Edited by: Shin Murakami, Touro University-California, USA
Reviewed by: Bogdan O. Popescu, Colentina Clinical Hospital, Romania; Agnes Lacreuse, University of Massachusetts, USA
*Correspondence: Sara B. Festini
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution and reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Sustained engagement in mentally challenging activities has been shown to improve memory in older adults. We hypothesized that a busy schedule would be a proxy for an engaged lifestyle and would facilitate cognition. Here, we examined the relationship between busyness and cognition in adults aged 50–89. Participants (
Everyday conversation frequently touches upon the busyness of daily schedules. People discuss their packed to-do lists and make inferences about the impact of their busy lifestyle on their health and mental function. Often busyness carries a negative connotation, as people tend to complain about their hectic schedules, yet, little scientific work has been done to empirically investigate the construct of busyness and its associations. To fill this gap, the present study: (1) assesses whether busier people tend to have better or worse cognitive performance; and (2) tests whether this relationship with cognition varies with age.
Busyness has been defined as the subjective evaluation of one’s ongoing activity patterns, including reflections about the quantity of one’s unscheduled time and comparisons to what is expected or standard (see Gershuny,
Given the pervasive discussion of busyness in everyday life, it is surprising that few studies have assessed busyness. We note that, unlike engagement, which typically bears a positive connotation, busyness carries a more negative undertone, and, at present, the cognitive associations of a busy lifestyle are empirically unknown. Related literature suggests that busyness either could be beneficial or harmful to cognition. Busyness could be related to increased levels of stress, which can have negative consequences on the brain and cognitive function (i.e., allostatic load, see McEwen,
Alternatively, more positively, busyness could be related to increased effortful engagement at work, home, and in leisure activities, which can have advantageous consequences on neural health and cognition. Recently, several studies experimentally manipulated lifestyle engagement levels and found benefits for intense, sustained engagement. In the Synapse Project, productive engagement groups that learned digital photography or quilting showed improvements in episodic memory relative to receptive control groups that did little new learning (Park et al.,
In addition to these experimental manipulations of engagement, many correlational studies report benefits of high levels of cognitive, social, and physical activities. Benefits include improved cognition, delayed cognitive decline, increased longevity, and reduced risk of various diseases, including dementias (e.g., see Christensen and Mackinnon,
Based on these experimental and correlational findings, if busyness serves as a proxy for intense, sustained lifestyle engagement, then we would predict that greater busyness would be associated with better cognition. Moreover, because busyness has been shown to differ between middle-aged adults and older adults (see Martin and Park,
A total of 330 participants from the Dallas Lifespan Brain Study (DLBS) spanning ages 50–89 were included in this analysis. The DLBS is a large-scale multi-modal assessment of cognition and brain health, structure, and function in healthy adults. The current sample included participants from a highly-screened, elite cohort, as well as a second cohort with more lenient screening criteria. The second cohort was recruited in order to achieve a broader range in variability in various demographic variables, including health, education, and socioeconomic status, and this cohort has participants spanning ages 50–89. This study was approved by the Institutional Review Board at the University of Texas at Dallas and at University of Texas Southwestern Medical Center. All participants provided informed written consent in accordance with the Declaration of Helsinki. See Table
Age group | Female ( |
Male ( |
Education (years) | MMSE | Shipley vocab | |
---|---|---|---|---|---|---|
50–59 | 86 | 52 | 34 | 15.69 | 28.64 | 33.47 |
60–69 | 99 | 62 | 37 | 15.77 | 28.45 | 34.49 |
70–79 | 90 | 55 | 35 | 15.11 | 28.13 | 34.20 |
80–89 | 55 | 32 | 23 | 15.87 | 27.55 | 34.05 |
Total | 330 | 201 | 129 | 15.59 | 28.26 | 34.08 |
A large neuropsychological battery was administered as part of the DLBS. In the present analyses we assessed five core cognitive constructs—processing speed, working memory, episodic memory, reasoning, and crystallized knowledge. We utilized cognitive tasks that loaded well on these constructs, as well as assessments of age, gender, the highest level of education completed (coded into years of education), and busyness.
Busyness ratings were obtained from the MPED Questionnaire (Martin and Park,
Participants viewed two strings of numbers and determined whether they were the same or different (adapted from Salthouse and Babcock,
Participants were given a list with a randomized set of digits. A key at the top of the page displayed nine geometric symbols that corresponded to a digit from 1 to 9. Participants were asked to draw the corresponding symbol below each digit as fast as possible (Wechsler,
Participants viewed an array of “boxes” on a computer screen and had to maintain the location of a blue token in working memory for accurate performance. The trials varied in set size from 3 to 8 boxes, and the dependent measure was the additive inverse of the number of errors committed. This task was from the Cambridge Neuropsychological Test Automated Battery (CANTAB; Robbins et al.,
A white square appeared on the screen at five different spatial locations, one location at a time. Participants had to update and maintain these spatial locations in working memory. The dependent measure was the number of correct spatial locations that were identified (CANTAB; Robbins et al.,
Participants viewed a complex abstract pattern for several seconds and had to select the same pattern out of a possible four choices either simultaneously with the target pattern, immediately following the target pattern, or after a 12 s delay. The dependent measure was the total number of items that were correctly matched (CANTAB; Robbins et al.,
The experimenter read a series of letters and numbers to the participant (e.g., 2-M-7-B). When the experimenter stopped speaking, the participant needed to mentally rearrange the information and to say the numbers in ascending order, followed by the letters in alphabetical order (e.g., 2-7-B-M). The dependent variable was the total number of items correctly answered. This task was from the Wechsler Adult Intelligence Scale (WAIS-III; Wechsler,
Participants verified whether a math problem was solved correctly and then read a word that followed the math problem. After 2–5 of these problems had been presented, participants wrote down all of the target words that they remembered in the order that they were presented (Turner and Engle,
Participants read aloud 12 words that were presented one-at-a-time on a computer screen. Immediately after the word list, participants recalled as many words as possible (CANTAB; Robbins et al.,
The experimenter read a list of 12 words aloud, one word every 1.5 s. The list contained four words in three semantic categories, which were presented in random order. Three different dependent measures were collected: (a) immediate recall, in which participants recalled as many words as possible immediately after hearing them, (b) delayed recall, in which participants recalled as many words as possible after a 20-min delay, and (c) delayed recognition, in which participants listened to the experimenter read a list of 24 words aloud (12 target words, 6 semantically-related foils, and 6 unrelated foils) and judged whether or not the word was originally studied. The delayed recognition test was always given after the delayed recall test (Brandt,
Participants attempted to learn the names of novel, imaginary space aliens in a visual-auditory paired-associate memory test (Woodcock and Johnson,
In this reasoning task, participants viewed visual patterns and selected a piece that best completed the given pattern (Raven et al.,
Out of five alternatives, participants were asked to determine which set of letters did not follow the same pattern as the others (Ekstrom et al.,
This is a computerized CANTAB version (Robbins et al.,
Out of five alternatives, participants selected the word that most closely matched the meaning of the target word (Ekstrom et al.,
Out of four alternatives, participants selected the word that most closely matched the meaning of the target word (Zachary and Shipley,
As part of the DLBS, participants visited the lab to perform cognitive sessions. The cognitive tasks were spaced over 2 days, in a 2–3 h session each day. Participants also completed a battery of surveys at home on an online system.
First, we examined whether busyness varied as a function of age, gender, or education. A bivariate Pearson correlation revealed that busyness decreased with age,
Next, to investigate the relationship between busyness and cognition, we first created constructs of the five cognitive domains by standardizing performance on each task and averaging the
First, we assessed the nature of the relationship of busyness and cognitive performance. We conducted bivariate correlations between busyness and each of the five cognitive constructs. Results consistently revealed significant positive relationships between busyness and cognition, such that busier people tended to have better cognitive performance. Specifically, greater busyness was associated with faster processing speed,
Next, hierarchical linear regressions were run to determine if busyness predicted significant additional variance in cognition that was unexplained by age and education. Across all five analyses, busyness explained significant additional variance in cognition. Notably, busyness had the largest effect on episodic memory, after controlling for age and education,
Unstandardized | Standardized | ||||||
---|---|---|---|---|---|---|---|
Cognitive construct | Predictor | ||||||
Processing speed | −0.043 | 0.004 | −0.489 | 0.276 | 0.276 | <0.001 | |
Education | 0.025 | 0.019 | 0.062 | 0.282 | 0.006 | 0.091 | |
0.203 | 0.066 | 0.148 | 0.303 | 0.021 | 0.002 | ||
Working memory | −0.033 | 0.003 | −0.491 | 0.279 | 0.279 | <0.001 | |
0.033 | 0.014 | 0.110 | 0.296 | 0.016 | 0.006 | ||
0.150 | 0.050 | 0.144 | 0.315 | 0.020 | 0.003 | ||
Episodic memory | −0.015 | 0.004 | −0.237 | 0.074 | 0.074 | <0.001 | |
0.042 | 0.022 | 0.126 | 0.100 | 0.025 | 0.019 | ||
0.271 | 0.064 | 0.275 | 0.173 | 0.073 | <0.001 | ||
Reasoning | −0.032 | 0.004 | −0.413 | 0.202 | 0.202 | <0.001 | |
0.065 | 0.018 | 0.180 | 0.240 | 0.039 | <0.001 | ||
0.166 | 0.061 | 0.140 | 0.259 | 0.018 | 0.007 | ||
Crystallized knowledge | Age | 0.011 | 0.005 | 0.120 | 0.004 | 0.004 | 0.227 |
0.165 | 0.020 | 0.406 | 0.182 | 0.177 | <0.001 | ||
0.184 | 0.072 | 0.131 | 0.198 | 0.016 | 0.011 |
Finally, a series of hierarchical linear regressions were performed to investigate if age by busyness interactions were present, which would suggest that busyness had a different effect as a function of age. In the regression models, the effects of age and busyness were entered, followed by the interaction term, after controlling for education. Both age and busyness were centered before computing the interaction. The age by busyness interaction was not significant for any cognitive construct: processing speed (
This study was conducted to examine if greater busyness was associated with superior or inferior cognition. Analysis of over 300 people from the DLBS revealed that greater busyness was correlated with better cognition, with the largest effects observed for episodic memory. Furthermore, busyness was similarly influential in adults aged 50–89, indicating that cognitive associations of lifestyle engagement were consistent across this age range. Next, we discuss how our results relate to prior literature, offer several potential mechanisms of the observed effects, and outline future lines of research to consider.
Although busyness is frequently discussed in everyday conversation, little empirical work has examined the cognitive repercussions of a busy lifestyle. Consistent with an engagement framework, the present study revealed that higher levels of busyness were associated with better cognition in adults aged 50–89, with the biggest effects observed for episodic memory. Individuals who reported greater day-to-day busyness tended to have better processing speed, working memory, episodic memory, reasoning, and crystallized knowledge, and these relationships persisted after controlling for age. Moreover, hierarchical regressions demonstrated that after accounting for variations in cognitive ability already explained by age and education, busyness accounted for significant additional variance in all cognitive domains. The most pronounced effects for episodic memory parallel the findings from several experimental engagement protocols (Carlson et al.,
We acknowledge that our analyses are correlational, and while it is notable that busyness and cognition are related, we are unable to determine if living a busy lifestyle improves cognition or if smarter individuals are capable of partaking in more activities, resulting in greater levels of busyness. This is a drawback of all correlational studies, which has been noted in many studies of activity levels (e.g., see Scarmeas and Stern,
Our findings are consistent with studies using other measures of engagement. With regard to assessments of activity levels, a myriad of studies have documented significant relationships between activity frequencies and cognitive function (for a review see Fratiglioni et al.,
All of these studies relate cognitive, social, and physical activity levels to current levels of cognition, as was done in our present busyness analyses, yet evidence also exists for a relationship between activity frequencies and maintenance of cognition over time. For instance, an engaged lifestyle and high cognitive performance at midlife predicted high cognitive performance in old age (Schaie,
However, not all prior studies have found such a pattern. For instance, Wilson et al. (
Finally, although our study found that greater busyness predicted better cognition, especially for episodic memory in the laboratory, this does not suggest that all everyday cognitive activities will exhibit similar benefits. Most notably, busyness may negatively impact prospective memory (i.e., remembering to complete tasks in the future), which was not assessed in the present study. Prior work has shown that busier people tend to have poorer medication adherence (Park et al.,
In this study, we also examined whether a significant interaction was present between busyness and age while predicting cognition—that is, whether busyness was differentially related to cognitive performance at different ages. Results indicated no significant interaction for any of our cognitive constructs. Correspondingly, our data suggest that greater busyness was similarly associated with better cognition across adults aged 50–89. This is consistent with Buchman et al. (
We acknowledge that the correlational nature of the present study does not allow us to definitively address mechanisms underlying the relationship between busyness and cognition. It is indeed possible that individuals with better cognition seek out or are able to sustain more busy lifestyles, rather than that high levels of busyness facilitate cognitive function. We also recognize that an additional, unexplored factor could be contributing to the observed association. Nevertheless, given evidence from prior research on learning and engagement training, we consider several possible mechanisms of how busyness could promote cognition. These potential mechanisms can be tested with future work.
First, prior studies have shown that new learning promotes the retention of new neurons in the hippocampus (see Churchill et al.,
Second, busyness could promote the development of
Third, in a similar vein, busyness may promote the development of
Fourth, in accordance with the
Finally, busyness may encourage the reliance on memory strategies and aids that may assist performance. In a prospective memory experiment, Uttl and Kibreab (
We investigated whether busyness was beneficial or detrimental to cognition in adults aged 50–89. Importantly, we document that busier people tend to have better cognition, especially episodic memory. Although correlational in nature, these results are in line with an engagement framework, and they have implications for the usefulness of engagement training programs. Additional experimental work should be conducted to determine if manipulations of busyness influence cognition in a similar manner. Overall, our findings offer encouragement to maintain active, busy lifestyles throughout middle and late adulthood.
SBF designed the study, analyzed the data, interpreted the results, and wrote the manuscript. IMMcD offered suggestions for analyses, helped interpret the results, and also critically edited the manuscript. DCP designed the study, helped interpret the results, and critically edited the manuscript. All authors approve the final version of the manuscript and agree to be accountable for the content of the work.
This study was supported by NIH Grant 5R37AG-006265-29 awarded to DCP. SBF is supported by the Aging Mind Foundation.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Portions of these data were presented at the Psychonomic Society’s 56th Annual Meeting in Chicago (November 2015).
1All effects remained significant when additionally controlling for cohort. Similarly, all effects remained significant when additionally controlling for gender, with the exception that the effect of busyness on processing speed became marginal,
2All interactions remained non-significant after additionally controlling for both cohort and gender.