Inverse Associations Between Parasympathetic Activity and Cognitive Flexibility in African Americans: Preliminary Findings

Abstract

Previous research suggests that vagal activity and executive function (EF) are positively associated. However, existing data exploring the relationship between HRV and EF remains limited. Though Blacks may have higher HRV, they remain underrepresented in previous research examining HRV and EF. The current study aimed to determine the relationship between EF and HRV in a sample of 92 Black undergraduates (Mean age=20.32, SD=2.28). Participants wore an 6-lead ambulatory electrocardiographic impedance monitoring system to obtain the root mean square of interbeat interval differences (rMSSD) and Cardiac Sympathetic Index. After baseline autonomic activity assessment, participants completed the Berg Card Sorting Test. Utilizing hierarchical regression analyses, HRV was negatively associated with correct responses (Beta = −0.40, SE= 0.04, p=0.01) and categories experienced (Beta= −0.37, SE=0.01, p=0.01), and positively associated with total errors (Beta = 0.39, SE =0.04, p=0.01). To further elucidate these findings, participants were assigned to coinhibition (n=12), parasympathetically dominant (n=34), sympathetically dominant (n=35), or coactivation (n=11) autonomic space subgroups. Participants in the sympathetically dominant subgroup completed more categories (M=6.86, SD=2.13) and committed fewer errors (M=30.63, SD=11.53) than their parasympathetic counterparts (M=5.74, SD=2.44; M=43.29, SD=18.83, respectively). This study suggests that a state of sympathetic arousal immediately prior to the administration of an EF task, may aid in better task performance.

Introduction

Heart rate variability (HRV) is an indicator of autonomic activity (Perini & Veicsteinas, 2003) and a psychophysiological parameter of cardiovascular health in both clinical and research domains (Acharya et al., 2006; Thayer, Yamamoto, & Brosschot, 2010). Specifically, HRV is the variation in time between each heartbeat. Each heartbeat is in response to environmental and internal demands, supported by parasympathetic and sympathetic branches of the autonomic nervous system. This process allows for the autonomic nervous system to modulate heart rate and cardiac activity. Examining the variability in heart rate provides an indicator of how the heart responds to shifts in parasympathetic and sympathetic balance, based on stimuli or bodily regulation.

Low levels of HRV are associated with negative cardiovascular health outcomes, such as myocardial infarction (Buccelletti et al., 2009), sudden cardiac death (Sessa et al., 2018), heart failure (Casolo et al., 1989), and inflammation (Cooper et al., 2015). Ultimately, previous research posits HRV as a potential proxy of neurophysiological mechanisms involved in innate inflammatory modulation (Williams et al., 2019). Thus, HRV may serve as a signal for various biobehavioral health conditions, including disease severity and mental health (Beauchaine & Thayer, 2015). Previous research has sought to understand the biobehavioral correlates of HRV, unearthing prefrontal and subcortical neural substrates that have neuroanatomical overlap with various neurocognitive processes (Jennings et al., 2015). As early as the seminal research conducted by Lacey and Lacey (1958), researchers have sought to establish an empirical connection between autonomic activity and executive function (EF). Overall, researchers reported HRV suppression during EF tasks (Lacey and Lacey, 1958; Richards and Casey, 1991, Vincent, Craik and Furedy, 1996). Despite this association between EF task performance and HRV suppression, limited studies have sought to utilize resting HRV as a correlate of neurocognitive performance. This is imperative, as HRV may be able to indicate not only autonomic health, but also have implications in decision making processes, including cognitive flexibility or the ability to adjust to a changing stimuli in the environment.

Overall, previous research presents an inconsistent relationship between resting HRV and EF. Hansen et al (2003) was one of the first studies to utilize HRV as a predictor variable of neurocognitive function, reporting a positive association between HRV, EF, and attention. Moreover, previous studies have identified an association between HRV and inhibition and neurocognitive flexibility tasks (Hovland et al., 2012). However, previous studies examining these variables have also rendered controversial results. In a large older sample of middle-aged participants, baseline HRV was not associated with EF after statistical correction for demographic and biological covariates (Kimhy et al., 2013). Upper quartiles of HRV were associated with better EF (e.g. cognitive flexibility and inhibition) when compared to their lowest quartile counterparts in large studies (Hazzouri et al., 2017). It should be noted that Hazzouri and colleagues (2017) reported these findings with the standard deviation of the interbeat intervals, but did not find this comparison for root Mean Square of successive differences in the same sample. Further, previous research suggests HRV is associated with neurocognitive processes that involve sustained attention (Luque-Casado, et al., 2015). Longitudinally, researchers have reported individuals with lower HRV were more likely to develop cognitive impairment than their high HRV counterparts (Kim et al., 2006). Ultimately, previous research has posited that HRV is associated with engagement and disengagement processes in EF, but vary based on the type of task being administered (Jennings et al., 2014).

Sample composition must be considered in examining the controversial nature of the association between HRV and EF. Given African Americans/Blacks (Blacks) experience disparities in cardiovascular morbidity in comparison to their White counterparts, researchers may need to account for this potential ethnic difference in cardiac modulation (Williams et al., 2016). Previous research also suggests Blacks have higher resting HRV than their White counterparts (LaBarron et al., 2015). These findings suggest HRV may not function as a risk factor for disease in Blacks the same way they do in Whites. Moreover, previous research also speculates that cardiac sympathetic activity may be augmented within Blacks (Allen et al., 2015). This gives some paucity to understanding how HRV may function within Blacks, as this association may be different or altered within this ethnic group.

The theory of cardiac autonomic space (Berntson, Caccioppo & Quigley, 1991) describes the complex interplay between the sympathetic and parasympathetic modes of control over heart rate. The branches can be both active (co-activation), either decrease in activity (co-deactivation) or the two branches can have an inverse relationship. These quadrants help elucidate autonomic activity and would prove fruitful in an exploratory study between HRV and EF. The majority of previous research has examined the association between HRV and affective-related attention or inhibitory neurocognitive mechanisms (Mathewson et al., 2010; Gillie, Vasey, & Thayer, 2014). Though there have been few studies that have identified an association between HRV and EF, scant literature has focused on the autonomic plane as a potential mechanism to characterize the individuals taking the EF tasks. The purpose of the current study was to explore the associations between HRV and EF in a sample of predominately Black undergraduate students. Based on previous research (Hansen et al., 2003; Hovland et al. 2012; Ottaviani et al., 2018), we hypothesize that higher HRV will be associated with higher executive function levels. We will then map the results in the autonomic plane to detail the relationship between HRV and EF.

Methodology

Participants

The current study included a total of 104 participants in the study. Participants with a history of psychiatric conditions and cardiovascular health conditions were excluded from the study (n=7). Additionally, there were five more participants who experienced computer errors, and were excluded from the study. Overall, a total of 92 participants (67 females; mean age 20.32 years, SD=2.89) comprised the current study. Participants were recruited from undergraduate Psychology courses at a Northeastern university in the United States and received class credit as compensation for participating in the study. This study was IRB approved and all participants gave informed consent.

Procedure

Data were collected in a single study visit. After obtaining informed consent, the participants completed a demographic questionnaire. After completing the demographic questionnaire, a trained research assistant attached a six-lead ambulatory monitoring system to the participant, in accordance with the user manual (Groot, Geus, & Vries, 1998), to assess their cardiac activity throughout the study session. Two of the electrodes were placed bilaterally along the rib cage, another at the bottom of the sternum and one at the base of the neck. The last two electrodes were placed on the back: one at the base of the neck and one just left of the midpoint of the spinal column. After the electrodes were attached, participants were escorted to an enclosed lab space with only a chair, a table with a monitor on top of it, and a keyboard. As the subject was seated, a trained researcher attached a blood pressure cuff to the participant’s non-dominant arm. Blood pressure and interbeat intervals were assessed during this ten-minute relaxation period, which served as baseline measurement. At the end of the baseline period, the researcher removed the blood pressure cuff and placed the keyboard in the participant’s lap and turned on the monitor. The participant was instructed on how to complete the computerized version of the Berg Card Sorting Task (BCST; Fox et al., 2013).

Measures

Autonomic Activity Assessment

The electrocardiographic signal obtained using a six-lead ambulatory monitoring system (Amsterdam, Netherlands; Vrije Universiteit) was used to derive the interbeat intervals. Ten minutes of interbeat intervals were collected prior to neurocognitive assessment. Five minutes of continuous interbeat interval were selected from the baseline assessment to derive the measures of HRV. The ambulatory cardiographic monitoring system also measures movement and the selection of the continuous data were obtained when motility was lowest. Utilizing the high frequency bandwidth (0.15-0.40 hz), researchers removed artifacts in the interbeat interval data. After correction for artifacts, the interbeat interval data was then transformed employing Fast Fourier transformation method via the CMetX software and protocol (Allen, Chambers, & Towers, 2007). The root mean square of successive differences was computed using the CMetX software. Root mean square of successive differences (rMSSD) is a time domain measure of parasympathetic activity and is positively related to other proposed cardiac measures of parasympathetic activity (Friedman et al, 2002).

From the same hand corrected five minutes of which was utilized for rMSSD, the CMetX software also computed Toichi’s Cardiac Sympathetic Index (CSI). This putative sympathetic metric of sympathetic activity is based on the Lorenz Plot of interbeat intervals (Toichi, Sugiura, Murai, and Sengoku, 1997). Previous research has identified inverse associations between CSI and parasympathetic activity, in addition to positive association with heart rate (Hibbert et al., 2012). The CSI is associated with various subcortical structures associated with autonomic outflow and sympathetic nervous system tone (Ruffle et al., 2018). Ultimately, the CSI is a ratio of R-R intervals and does not have any units. We employed the CSI to index sympathetic activity and to help map participants into subgroups based on autonomic space.

Executive Function Assessment

The Berg Card Sorting Task (BCST) was used to assess processes of EF used for individuals (Fox et al., 2013). The BCST is a measure similar to the Wisconsin Card Sorting Task, measuring set-shifting ability and cognitive flexibility, the ability to attenuate impulsive responses and neurocognitive flexibility (Strauss, Sherman and Spreen, 2006). The BCST is a computerized task created by The Psychology Experiment Language (Mueller et al., 2007). During this task, participants were asked to match cards with stimuli (shape, color or number) on them, to four reference cards based on color, shape or number. After each response, participants were provided feedback, “correct” or “incorrect”. The feedback provided participants the opportunity to learn the correct sorting principle. After the participant reached the criterion for correct responses (ten), the rule changed to the next category. Participants were not aware of the rule change criterion or the rule change. There was no time limit on this task. There were four variables as outcome variables. Categories experienced were the different sets (rules) the participant was able to complete successfully. All the correct responses were totaled to create the total correct responses score. All errors were totaled together to create the total errors variable. Lastly, perseverative errors occurred when participants continued to operate under a rule, after it was changed. These errors were totaled for the perseverative errors variable.

Covariate Measurement

Covariates were assessed based on their association with both EF and HRV. Previous work by this research team has reported an association between depressive symptoms and HRV (Keen et al., 2015). Additionally, previous research suggests depressive symptoms are also associated with EF (Snyder, 2013). Considering the potential influence depressive symptoms may have on HRV and EF, the Beck Depression Inventory (BDI-II) was used to determine depressive symptom severity (Beck, Steer, & Brown, 1996). The BDI-II is a self-report scale used to measure depressive symptoms such as appetite changes, sleep changes, sad mood, hopelessness, helplessness, suicidal ideations and attention difficulties in adults. The BDI-II is comprised of 21 questions and responses range from 0-3. Scores of 0-19 indicate mild to moderate depressive symptoms and scores above 19 indicate severe depressive symptoms (Beck, Steer & Brown, 1996).

Demographically, the literature suggests associations between age, sex, blood pressure, and body mass index (BMI) with both HRV and EF. Age and sex were determined via the demographic questionnaire. Age, measured in years, was self-reported by the participant. Sex, was dummy coded into “0” for male, and “1” for female. Body mass index (BMI) was calculated using the formula weight (in pounds) by the square of height with quotient multiplied by 703. Systolic and diastolic blood pressures were obtained via a blood pressure monitoring system. Blood pressure readings were taken for ten consecutive minutes, and then averaged for data analyses.

Data Analysis

Means and standard deviations of all variables were calculated to determine normality in distribution. Pearson correlations were utilized to investigate zero-order associations between HRV and BCST performance. Hierarchical linear multiple regression analyses were utilized to determine the association between HRV levels and EF performance, in the presence of demographic and physiological covariates. To divide the participants into the four quadrants of autonomic space, researchers utilized the means for rMSSD and Cardiac Sympathetic Index (CSI) to artificially map participants into quadrants. Participants were placed in the Coinhibition, Coactivation, Sympathetic Nervous System Dominant, or Parasympathetic Nervous System Dominant quadrants based on if they were above or below the mean on both rMSSD and CSI. The Kruskal Wallis H test, with Bonferonni corrected post hoc comparisons, was employed to determine if there was a significant difference in BCST performance among the autonomic balance subgroups. Eta square statistics were calculated for the Kruskal Wallis H test to determine effect sizes (Murphey & Myors, 2004).

Results

Sample Characteristics

As seen in Table 1, 25 (27%) males and 67 (73%) females participated in the study. The sample’s mean age was 20.32 years (SD = 2.28), with the majority of participants self-reporting as African American/Black (n=73). Overall, the sample had a BMI of 25.66 (SD = 6.52), which by the World Health Organization’s standards is considered right on the cutoff between normal and overweight (WHO, 2000). Participants averaged 112.62 (SD = 9.14) mmHG in SBP and 68.96 (SD = 6.52) mmHG in DBP.

Table 1:

Sample Characteristics

	M/F (SD/%)
Age	20.32 (2.28)
Sex
Females	67 (72.8%)
Males	25 (27.2%)
Ethnicity
African American/Black	73 (80%)
African	5 (5%)
Mixed	11 (12%)
Other	3 (3%)
SBP mmHg	112.62 (9.14)
DBP mmHg	68.96 (6.52)
BMI	25.66 (6.52)
BDI	6.43 (4.83)
Heart Rate	73.61 (9.05)
rMSSD ms	79.61 (38.73)
BCST-Cat	6.08 (2.62)
BCST-Corr	90.35 (15.17)
BCST-TE	35.18 (17.21)
BCST-PE	19.53 (12.25)

Note: N = SBP = Systolic Blood Pressure; DBP = Diastolic Blood Pressure; BMI = Body Mass Index; BDI = Beck Depression Inventory Total Score; HRV = root Mean Square of Successive Interbeat Interval Differences. BCST-Cat = Berg Card Sorting Test Categories Completed; BCST-Corr = Berg Card Sorting Test Total Correct Responses; BCST-TE = Berg Card Sorting Test Total Errors; BCST-PE = Berg Card Sorting Test Perseverative Errors

Neurocognitive Measures Regressed on Heart Rate Variability in the Presence of Covariates

Hierarchical regression analysis was performed to determine if HRV was associated with EF. As seen in Table 3, HRV was associated with BCST categories completed (β = −0.37, p = 0.01; R² = 0.16), total correct responses after adjusting for demographic and physiological covariates (β = −.40, p = 0.01; R²=0.16). After adjusting for covariates, HRV held a positive relationship with BCST total errors (β = .39, p = .01; R²= 0.17). However, HRV was not associated with perseverative errors (β = .08, p = .45; R² = 0.06).

Table 3:

Executive Function Task Performance Regressed on Heart Rate Variability and Covariates

		R-square	Beta	p-value
BCST-Cat		0.16
	Age		0.09 (0.09)	0.37
	Sex		0.04 (0.73)	0.73
	SBP		0.06 (0.04)	0.65
	DBP		−0.22 (0.05)	0.12
	BMI		0.02 (0.04)	0.85
	BDI		−0.11 (0.05)	0.29
	HRV		−0.37 (0.01)	0.01
BCST-Corr		0.16
	Age		0.14 (0.56)	0.19
	Sex		−0.01 (4.24)	0.90
	SBP		0.07 (0.25)	0.63
	DBP		−0.26 (0.32)	0.06
	BMI		0.06 (0.25)	0.55
	BDI		0.05 (0.33)	0.63
	HRV		−0.40 (0.04)	0.01
BCST-TE		0.17
	Age		−0.16 (0.63)	0.12
	Sex		0.01 (4.77)	0.96
	SBP		−0.06 (0.28)	0.65
	DBP		0.27 (0.36)	0.05
	BMI		−0.05 (0.28)	0.58
	BDI		0.01 (0.37)	0.87
	HRV		0.39 (0.04)	0.01
BCST-PE		0.06
	Age		−0.13 (0.48)	0.22
	Sex		−0.15 (3.61)	0.23
	SBP		−0.25 (0.21)	0.10
	DBP		0.22 (0.27)	0.13
	BMI		−0.04 (0.21)	0.70
	BDI		0.03 (0.28)	0.78
	HRV		0.08 (0.03)	0.45

Note: BCST-Cat = Berg Card Sorting Test Categories Completed; BCST-Corr = Berg Card Sorting Test Total Correct Responses; BCST-TE = Berg Card Sorting Test Total Errors; BCST-PE = Berg Card Sorting Test Perseverative Errors; SBP = Systolic Blood Pressure; DBP = Diastolic Blood Pressure; BMI = Body Mass Index; BDI = Beck Depression Inventory Total Score; HRV = root Mean Square of Successive Interbeat Interval Differences

Differences in Executive Function Across Autonomic Balance Subgroups

The four subgroups represented the four quadrants of autonomic space. The largest group was the Sympathetic Nervous System Dominant (SNS Dominant; n=35), followed by Parasympathetic Nervous System Dominant (PNS Dominant; n= 34), Coinhibition (n=12), and Coactivation (n=11). A one-way analysis of variance (Kruskal Wallis H test) was conducted to compare the effect of location in autonomic space on EF task performance. There was a significant difference in BCST categories completed [χ² (3) = 14.31, p = 0.01, eta square = 0.14], correct responses [(χ² (3) = 8.30, p = 0.04, eta square = 0.08], and total errors ([χ² (3) = 12.79, p = 0.01, eta square = 0.13] based on participants location in autonomic space. However, there was no difference in perseverative errors [(χ² (3) = 6.24, p = 0.10, eta square = 0.07]. These results can are depicted in Figure 1.

Figure 1.

Analysis of variance, with Bonferonni post hoc comparisons, examining executive function across autonomic groups. The SNS Recipriocal group was significantly different than the PNS Reciprocal group for each executive function performance metric. Coinhibition group was significantly lower than the PNS Reciprocal group for total errors. BCST-Cat = Berg Card Sorting Test Categories Completed; BCST-Corr = Berg Card Sorting Test Total Correct Responses; BCST-TE = Berg Card Sorting Test Total Errors; BCST-PE = Berg Card Sorting Test Perseverative Errors

Discussion

The current study was designed to examine the relationship between resting autonomic activity and cognitive flexibility. HRV was inversely related to the BCST categories completed, total correct responses and positively related to the BCST total errors, after adjusting for covariates. The inverse relationship between HRV and cognitive flexibility task performance was counter to our hypothesis. However, given this is a sample primarily comprised of young adults without cardiovascular diseases, it is plausible that increases in sympathetic activity could be associated with preparedness and more adaptive behaviors. In contrast, sympathetic activation in older individuals may be due to the body attempting to adapt to various cardiovascular-related health conditions, including aging (Acharya et al., 2006; Thayer et al., 2008). This finding runs counter to results reported elsewhere. Hansen, Johnsen & Thayer (2003; 2004) reported positive associations between HRV and measures of neurocognition. However, consistent with previous research, HRV was not associated with perseveration or set shifting ability (Britton et al., 2008; Gathright et al. 2016). Given the BCST provides a subset of EF metrics, our findings suggest a potential differentiation in the EF components.

The current findings suggest that participants with lower HRV performed better on a cognitive flexibility-based neurocognitive task, which runs counter to previous research (Hansen, Johnsen & Thayer, 2003; 2004; Kim et al, 2006; Ottaviani et al., 2018). However, previous research has posited that acceleration of heart rate may be indicative of action or complex neurocognitive processes (Jennings et al., 2005). Additionally, though Hazzouri and colleagues reported higher quartiles of HRV performed better on higher order or more complex EF than their lowest quartile counterparts, these findings were with the standard deviation of interbeat intervals and no significant results were reported with rMSSD. This is critical, as the standard deviation of interbeat intervals utilize both sympathetic and parasympathetic activity (Shaffer and Ginsberg, 2017). The complex EF processes would include cognitive flexibility, as it incorporates various EFs to complete a specific task or behavior. So having lower HRV at baseline may be linked to an individual’s ability to complete a complex task, not lower cognitive processes, such as attention (Luque-Casado et al., 2016). Though previous models have posited that these measures are connected to more basic forms of neurocognition, such as attention and information processing speed, other heart-brain conceptualizations suggest that this may vary based on the complexity of the neurocognitive task. The transition from resting HRV levels to neurocognitive task-engaged HRV levels could provide interesting results, as this would allow the ability to map the participants in the autonomic plane during lower level and higher level neurocognitive processes.

Additionally, analyses of variance were employed to determine changes in vagal activity across groups of participants who differed with respect to relative levels of resting cardiac sympathetic and parasympathetic activity. The subgroups were created by “mapping” each individual along dimensions of sympathetic and parasympathetic autonomic space (Berntson, Boysen, & Cacioppo, 1991). A participant’s location in autonomic space provides key information regarding the mode of autonomic control driving this relationship. This procedure provides an alternative perspective of autonomic activation and control. These two variables represent the sympathetic and parasympathetic axes in autonomic space. The SNS Dominant group and the PNS Dominant group performed differently on the cognitive flexibility task (BCST correct and total errors) as a function of location in autonomic space. These findings run counter to outcomes reported by other researchers in which PNS indices of activity, not SNS indices, are associated with EF (Hansen, Johnsen & Thayer, 2003; 2004; Kimhy et al, 2013). This discrepancy in results could be due to sample composition. Specifically, previous findings are reported in middle-aged samples with some participants reporting cardiovascular conditions (Kimhy et al., 2013) or in young adult men on a military base (Hansen Johnsen & Thayer, 2003). The current sample consists of young adult male and female college students, absent cardiovascular diseases. Further, there is the consideration of the current sample being primarily comprised of Blacks. Though other studies may not present this data, previous research suggests that African Americans may have greater HRV levels than their White counterparts (Hill et al., 2015). Moreover, sympathetic activation may be related to the novelty of the laboratory setting or to preparation for engagement in the task (Friedman and Thayer, 1998). While only a few studies have focused on the relationship between baseline HRV and neurocognition, none of them have attempted to relate a participant’s location in autonomic space to performance on an EF measure. Sympathetic modulation is associated with faster reaction times (Luft, Takase, & Darby, 2009).

Utilizing Toichi’s CSI to contextualize the sympathetic activity in addition to the parasympathetic activity (rMSSD) was an attempt to elucidate the unexpected inverse association between HRV and cognitive flexibility. Given previous research suggests Black’s resting autonomic activity may differ from their White counterparts, it seemed imperative to further understand the relationship reported in our current sample. Much of the literature reports a positive relationship between HRV and EF (Hansen et al., 2003), while others report no association (Kimhy et al., 2013). However, this literature very rarely examines the balance within the autonomic nervous system to understand the connection with EF. This becomes especially important, when cognitive flexibility requires more cognitive effort (Luft et al., 2009). This notion is consistent with previous research presenting sympathetic activity increases at baseline are associated with poorer complex neurocognitive tasks in university students (Duschek et al., 2009). If it is plausible for Blacks to exhibit both higher HRV and higher vascular dysfunction relating to sympathetic activity (Dorr et al., 2007), the current findings may align with this potential paradox within Black undergraduate students.

The current study has a few limitations, the first of which concerned the composition of the study sample. There were a disproportionate number of females relative to male participants in the current sample. Potential interactions may exist between sex and depressive symptoms that influence HRV modulation. Though having a predominantely Black sample is a strength, it does limit the generalizability to other samples. Generalizability is also limited by the cross sectional nature of the study and the small autonomic space subgroups. Additionally, we did not assess for any medication that could alter cardiac activity. As significant results were not found with perseveration or set shifting ability, having a more complete EF neuropsychological battery is needed to determine exactly which processes were associated with HRV. The current findings present an association with the general construct of EF, which could be interpreted as HRV is associated with the interplay of various EF component processes or that there may be a specific EF component driving this relationship via the task complexity. Though the procedure to determine autonomic space was created artificially, the use of well accepted time domain measures of autonomic activity would aid in replication.

Overall, the current study shows that tonic baseline measurements HRV can be used to predict neurocognitive function. Survival can only be achieved by employing EF, (i.e. cognitive flexibility, decision making and behavioral inhibition) during novel experiences (Thayer and Lane, 2009). In summation, utilizing HRV as a predictor of neurocognition may provide insight into how an individual perceives their environment. Given the association between decision making processes and autonomic activity, researchers may be able to identify different levels of autonomic activity that directly impact behavior.

Highlights.

1. Inverse association between parasympathetic modulation and EF component cognitive flexibility, not perseveration.

2. Autonomic space subgrouping suggested sympathetically dominant individuals performed better than their parasympathetic counterparts.

3. These findings were in a sample comprised of only African Americans/Black young adults.