol. Sci. Soc. Sci., July 1, 2003; 58(4): P203 - 211. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Services Download to citation manager Citing Articles Citing Articles via HighWire PubMed PubMed Citation