Keywords: chronic conditions, functional magnetic resonance imaging, neuromarkers, physical activity
Abstract
Although research has largely focused on the effects of physical activity (PA) on the brain, less is known about the influence of the brain on engagement in healthy-living behaviors, such as regular PA. In this secondary analysis of a study of brain activity and participation in healthy-living behaviors, we examined relationships between the activation of selected brain networks and PA in persons self-managing chronic conditions. Fifty-eight individuals with chronic conditions underwent functional magnetic resonance imaging while exposed to a protocol consisting of listening to emotion-focused and analytic-focused information and measures of activation of three neuromarkers were obtained: default mode network (DMN), task-positive network (TPN), and ventromedial prefrontal cortex (vmPFC). In an exploratory analysis, we assessed differences in neuromarker activation between two PA levels (representing higher and lower accelerometry-measured PA levels) of 1) moderate-to-vigorous physical activity (MVPA) minutes, 2) metabolic equivalents expended (METs), and 3) daily steps. Results showed positive associations between MVPA and DMN (r = 0.31, P = 0.018), steps and DMN (r = 0.28, P = 0.035), and MVPA and vmPFC (r = 0.29, P = 0.026). No associations were found between the TPN and any of the PA measures. Individuals with high MVPA and METs had higher DMN values compared with those with low MVPA (t = −2.17, P = 0.035) and METs (t = −2.02, P = 0.048). No differences in TPN and vmPFC were found among PA levels. These results suggest that providing health information that activates the emotion-focused brain network may be more useful than analytic-focused information (centered on logic and reasoning) to assist people with chronic conditions to engage in more PA.
NEW & NOTEWORTHY The influence of the brain on engagement in regular physical activity (PA) has not been well studied. We examined relationships between the activation of three neuromarkers and two PA levels in 58 persons self-managing chronic conditions. Findings suggest that individuals who optimally process health-information when the emotional tone is high (Empathic Network; DMN) may engage in more PA compared with individuals who respond to health information when the emotional tone is low (Analytic Network; TPN).
INTRODUCTION
Neuroscience can provide new insights into the mechanisms underlying the links between cognition, behavior, and health. Understanding these links can assist in the development of more effective interventions to assist people to engage in healthy-living behaviors, such as physical activity (PA). Although the positive effects of PA on health have been well established (1, 2), less is known about how cognition can affect PA participation. Prior research has largely focused on the effects of PA on the brain (3, 4), with less focus on how brain functioning affects individuals’ adherence to a PA regimen. How individuals process and respond to information (e.g., patient or public health education) is an important factor influencing health-promoting behaviors. Cognitive processes are greatly involved in understanding and responding to health information, thus, understanding the neural activity underlying those cognitive responses is desirable and may represent a neurobiological explanation for why responses to health information and subsequent behavior differ from individual to individual. The purpose of this study was to examine the relationships between the activation of selected brain networks and two levels of PA in adults with chronic conditions receiving different types of health information. In this paper, the high physical activity group is defined as ≥ 29 moderate to vigorous physical activity (MVPA) minutes per day, a level consistent with the recommended amount of daily PA to achieve a therapeutic effect (5).
Over the past two decades, there is increasing evidence that cognitive and behavioral propensities emerge from the interactions both between and within neural networks rather than in discrete brain regions acting in isolation from each other (6, 7). The theoretical basis for the work in our research center derives from the Opposing Domains hypothesis (6, 8) that suggests that information that is emotion-focused will focally engage brain areas associated with motivation, valuing, and self- and social-referencing, and disengage brain areas associated with task performance and nonsocial reasoning (analytic thinking). Conversely, analytic information engages brain areas associated with task performance and nonsocial reasoning and disengages brain areas associated with motivation, valuing, and self- and social-referencing (emotional/empathic thinking). The Opposing Domains hypothesis is supported by a number of prior studies examining neural processing of health information and subsequent behavior change (9–13). In particular, activity in the medial parietal and medial prefrontal regions in response to health information processing predict the performance of health-promoting behaviors (10, 11). These brain regions are positively associated with social and emotional cognition and are part of what is commonly known as the Default Mode Network (DMN; Empathy Network). Our investigations focus on the dorsal (superior) parts of the DMN, which have been described and termed the Empathy Network (8, 14, 15). In addition to the Empathy Network, another core of the DMN, the ventromedial prefrontal cortex (vmPFC), which is implicated in the valuation of information and rewards (e.g., weighing the benefits vs. barriers for behavior change), has been examined for its association with the prediction of behavior (12, 16).
In contrast to the Empathy Network, the Analytic Network, which is also known as the task-positive network (TPN), involves activation of the lateral parietal and lateral prefrontal cortices of the brain during executive functioning, nonsocial reasoning, logical and scientific reasoning, and inhibitory control (17, 18). Typically, when activation of the Analytic Network is pronounced, activity in the Empathy Network diminishes and vice versa. This pattern of differential activation indicates that the networks are anti-correlated with one other. According to the Opposing Domains hypothesis, we predict that the strength of the anti-correlation of the Empathy and the Analytic Networks represents our brain biomarker of optimal engagement in health-promoting behaviors, such as PA. Thus, we expect that individuals who strongly activate the Empathy Network when the emotional tone of health information is high and strongly activate the Analytic Network when the emotional tone of information is low, will optimally process all health information and engage more in recommended daily MVPA.
This reciprocal inhibitory relationship between these two opposing brain networks has been supported in several studies using functional magnetic resonance imaging (fMRI) (19–21). Since health information tends to vary in the degree to which it is analytic or empathic, the ability of an individual to differentiate analytic processing from emotion/empathic processing (indicated by activation of the respective brain networks) may influence the likelihood of their use of both types of information for optimal engagement in healthy behaviors. This premise is supported by the work of Jack and colleagues (14, 15) that showed that personalized coaching containing both empathic and analytic content, resulted in more goal setting and action-taking by individuals. In particular, a dose effect was found on brain regions associated with the Empathy Network.
Task differentiation, which we define as the ability to differentiate analytic processing from emotion/empathic processing in response to different types of health information (analytic and emotional/empathic) is measured in this study using an fMRI paradigm developed by our team in which we examined task differentiation associated with two large-scale neural networks, the Analytic Network and the Empathy Network, in response to the two types of health information (see Fig. 1). The magnitude of the task differentiation in each of these networks when the information content is analytic (Analytic > Empathy) or emotional (Empathy Network > Analytic Network) are two of our markers for optimal ability to engage in health-promoting activities. The task differentiation values are at a continuous level of measurement, with higher values representing better task differentiation. A third marker is activation of the vmPFC, in which we expect to have high activation in response to emotional information and minimal activation in response to analytic information. Our premise is that individuals who optimally process health information, that is they are equally adept at attending to and responding to health information that comprises both analytic and emotion/empathic components, will be more likely to effectively act on that information. We expect that health information having different tones of emotion or task orientation may have markedly different effects on brain areas that predict people’s actions on that information and, for example, engage in more PA, and that individuals who show the strongest differential neural response to the two different types of information will exhibit more PA.
Figure 1.
Different types of health information processed in different ways in the brain. Lateral and medial views of the brain illustrate different brain areas engaged by analytic Fact Focused health information (blue), and empathic Coping Stories about dealing with chronic illness (orange/yellow). Figure is reprinted from Ref. 7 by permission of Wolters Kluwer Health.
METHODS
Study Design
This study is a secondary analysis using baseline data of three pilot studies associated with a National Institutes of Health-funded Center of Excellence for Self-Management Research (SMART) Center at Case Western Reserve University (22). Each study used a common set of measures to obtain data on selected brain network activity and healthy-living behaviors.
Sample
A sample of 58 adults with a chronic condition were included in this study: 28 people living with human immunodeficiency virus (PLHIV), 16 adults recovering from a cardiac event, and 14 adults with uncontrolled hypertension (HTN). Individuals were recruited by clinic registries and flyers. A screening phone call and medical record review were used to determine eligibility. Eligible participants in the HIV study were ≥ 18 yr of age and HIV+, receiving antiretroviral therapy, at risk for cardiovascular disease based on the Framingham risk score, on a stable dose of statins, and had a recent viral load <400 copies/mL. Individuals were excluded if they were not able to engage in scheduled exercise, had ≥ 150 min of MVPA or 75 min of vigorous exercise per week, had uncontrolled diabetes, or were enrolled in a formal exercise, diet, or weight loss program (23). Eligible participants in the study of adults recovering from a cardiac event were ≥ 40 yr of age, experienced a first cardiac event (myocardial infarction or revascularization), and planned to participate in a cardiac rehabilitation program (24). Individuals were excluded from the study if they experienced cardiac arrest. Eligible participants in the HTN study were Black adults who were ≥ 25 yr of age, had HTN based on a blood pressure >140/80 mmHg, were on at least one anti-HTN medication, and owned a smartphone (25). Individuals were excluded from all pilot studies if they were pregnant, could not speak or write in English, or had a medical history that was contraindicated for functional magnetic resonance imaging (fMRI) (22).
Procedures
This study was approved by the Institutional Review Board at University Hospitals of Cleveland. Written informed consent was obtained before study participation. Eligible participants completed a baseline visit that included collection of anthropometrics and a set of survey questionnaires. Participants in the PLHIV and HTN groups were screened by the Montreal Cognitive Assessment (MoCA) (26) and participants in the cardiac sample were screened by medical chart review to assess cognitive status and to ensure that participants were able to participate in the fMRI scan and study protocol. Participants then completed an fMRI scan using a protocol designed to assess neural processing in response to exposure to different types of self-management health information. Actigraphy monitors were provided to subjects at the visit to take home. Baseline visits ranged from 1–2 h and the fMRI protocol was ∼1 h in length. Participants received a small cash honorarium for completing the baseline visit, actigraphy monitoring, and fMRI protocol.
Measures
Demographic characteristics were collected by self-report. Body mass index (BMI) was collected by a research nurse using a standard laboratory protocol in kg/m2. Handedness was measured by the self-reported Edinburg Handedness Inventory-Short Form (27).
Neuromarkers.
Our research center team designed a 7min imaging protocol (7, 28) to measure the ability to differentiate emotion/empathic processing from analytic processing (6) in response to three different types of health information to promote healthy-living behaviors in persons managing chronic conditions, two conditions of relevance to the current investigation, Analytic (fact focused) and Emotion-focused (coping stories). A third condition containing both analytic and emotion-focused content was not used in this analysis. While undergoing fMRI participants were provided with a series of video clips that included animations of anatomical and physiological explanations of health and disease processes related to self-management of chronic conditions (analytic-focused information) and short stories of individual/families’ experiences in coping with and managing chronic conditions (empathic-focused information). There were 48 video clips of 23 s in length to which subjects were exposed during imaging in the same fixed sequence design of four runs and temporal spacing standardized using Eprime software (v 2.0.10) (7, 28). The fMRI protocol has been previously described (23, 28, 29). Brain scans were performed using a Siemens 3 T Skyra scanner with each scan beginning with a T1-weighted structural MRI sequence to establish an anatomical baseline, followed by four T2-weighted functional task runs described above. The brain scans took approximately 1 h to complete.
Physical activity.
Physical activity was measured using the triaxial Actigraph wGT3X-BT device (Pensacola, FL). Participants were asked to wear the device on their nondominant hip for 7 days (5 weekdays and 2 weekend days). This actigraphy protocol is a standard and reliable approach for PA measurement in adults (30). To be included in the analysis, participants had to wear the device for at least three days (at least 1 weekday and 1 weekend day) and ≥ 6 h per day. Nonwear intervals were ≥ 60 min of wear length and a spike tolerance of 100 counts per minute (31). Intervals that met the wear criteria were analyzed with ActiLife v6.13.3 using a 60-s epoch length, normal activity filter, and a sampling frequency of 30 Hz. The Freedson VM3 (32, 33) algorithm was used to calculate the number of kilocalories burned and the Freedson Adult et al. (34) algorithm was used to calculate the MVPA and METs.
Statistical Analysis
Imaging data were preprocessed using the Washington University in St. Louis program, fidl. First-level analyses were also carried out using fidl. A general linear model (GLM) with assumed Hemodynamic Response Functions (HRFs) was used to estimate the average magnitude of each participant’s response to the video conditions. The rest condition was not explicitly modeled, and so was implicitly captured by the baseline estimate. Voxel-based first-level estimates of response magnitude were entered into a second-level analysis to allow random effects analysis of effects attributable to the population. Values were averaged over voxels in the prespecified regions for the Analytic- and Emotion-focused conditions. These were extracted from the general linear model as average estimates and were then entered in a separate analysis program with the behavioral measures.
This analysis includes individuals with complete actigraphy and fMRI data. Descriptive statistics (mean, standard deviation, range) for continuous variables or frequencies and percentages for categorical variables were calculated on demographic and clinical characteristics, PA, and neuromarker data. Two-tailed independent samples t-tests were used to examine mean differences in the three neuromarkers of interest in two levels of PA groups. The two levels of PA groups were determined using the median-split classifying physical activity as high (≥ 29 MVPA min./day; ≥ 5,024 steps/day; ≥ 1.0991 METs/day) or low (< 29 MVPA min/per day; < 5,024 steps/day; <1.0991 METs per day). The high MVPA threshold is consistent with the recommended guidelines (i.e., ≥30 min per day) (5). Prior to tests of differences, the equivalence of the study group demographic and clinical characteristics was assessed using two-tailed independent samples t tests (continuous variables) or Pearson chi-square (categorical variables). In an exploratory analysis, Spearman’s Rho correlation coefficients were used to examine the relationships among the neuromarkers and physical activity measures. Effect sizes (Cohen’s d) were calculated using IBM SPSS version 28 (Armonk, NY, 2021) and were interpreted as small (0.2), moderate (0.5), and large (0.8) (35). Analyses were conducted with IBM SPSS version 28 (Armonk, NY, 2021). A P value < 0.05 was considered statistically significant.
RESULTS
Characteristics of the Sample
The sample included 58 adults with chronic conditions ranging in age from 25 to 85 yr. The sample was primarily male, Black, single, unemployed, and had some college education. The majority were on Medicaid/Medicare insurance plans. Table 1 displays the demographic and clinical characteristics of the sample. Table 1 also displays the results of the tests for equivalence of the characteristics of the two PA study groups. There were no significant PA group differences in age, gender, race/ethnicity, marital status, employment, insurance type, cognitive functioning (MoCA), BMI, or handedness choice. There was a significant group difference for self-reported education.
Table 1.
Sample demographic and clinical characteristics
| Demographics, n (%) | Whole Group | Low MVPA, min/day | High MVPA, min/day | P Valuea |
|---|---|---|---|---|
| Study sample size | ||||
| Study 1b | 28 (48.3) | 8 (27.6) | 20 (69.0) | |
| Study 2c | 16 (27.6) | 10 (34.5) | 6 (20.7) | |
| Study 3d | 14 (24.1) | 11 (37.9) | 3 (10.3) | |
| Age, yr, means ± SD | 56.9 ± 12.14 | 59.9 ± 13.75 | 54.0 ± 9.65 | 0.065 |
| Gender | ||||
| Female | 24 (41.4) | 14 (48.3) | 10 (34.5) | 0.379 |
| Male | 33 (56.9) | 15 (51.7) | 18 (62.1) | |
| Transgender | 1 (1.7) | 0 (0) | 1 (3.4) | |
| Race/ethnicity | ||||
| African American/Black | 46 (79.3) | 23 (79.3) | 23 (79.3) | 0.580 |
| Asian/Pacific Islander | 1 (1.7) | 0 (0) | 1 (3.4) | |
| White/Angelo (Non-Hispanic) | 11 (19) | 6 (20.7) | 5 (17.2) | |
| Marital status | ||||
| Married | 16 (27.6) | 10 (34.5) | 6 (20.7) | 0.240 |
| Single | 42 (72.4) | 19 (65.5) | 23 (79.3) | |
| Education | ||||
| Did not finish high school | 10 (17.2) | 1 (3.4) | 9 (31.0) | 0.022 |
| High school diploma/GED | 8 (13.8) | 3 (10.3) | 5 (17.2) | |
| Some college/associate degree | 22 (37.9) | 13 (44.8) | 9 (31.0) | |
| 4-yr college degree or higher | 18 (31.0) | 12 (41.4) | 6 (20.7) | |
| Employment | ||||
| Employed | 21 (36.8) | 12 (42.9) | 9 (31.0) | 0.355 |
| Unemployed (includes retirees) | 36 (63.2) | 16 (57.1) | 20 (69.0) | |
| Insurance type (n = 57) | ||||
| Medicaid/Medicare | 37 (64.9) | 15 (53.6) | 22 (75.9) | 0.185 |
| Private | 13 (22.8) | 9 (32.1) | 4 (13.8) | |
| Other | 7 (12.3) | 4 (14.3) | 3 (10.3) | |
| Clinical characteristics | ||||
| Cognitive functioning (MoCA) (n = 41), means ± SD | 25.39 ± 3.44 | 25.26 ± 4.12 | 25.50 ± 2.82 | 0.829 |
| BMI, kg/m2, means ± SD | 31.66 ± 8.65 | 32.76 ± 9.79 | 30.57 ± 7.36 | 0.340 |
| Handedness choice (Edinburg) (n = 30), n (%) | ||||
| Left-handed | 4 (13.3) | 2 (9.5) | 2 (22.2) | 0.443 |
| Mixed handed | 2 (6.7) | 2 (9.5) | 0 (0) | |
| Right-handed | 24 (80) | 17 (81.0) | 7 (77.8) |
n = 58, n = 29. Results from the two-tailed independent samples t tests and Pearson chi-square are displayed. Sample size is 58 unless otherwise specified. Data are presented as n (%) or means ± SD. aCompares low and high physical activity groups based on the median MVPA. bPeople living with HIV (PLHIV). cParticipants recovering from a cardiac event. dParticipants with uncontrolled hypertension.
Table 2 provides the descriptives for the neuromarkers and physical activity measures. All neuromarker ranges were within expected values (DMN values ranged from −0.21 to 0.63; TPN ranged from −0.22 to 0.50; and vmPFC ranged from −0.51 to 0.75) providing some validation for our fMRI neural differentiation protocol. Also providing validation of the fMRI protocol is that the associations displayed in Table 3 among the neuromarkers were as expected, in that the TPN was inversely associated with the DMN and vmPFC, and the DMN and vmPFC had a high positive association. Table 2 also shows the mean values of the physical activity measures for the whole sample (MVPA ranged from 0.57 to 126.67; METs ranged from 1 to 1.55; and daily steps ranged from 971 to 15,205) and for the two PA groups. As shown in Table 3, the associations between MVPA, METs, and daily steps were as expected. METs had a high positive association with MVPA and steps and MVPA had a high positive association with steps.
Table 2.
Neuromarker and physical activity measures
| Variables | Whole Group | |
|---|---|---|
| Neuromarkers | Means ± SD | Range |
| DMN | 0.16 ± 0.18 | −0.21–0.63 |
| TPN | 0.17 ± 0.15 | −0.22–0.50 |
| vmPFC | 0.10 ± 0.24 | −0.51–0.75 |
| Physical activity measures | ||
| MVPA, min/day | 32.75 ± 27.48 | 0.57–126.67 |
| High group (n = 29) | 52.98 ± 24.58 | 31–126.67 |
| Low group (n = 29) | 12.52 ± 9.26 | 0.57–27.67 |
| METs/day | 1.13 ± 0.12 | 1–1.55 |
| High group (n = 29) | 1.21 ± 0.12 | 1.10–1.55 |
| Low group (n = 29) | 1.04 ± 0.03 | 1–1.10 |
| Steps/day | 5,161 ± 2,958 | 971–15,205 |
| High group (n = 29) | 7,334 ± 2,638 | 5,036–15,205 |
| Low group (n = 29) | 2,989 ± 1,041 | 971–5,012 |
n = 58. The median-split method was used to classify physical activity as high (≥29 MVPA min/day; ≥ 5,024 steps/day; ≥1.0991 METs/day) or low (<29 MVPA min/per day; <5,024 steps/day; <1.0991 METs per day). DMN, default mode network; TPN, task-positive network; vmPFC, ventromedial prefrontal cortex; MVPA, moderate-to-vigorous physical activity; METs, metabolic equivalents.
Table 3.
Neuromarker and physical activity indices Spearman’s Rho correlation coefficients
| Variables | TPN |
DMN |
vmPFC |
|||
|---|---|---|---|---|---|---|
| Correlation Coefficient | P Value | Correlation Coefficient | P Value | Correlation Coefficient | P Value | |
| Neuromarkers | ||||||
| DMN | −0.38 | 0.004 | — | — | 0.67 | <0.001 |
| TPN | — | — | −0.38 | 0.004 | −0.26 | 0.046 |
| vmPFC | −0.26 | 0.046 | 0.67 | <0.001 | — | — |
| Physical activity indices | ||||||
| MVPA, min/day | 0.02 | 0.904 | 0.31 | 0.018 | 0.29 | 0.026 |
| METs/day | 0.06 | 0.657 | 0.25 | 0.057 | 0.24 | 0.065 |
| Steps/day | −0.01 | 0.970 | 0.28 | 0.035 | 0.24 | 0.066 |
n = 58. Bold values are significant Spearman correlations at the 0.10 level. DMN, default mode network; METs, metabolic equivalents; MVPA, moderate-to-vigorous physical activity; TPN, task-positive network; vmPFC, ventromedial prefrontal cortex.
Associations between Neuromarkers and Physical Activity
The associations between the neuromarkers and the physical activity measures are displayed in Table 3. Positive moderate associations were found between MVPA and the DMN and vmPFC and between daily steps and the DMN and vmPFC. A significant association was not found between steps and vmPFC; however, the close-to-significant P value suggests further exploration of this relationship. No associations were found among METs and the three neuromarkers. No associations were found between the TPN and any of the PA measures.
Neuromarker Differences between High and Low Physical Activity Groups
Table 4 displays the results of the tests of differences in the neuromarkers between the two PA groups. Higher levels of the neuromarkers indicate greater differentiation of the network. Individuals in the high MVPA and METs groups had significantly higher DMN values compared with those in the low MVPA and METs groups. No differences in the TPN and vmPFC were found between those in the high and low MVPA and METs groups. No differences in any of the neuromarkers were found in the daily steps groups.
Table 4.
Neuromarker differences between high and low physical activity groups
| Variables | DMN |
TPN |
vmPFC |
||||||
|---|---|---|---|---|---|---|---|---|---|
| Means ± SD | t (56), P Value | Cohen’s d | Means ± SD | t (56), P Value | Cohen’s d | Means ± SD | t (56), P Value | Cohen’s d | |
| MVPA, min/day | |||||||||
| High | 0.21 ± 0.16 | t = −2.17, P = 0.035 | −0.57 | 0.16 ± 0.15 | t = 0.48, P = 0.634 | 0.13 | 0.16 ± 0.24 | t = −1.99, P = 0.052 | −0.52 |
| Low | 0.11 ± 0.20 | 0.18 ± 0.14 | 0.04 ± 0.23 | ||||||
| METs/day | |||||||||
| High | 0.20 ± 0.14 | t = −2.02, P = 0.048 | −0.53 | 0.17 ± 0.15 | t = 0.32, P = 0.749 | 0.08 | 0.15 ± 0.24 | t = −2.01, P = 0.089 | −0.45 |
| Low | 0.11 ± 0.21 | 0.18 ± 0.15 | 0.04 ± 0.23 | ||||||
| Steps/day | |||||||||
| High | 0.20 ± 0.16 | t = −1.68, P = 0.100 | −0.44 | 0.16 ± 0.15 | t = 0.66, P = 0.512 | 0.17 | 0.15 ± 0.23 | t = −1.63, P = 0.109 | −0.43 |
| Low | 0.12 ± 0.20 | 0.19 ± 0.14 | 0.05 ± 0.25 | ||||||
n = 58. Results from the two-tailed independent samples t tests are displayed. The median-split method was used to classify physical activity as high (≥29 MVPA min/day; ≥5,024 steps/day; ≥1.0991 METs/day) or low (<29 MVPA min/per day; <5,024 steps/day; <1.0991 METs per day). Bolded values are significant at the 0.05 level. DMN, default mode network; TPN, task-positive network; vmPFC, ventromedial prefrontal cortex; MVPA, moderate-to-vigorous physical activity; METs, metabolic equivalents.
DISCUSSION
This study generated preliminary data on associations between neuromarkers and PA. Our finding of positive associations between two of our measures of physical activity (MVPA and daily steps) and the DMN and vmPFC (emotion-focused network) suggests that optimal processing of emotional-focused health information is associated with greater PA. We did not find any significant associations between the PA measures and the TPN, suggesting that in this sample, differentiation of analytic information was not associated with PA. Comparison of neuromaker activation in the two levels of PA showed a similar finding. Significantly higher levels of DMN differentiation were found in the high MVPA and METs groups than in the low PA groups, with moderate effect sizes. Our findings are consistent with those of Falk and colleagues (36) who found that self-affirmation statements designed to increase self-valuing and decrease perceptions of threat-activated neural activity in the vmPFC and resulted in less sedentary activity.
These findings suggest that individuals may be better able to optimally process health information and act on it when the emotional tone of the information is high (Empathic Network) versus when the emotional tone of the information is low (Analytic Network). Therefore, interventions that present health information to individuals with a high emotional tone (e.g., motivation, valuing, self- and social-referencing) may help individuals engage in more PA. Should our findings be upheld in future larger studies and in other populations, it may signal significant changes in the current content and approach to delivering health information when the aim is to engage individuals to take action on behalf of themselves. Currently, most public and patient health information is predominately analytic in nature. When a person is diagnosed with a chronic condition, such as diabetes, hypertension, asthma, etc., their health education begins with descriptions of the physiology and anatomy underlying the problem, self-care tasks to do at home, and medication and its side effects. Alternatively, our findings suggest that information to help manage the emotions associated with a new or ongoing health problem may be a better approach to garner engagement of an individual in self-management of their condition. Examples of health information that are high in emotional tone are short examples/stories of how others have successfully navigated a similar challenge, garnering social support through family, friends, and support groups, listening and responding to patient concerns, and showing compassion. This is not to say that analytic information about managing a chronic disease is not important; however, the predominance of analytic information and the low attention to emotional support information in health communications may be contributing to the low level of adherence to public health guidelines and engagement in self-management of chronic conditions.
Interpretation of our findings should consider the limitations of the study. Our sample consisted of adults with chronic conditions, thus limiting the generalization of the findings to other populations. More research is needed to replicate and further explore these findings in a larger and more diverse sample of individuals and compare these findings to age-sex-matched healthy populations. Future studies should include a large enough sample to account for confounding variables, such as type of chronic condition, and time since diagnosis, which could have resulted in over- or underestimated PA levels (37) or influenced brain functioning. For example, the cardiac population recovering from a cardiac event included in this study may have been individuals with a high level of intrinsic motivation and eagerness to participate in more PA. In addition, our measures of PA, although objectively measured, had some limitations. We used an actigraphy minimum wear time of ≥ 3 days and ≥ 360 min per day (30) which is shorter than the wear time recommended for some actigraphy studies. We also did not control for the type of PA and structure (free vs. structured living environments). Finally, we are aware that although we named our two categories of PA as high and low PA, the high PA group data represents those exercising at the higher half of this particular sample and does not necessarily meet the usual definition of high PA.
Our findings support further investigation of the theory of Opposing Domains. Our fMRI protocol was validated in this study and the Opposing Domains theory was supported in that there was, as expected, anti-correlation between the TPN and the DMN and vmPFC neuromarkers and high positive correlations between the DMN and the vmPFC. These associations are supported by the Opposing Domains hypothesis (6, 8) that suggests that emotion-focused information will engage emotional/empathic thinking neuromarkers (DMN and vmPFC) and disengage the analytic thinking neuromarker (TPN). Conversely, analytic-focused information will engage the analytic thinking neuromarker and disengage emotional/empathic thinking neuromarkers. Areas of future investigation include explorations of the dose and ratio of empathic and analytic information for optimal engagement of the brain to aid health-promoting behaviors of individuals. Future studies might address whether or not there is a desirable sequence in the delivery of empathic versus analytic information and how this may vary among individuals and health conditions. Studies can also explore how we can enhance optimal task differentiation among the Empathy and Analytic Networks. Finally, because the literature indicates high correlations in adherence levels among different health-promoting behaviors (e.g., PA, healthy eating, sleeping, medication taking) (38, 39), we recommend that future studies assess the relationships between neural activation in response to different types of health information and the performance of additional healthy-living behaviors and in other populations.
We chose the performance of PA as the focal behavior of this study in that it is a key health behavior that is part of the health information provided to almost all patients with chronic conditions, yet remains a difficult behavior for individuals to adopt and maintain. Our results suggest that including information that is more emotion-focused than is currently included in messaging about the importance of PA may improve PA participation. Engagement of the Empathic Network through encouragement of the use of an “exercise buddy,” using small, frequent self-rewards, acknowledging the personal challenges in maintaining an exercise regime, and hearing success tips from others may be useful information to our current approach of citing the health benefits of PA. Our findings support that neuromarker activation, specifically the DMN, does make a difference in PA engagement and is linked to increased PA levels in persons self-managing chronic conditions. Considering these findings, we suggest that future studies replicate and validate our results in larger and more diverse chronic condition samples to help individuals optimally process health information and engage in levels of PA consistent with recommended guidelines.
DATA AVAILABILITY
Data will be made available upon reasonable request.
GRANTS
Research reported in this manuscript was supported by the National Institutes of Health, National Institute of Nursing Research, under Award Numbers P30NR015326 and T32 NR015433.
DISCLOSURES
No conflicts of interest, financial or otherwise, are declared by the authors.
AUTHOR CONTRIBUTIONS
M.L.W., S.M.M., C.H.S., and K.L.W. conceived and designed research; M.L.W., S.M.M., C.H.S., and K.L.W. performed experiments; M.L.W. and S.M.M. analyzed data; M.L.W. and S.M.M. interpreted results of experiments; S.M.M. prepared figures; M.L.W. and S.M.M. drafted manuscript; M.L.W., S.M.M., C.H.S., and K.L.W. edited and revised manuscript; M.L.W., S.M.M., C.H.S., and K.L.W. approved final version of manuscript.
ACKNOWLEDGMENTS
The neuroimaging work made use of the High-Performance Computing Resource in the Core Facility for Advanced Research Computing at Case Western Reserve University.
REFERENCES
- 1. Haskell WL, Blair SN, Hill JO. Physical activity: health outcomes and importance for public health policy. Prev Med 49: 280–282, 2009. doi: 10.1016/j.ypmed.2009.05.002. [DOI] [PubMed] [Google Scholar]
- 2. Lavie CJ, Ozemek C, Carbone S, Katzmarzyk PT, Blair SN. Sedentary behavior, exercise, and cardiovascular health. Circ Res 124: 799–815, 2019. doi: 10.1161/CIRCRESAHA.118.312669. [DOI] [PubMed] [Google Scholar]
- 3. Erickson KI, Hillman C, Stillman CM, Ballard RM, Bloodgood B, Conroy DE, Macko R, Marquez DX, Petruzzello SJ, Powell KE; For Physical Activity Guidelines Advisory COMMITTEE*. Physical activity, cognition, and brain outcomes: a review of the 2018 physical activity guidelines. Med Sci Sports Exerc 51: 1242–1251, 2019. doi: 10.1249/MSS.0000000000001936. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Bherer L. Cognitive plasticity in older adults: effects of cognitive training and physical exercise. Ann N Y Acad Sci 1337: 1–6, 2015. doi: 10.1111/nyas.12682. [DOI] [PubMed] [Google Scholar]
- 5.U.S. Department of Health and Human Services. Physical Activity Guidelines for Americans (Online). Washington, DC: U.S. Department of Health and Human Services, 2018. [Accessed: September 1, 2023]. https://health.gov/sites/default/files/2019-09/Physical_Activity_Guidelines_2nd_edition.pdf [Google Scholar]
- 6. Jack AI, Dawson AJ, Begany KL, Leckie RL, Barry KP, Ciccia AH, Snyder AZ. fMRI reveals reciprocal inhibition between social and physical cognitive domains. NeuroImage 66: 385–401, 2013. doi: 10.1016/j.neuroimage.2012.10.061. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Moore SM, Musil CM, Jack AI, Alder ML, Fresco DM, Webel A, Wright KD, Sattar A, Higgins P. Characterization of brain signatures to add precision to self-management health information interventions. Nurs Res 68: 127–134, 2019. doi: 10.1097/NNR.0000000000000331. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Friedman J, Jack A, Rochford K, Boyatzis R. Antagonistic neural networks underlying organizational behavior. In: Organizational Neuroscience. Bingley: Emerald Group Publishing Limited, 2015, p. 115–141. [Google Scholar]
- 9. Whelan ME, Morgan PS, Sherar LB, Orme MW, Esliger DW. Can functional magnetic resonance imaging studies help with the optimization of health messaging for lifestyle behavior change? A systematic review. Prev Med 99: 185–196, 2017. doi: 10.1016/j.ypmed.2017.02.004. [DOI] [PubMed] [Google Scholar]
- 10. Falk EB, Berkman ET, Mann T, Harrison B, Lieberman MD. Predicting persuasion-induced behavior change from the brain. J Neurosci 30: 8421–8424, 2010. doi: 10.1523/JNEUROSCI.0063-10.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Vezich S, Falk EB, Lieberman MD. Persuasion neurosciece: new potential to test dual-process theories. In: Social Neuroscience: Biological Approaches to Social Psychology, edited by Harmon-Jones E, Inzlicht M.. New York: Routledge, 2015. [Google Scholar]
- 12. Tompson S, Lieberman MD, Falk EB. Grounding the neuroscience of behavior change in the sociocultural context. Curr Opin Behav Sci 5: 58–63, 2015. doi: 10.1016/j.cobeha.2015.07.004. [DOI] [Google Scholar]
- 13. Cooper N, Bassett DS, Falk EB. Coherent activity between brain regions that code for value is linked to the malleability of human behavior. Sci Rep 7: 43250, 2017. doi: 10.1038/srep43250. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Boyatzis RE, Rochford K, Jack AI. Antagonistic neural networks underlying differentiated leadership roles. Front Hum Neurosci 8: 114, 2014. doi: 10.3389/fnhum.2014.00114. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Jack AI, Boyatzis RE, Khawaja MS, Passarelli AM, Leckie RL. Visioning in the brain: an fMRI study of inspirational coaching and mentoring. Soc Neurosci 8: 369–384, 2013. doi: 10.1080/17470919.2013.808259. [DOI] [PubMed] [Google Scholar]
- 16. Falk EB, Bassett DS. Brain and social networks: fundamental building blocks of human experience. Trends Cogn Sci 21: 674–690, 2017. doi: 10.1016/j.tics.2017.06.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Duncan J, Owen AM. Common regions of the human frontal lobe recruited by diverse cognitive demands. Trends Neurosci 23: 475–483, 2000. doi: 10.1016/s0166-2236(00)01633-7. [DOI] [PubMed] [Google Scholar]
- 18. Shulman GL, Corbetta M, Buckner RL, Fiez JA, Miezin FM, Raichle ME, Petersen SE. Common blood flow changes across visual tasks: I. Increases in subcortical structures and cerebellum but not in nonvisual cortex. J Cogn Neurosci 9: 624–647, 1997. doi: 10.1162/jocn.1997.9.5.624. [DOI] [PubMed] [Google Scholar]
- 19. Mars RB, Neubert FX, Noonan MP, Sallet J, Toni I, Rushworth MFS. On the relationship between the “default mode network” and the “social brain”. Front Hum Neurosci 6: 189–189, 2012. doi: 10.3389/fnhum.2012.00189. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Spunt RP, Lieberman MD. The busy social brain: evidence for automaticity and control in the neural systems supporting social cognition and action understanding. Psychol Sci 24: 80–86, 2013. doi: 10.1177/0956797612450884. [DOI] [PubMed] [Google Scholar]
- 21. Tavares P, Lawrence AD, Barnard PJ. Paying attention to social meaning: an FMRI study. Cereb Cortex 18: 1876–1885, 2008. doi: 10.1093/cercor/bhm212. [DOI] [PubMed] [Google Scholar]
- 22. Moore SM, Musil CM, Alder ML, Pignatiello G, Higgins P, Webel A, Wright KD. Building a research data repository for chronic condition self-management using harmonized data. Nurs Res 69: 254–263, 2020. doi: 10.1097/NNR.0000000000000435. [DOI] [PubMed] [Google Scholar]
- 23. Webel AR, Schreiner N, Salata RA, Friedman J, Jack AI, Sattar A, Fresco DM, Rodriguez M, Moore S. The effect of an HIV self-management intervention on neurocognitive behavioral processing. West J Nurs Res 41: 990–1008, 2019. doi: 10.1177/0193945918823347. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Wierenga KL, Fresco DM, Alder M, Sattar A, Moore SM. Preliminary efficacy of an emotion regulation intervention on physical activity and depressive and anxious symptoms in individuals in cardiac rehabilitation. J Cardiovasc Nurs 37: 296–305, 2022. doi: 10.1097/JCN.0000000000000837. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Still CH, Jack AI, Wright KD, Sattar A, Moore SM. Neural processing of health information and hypertension self-management in African Americans. Nurs Res 71: 303–312, 2022. doi: 10.1097/NNR.0000000000000592. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Rossetti HC, Lacritz LH, Cullum CM, Weiner MF. Normative data for the Montreal Cognitive Assessment (MoCA) in a population-based sample. Neurology 77: 1272–1275, 2011. doi: 10.1212/WNL.0b013e318230208a. [DOI] [PubMed] [Google Scholar]
- 27. Veale JF. Edinburgh Handedness Inventory—short form: a revised version based on confirmatory factor analysis. Laterality 19: 164–177, 2014. doi: 10.1080/1357650X.2013.783045. [DOI] [PubMed] [Google Scholar]
- 28. Jones LM, Wright KD, Jack AI, Friedman JP, Fresco DM, Veinot T, Lu W, Moore SM. The relationships between health information behavior and neural processing in African Americans with prehypertension. J Assoc Inf Sci Technol 70: 968–980, 2019. doi: 10.1002/asi.24098. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Still CH, Margevicius SP, Wright JT Jr, Ruksakulpiwat S, Moore SM. A pilot study evaluating the effects of a technology-based and positive psychological training intervention on blood pressure in African Americans with hypertension. J Prim Care Community Health 12: 21501327211056186, 2021. doi: 10.1177/21501327211056186. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Hart TL, Swartz AM, Cashin SE, Strath SJ. How many days of monitoring predict physical activity and sedentary behaviour in older adults? Int J Behav Nutr Phys Act 8: 62, 2011. doi: 10.1186/1479-5868-8-62. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Troiano RP, Berrigan D, Dodd KW, Mâsse LC, Tilert T, McDowell M. Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc 40: 181–188, 2008. doi: 10.1249/mss.0b013e31815a51b3. [DOI] [PubMed] [Google Scholar]
- 32. Sasaki JE, John D, Freedson PS. Validation and comparison of ActiGraph activity monitors. J Sci Med Sport 14: 411–416, 2011. doi: 10.1016/j.jsams.2011.04.003. [DOI] [PubMed] [Google Scholar]
- 33.ActiGraph, LLC. Kcal estimates from activity counts using the potential energy method. Pensacola, FL: ActiGraph, LLC. http://actigraphcorp.com/research-database/kcal-estimates-from-activity-counts-using-the-potential-energy-method/ [2021 Sep 13]. [Google Scholar]
- 34. Freedson PS, Melanson E, Sirard J. Calibration of the Computer Science and Applications, Inc. accelerometer. Med Sci Sports Exerc 30: 777–781, 1998. doi: 10.1097/00005768-199805000-00021. [DOI] [PubMed] [Google Scholar]
- 35. Cohen J. Statistical Power Analysis for the Behavioral Sciences. Hillsdale, NJ: Lawrence Erlbaum Associates, 1988. [Google Scholar]
- 36. Falk EB, O'Donnell MB, Cascio CN, Tinney F, Kang Y, Lieberman MD, Taylor SE, An L, Resnicow K, Strecher VJ. Self-affirmation alters the brain's response to health messages and subsequent behavior change. Proc Natl Acad Sci USA 112: 1977–1982, 2015. doi: 10.1073/pnas.1500247112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Alder M. L., Still C. H., Wierenga K. L., Pignatiello G. A., Moore S. M. (2022). Differences among physical activity actigraphy algorithms in three chronic illness populations. Chronic Illness: 0(0), 17423953221137889, 2022. doi: 10.1177/17423953221137889. [DOI] [PubMed] [Google Scholar]
- 38. Jones H, Edwards L, Vallis TM, Ruggiero L, Rossi SR, Rossi JS, Greene G, Prochaska JO, Zinman B; Diabetes Stages of Change Study. Changes in diabetes self-care behaviors make a difference in glycemic control: the Diabetes Stages of Change (DiSC) study. Diabetes Care 26: 732–737, 2003. doi: 10.2337/diacare.26.3.732. [DOI] [PubMed] [Google Scholar]
- 39. Prochaska JO, Velicer WF, Redding C, Rossi JS, Goldstein M, DePue J, Greene GW, Rossi SR, Sun X, Fava JL, Laforge R, Rakowski W, Plummer BA. Stage-based expert systems to guide a population of primary care patients to quit smoking, eat healthier, prevent skin cancer, and receive regular mammograms. Prev Med 41: 406–416, 2005. doi: 10.1016/j.ypmed.2004.09.050. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Data will be made available upon reasonable request.
