Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Jan 4.
Published in final edited form as: Body Image. 2022 Jul 28;42:307–314. doi: 10.1016/j.bodyim.2022.06.010

Does seeking the brawns impact the brain?: An investigation of muscularity-oriented disordered behaviors and cognitive function

Harley M Layman a,*, Natalie G Keirns a, Misty AW Hawkins a, Jason M Nagata b
PMCID: PMC9812210  NIHMSID: NIHMS1841555  PMID: 35908297

Abstract

Shifts in body-image ideals over the past 30 years towards leaner, muscular bodies have revealed new health behaviors that may be related to cognitive function. This study objective was to investigate prospective associations between a drive for muscularity and/or muscularity-oriented disordered behaviors (MODBs) with cognition. Data were drawn from Add Health, a nationally representative longitudinal cohort dataset. Drive for muscularity and MODB engagement were assessed in emerging adulthood (ages 18–26). Cognition was measured via immediate word recall, delayed-word recall, and number recall at 7-years later (ages 24–32). Analyzes were conducted in 1976 participants with available data. A one-way ANCOVA revealed that those with a drive for muscularity had lower immediate word recall (F(3, 12,819) = 3.845, p = .009) and delayed word recall (F(3, 12,807) = 5.933, p < .001) scores than other weight goal groups adjusting for covariates. Hierarchical linear regressions between individual MODBs and cognitive outcomes showed that legal performance-enhancing substance use (βs = 0.06–0.07, p < .05) and exercise (β = 0.06, p < .05) were positively associated with some cognition scores. Conversely, lifting weights (β = − 0.06, p < .05) and eating different foods than usual (β = − 0.05, p < .05) exhibited negative associations with some of the cognitive outcomes. Future research should be conducted to examine other potential outcomes related to the drive for muscularity and associated MODBs.

Keywords: Drive for muscularity, Behaviors, Cognition, Longitudinal

1. Introduction

Even though cognitive function has been previously shown to have established relationships with body composition (Anstey et al., 2011; Yang et al., 2018), eating disorders (Smith et al., 2018), and weight stigma (Guardabassi & Tomasetto, 2018; Nikendei et al., 2011), few studies have examined the link between cognitive function and an emerging and related concept in the health literature: Muscularity-Oriented Disordered Behaviors (MODBs). MODBs are theorized to be the result of current trends toward a “muscular ideal.” Men’s body image ideals have become increasingly muscular over time (Leit et al., 2001; Pope Jr et al., 1999), and women’s ideal body image has moved toward an emphasis on a toned, “fit” physique, suggesting that both genders are impacted by an increased focus on muscularity (Homan, 2010). While the drive for muscularity has been shown to be stable over the past 30 years in men, women have demonstrated a decreased drive for thinness and an increased preference for a lean-muscular body during this time frame (Bozsik et al., 2018; Karazsia et al., 2017). Interestingly, both men and women show a similar drive for muscularity when focusing on muscle tone and not just muscle size – with many comparable cognitive and behavioral correlates across genders (e.g., physical activity and diet changes; Kyrejto et al., 2008). Such findings suggest that examining the drive for muscularity and associated behaviors among both men and women is warranted.

Nagata et al. (2019) propose that endorsement of goals such as actively trying to gain weight or “bulk up” provides indications that a person wishes to achieve increased muscularity. In this case, “gaining weight” refers to gaining muscle mass, not fat mass. Several behaviors have been associated with the desire for increased muscularity. These behaviors to gain lean muscle mass, or MODBs, can be broadly categorized as: 1) alterations to one’s diet, 2) changes to strength/resistance activity regimens, and/or 3) use of legal and/or illegal performance-enhancing substances (PES). Information on each of these behavioral domains and the extant evidence linking them to cognitive performance is described below.

Dietary MODBs, or muscularity-oriented disordered eating (MODE), include: over-regulation of protein consumption and restriction of other dietary macronutrients, consuming protein supplements, engaging in “cheat meals”, eating beyond the point of feeling full, and maintaining access to preplanned foods (Murray et al., 2017; Pila et al., 2017). About two-thirds of male and female adolescents in the United States engage in eating behaviors to increase muscle size or tone (Eisenberg et al., 2012). In a sample of college men, MODE was positively related to set-shifting difficulties (Griffiths et al., 2016). On the other hand, Whey protein consumption is positively associated with cognitive domains such as memory in elderly individuals, at least acutely (Kita et al., 2019). Likewise, in healthy male young adults, a high protein diet improved reaction time compared to a usual protein diet (Jakobsen et al., 2011). However, it is unknown if high protein consumption driven by more pathological eating cognitions is related to executive function. Additional work is needed on dietary MODBs or MODEs as they relate to cognitive domains. It is also important to consider all MODBs, and not just MODE, given that MODE rarely occurs in isolation. Specifically, MODE may be used to supplement strength/resistance training regimens associated with a drive for muscularity and has been shown to be positively associated with weightlifting (Nagata et al., 2019).

In examining strength/resistance activity MODBs, an estimated 15.8 % of young adult males and 2.5 % of young adult females reported exercising to gain weight or build muscle (e.g., weightlifting) (McVey et al., 2005; Nagata et al., 2019), and dissatisfaction with muscularity has been shown to predict exercising to the point of physical injury (Karazsia & Crowther, 2010). While consequences of overtraining and excessive weightlifting on physical functioning have been observed in young people (Weinstein & Weinstein, 2014), investigations of cognitive function and anaerobic, resistance, or strength training is largely restricted to elderly samples. This work has found positive associations between strength training and improved memory, working memory, and verbal reasoning (Cassilhas et al., 2007; Lachman et al., 2006), though it may not depict a universal relationship between strength training and cognitive function, as elderly samples are subject to cognitive aging. Further research is necessary to determine if MODBs such as excessive exercise and weightlifting, specifically, are associated with cognitive function in young adults.

Finally, other MODBs associated with the desire to increase muscularity include performance-enhancing substance (PES) use, both legal and illegal. Legal PESs include substances such as creatine and amino acids, and illegal PESs include anabolic-androgenic steroids (AAS) (Smolak et al., 2005). A systematic review presents mixed findings regarding the beneficial cognitive effects of creatine, with some studies suggesting associated improvements in short- and long-term memory and some studies showing no effect (Avgerinos et al., 2018). On the other hand, AAS-users exhibit visuospatial memory deficits compared to weightlifters that report no AAS use, and within the AAS-user group, total reported lifetime AAS-dose was correlated with deficits of both immediate and delayed Pattern Recognition Memory (Kanayama et al., 2013). Another recent study has observed that long-term AAS users performed worse than the non-AAS group on tests assessing working memory, processing speed, problem-solving, fine motor speed, and executive function (Bjørnebekk et al., 2019). Overall, these findings suggest that creatine supplements may improve short-term memory in healthy adults; however, the impact on other cognitive domains is uncertain. In contrast, AAS-use may have negative effects on cognitive function, particularly memory. Together, these results suggest that PES-related MOBDs need to be examined individually in relation to cognitive impairment.

1.1. Current study

The proposed study advances the literature by conducting a preliminary examination of the relationship between MODBs and cognitive function. Specifically, the current study will be the first to investigate if those who endorse a drive for muscularity and/or specific MODBs show differential cognitive performance on a working memory task and two verbal memory tasks over time in a large nationally-representative sample of young adults. The aims of this study include: Preliminary Aim 1) to examine differences in cognitive performance between weight goal groups (Goal: Weight Loss, Weight Gain, Weight Maintenance, or Weight Neutral), and Primary Aim 2) to investigate if engagement in specific MODBs is associated with cognitive performance within those who endorse a drive for muscularity. No a priori hypotheses were made, given the lack and heterogeneity of the previous literature in this area.

2. Method

2.1. Study design and sample

Data used in the current study came from the National Longitudinal Study of Adolescent to Adult Health (Add Health). Add Health data is collected from a nationally representative cohort of youth that has been followed from adolescence into adulthood in the United States (Harris et al., 2016). The original cohort had 20,743 adolescents assessed when they were 11–18 years old. Systematic sampling methods and implicit stratification were utilized in the original sample (1994–1995, Wave I) to ensure that the schools included in the study were representative of U.S. schools with respect to region of the country, urbanicity, size, type, and ethnicity. Eighty high schools and 52 middle schools were used in the original sample. Additional information regarding the Add Health study design is detailed by Harris et al. (2016), including information on participant privacy and ethical data collection. All study procedures were reviewed and approved by the University of North Carolina Institutional Review Board, and all participants provided written informed consent.

The current study primarily uses data from the restricted-use Wave III (18–26 years, 2001–2002) and Wave IV (24–32 years, 2008) datasets. These waves contain the assessments that measure participation in muscularity-disordered behaviors and/or include evaluations of cognitive performance. For this study’s purposes, Wave III served as our “baseline assessment” and is referred to as baseline. Wave IV served as follow-up, with assessments occurring approximately 7 years after the baseline. We also include the Peabody Vocabulary Test from Wave I as an approximation of adolescent intellectual function/verbal ability and key covariate for future cognitive performance. See Fig. 1 for reference to Add Health study timeline and current study variables.

Fig. 1.

Fig. 1.

Open in a new tab

Add Health study design and current study variables.

The original Wave IV sample consisted of 15,701 respondents, respectively. We excluded participants who did not undergo cognitive testing. All Wave IV participants who completed cognitive testing were analyzed for Preliminary Aim 1; from the Wave IV sample that completed cognitive testing, participants who chose “trying to gain weight or bulk up” on the Drive for Muscularity measure at Wave III-baseline were retained for Primary Aim 2 analyses (see Drive for Muscularity in Measures section for more details). The final sample for Primary Aim 2 consisted of 1976 participants who were 28.8 ± 1.8 years old, 81.9 % male, and had a mean BMI of 25.3 ± 4.6 (see Fig. 2 for how the final Wave IV sample was selected). Males were overrepresented in this sample, as is expected in studies that assess a drive for muscularity during this time period (2001–2002).

Fig. 2.

Fig. 2.

Open in a new tab

Participant flow chart, Wave IV.

2.2. Measures

2.2.1. Baseline measures: emerging adulthood (18–26 years)

We identified muscularity-oriented disordered eating behaviors using the procedures set forth by Nagata et al. (2019) in a previous analysis of muscularity-oriented disordered behaviors in the Add Health data. Participants were first categorized by whether they were attempting to gain weight (i.e., assumed muscle mass), and those endorsing a positive response were asked follow-up questions regarding specific behaviors. Item-level descriptions and response options are described below.

2.2.1.1. Drive for muscularity.

At baseline, (young adult) participants were asked “What are you currently doing about your weight?” Response options included: “trying to lose weight” (1), “trying to gain weight or bulk up” (2), “trying to stay the same weight” (3), or “not trying to do anything about weight” (4). Corresponding weight goal groups used to analyze Preliminary Aim 1 were classified as: Weight Loss, Weight Gain, Weight Maintenance, or Weight Neutral. If participants chose the response “trying to gain weight or bulk up,” it was coded as a weight gain or muscle-building attempt, which is indicative of a drive for muscularity. Only participants who selected this answer choice were included in our Primary Aim 2 analyses.

2.2.1.2. Muscularity-oriented disordered behaviors.

Those who reported weight gain or muscle-building attempt were then asked a follow-up question: “During the past seven days, which of the following things did you do in order to gain weight or build muscle? Check all that apply.” Participant response options included: (a) ate different foods than usual, (b) exercised, (c) lifted weights, (d) took food supplements, or (e) ate more. Additionally, participants were asked, "In the past year, have you used anabolic steroids or other illegal performance-enhancing substances for athletes?" (i.e., illegal PES use at time of survey) and “In the past year, have you used a legal performance-enhancing substance for athletes (such as Creatine Monohydrate, or Andro)?” (i.e., legal PES-use at time of survey). All response options were: 0 = no, 1 = yes. Together, these items represent seven different potential muscularity-oriented disordered behaviors (MODBs).

2.2.2. Follow-up measures: 7-year follow-up (24–32 years)

2.2.2.1. Working memory.

Working memory, an aspect of executive attention abilities, was assessed using a digit-span backwards task at Wave IV (7-year follow-up). The digit-span backwards task is a standardized measure that is utilized to assess working memory in the Wechsler Adult Intelligence Scale (WAIS-IV). The task involved an interviewer reading strings of numbers aloud, with 1-second intervals between each number. The participant was then asked to recall the string of numbers in reverse order. The task began with a two-number string and consisted of seven levels possible. At each level, the participant had two trials to recall the number string backwards correctly. If the correct response was given on the first trial, the second trial of that level was not administered and the interviewer would then move to the number string at the next level. If the participant was unable to accurately recall a number string in both trials, the task was concluded. The possible range of scores was from 0 to 7, where higher scores demonstrate better working memory (i.e., greater executive attention). We will refer to this measure as “number recall”.

2.2.2.2. Verbal memory: short-term and long-term.

Verbal memory (i.e., word recall) scores were determined using the Rey Auditory-Verbal Learning Test at Wave IV (7-year follow-up). To evaluate immediate word recall, the interviewer read a list of 15 common words aloud with 1-s intervals between each word. The participant was then instructed to recall as many of the 15 words as possible in a 90-s period, or until they indicated that they could not remember any other words. The participant received one point for each correct word recalled and higher scores indicate better short-term verbal memory.

Delayed word recall was assessed later in the home interview. Participants were asked to recall as many of the words from the previous list in the first recall task in a 60-s period. The participant received one point for each correct word recalled, with higher scores indicating better long-term verbal memory.

2.2.3. Key covariates

2.2.3.1. General verbal ability.

General verbal ability in adolescence was assessed using a modified version of the Peabody Picture Vocabulary Test at Wave I. In this task, the interviewer reads a word aloud and asks the participant to pick one of the four pictures in front of them that best fits this meaning. The task consisted of 87 items and raw scores were standardized by participant age. This measure was included in our analyzes to control for general verbal ability in Wave I (adolescence) on verbal memory and working memory in Waves IV (adulthood).

2.2.3.2. Measured body mass index.

BMI was calculated by dividing weight (in kilograms) by height (in meters) squared in Wave IV. An interviewer measured weight using a digital scale and height using a steel tape measure.

2.2.3.3. Demographics.

Demographics assessed include age, sex, and highest education achieved. Age was calculated by subtracting the participant's self-reported birth year (Wave IV) from the year that the Wave IV data were collected. Sex and education achieved were collected via self-report measures in Wave IV.

2.3. Procedure

Wave III and Wave IV participants were drawn from the original Add Health Wave I participant pool. At Wave III, an interviewer traveled to participants’ homes and administered an in-home interview. Additionally, sensitive items were asked using a self-report portion during the assessment. Wave IV consisted of a 90-min computer-based self-report instrument and a 30-min biomarker collection period in which anthropometric data was collected and cognitive assessments were performed.

2.4. Data cleaning

First, the subsample that underwent cognitive testing was extracted from the larger Add Health dataset for Preliminary Aim 1 analyses. From this subsample, we selected all participants who endorsed the drive for muscularity (see Fig. 2 for sample selection) for Primary Aim 2 analyses. The responses of these participants to the seven categorical muscularity-oriented disordered behaviors items were scored individually (1 point for yes or 0 points for no). All data and variables were reviewed prior to scoring analysis to assure they were complete and met assumptions for bivariate correlation, as well as linear regression.

2.5. Data analysis

To assess Preliminary Aim 1, cognitive scores across the four weight groups (i.e., Weight Loss, Weight Gain, Weight Maintenance, and Weight Neutral) were compared (see Fig. 3) using an ANCOVA with BMI, age, gender, and education entered as covariates. Then participants who selected the answer choice “trying to gain weight or bulk up” on the measure of drive for muscularity were selected for Primary Aim 2 analyses. Descriptive statistics and bivariate correlations (Pearson’s or point-biserial depending on the nature of the variables) were conducted among all variables to understand how many people in the “trying to gain weight or bulk up” sample were participating in different muscularity-oriented disorders behaviors and how these behaviors may be related to their cognitive function.

Fig. 3.

Fig. 3.

Open in a new tab

Mean Wave IV cognition scores across Wave III weight goal groups. Covariates included age, sex, BMI, and education.

Next, multiple hierarchical linear regressions were performed to examine the amount of variance accounted for in each of the three cognitive variables by individual muscularity-oriented behaviors. The independent variables were eating different foods than usual for weight gain, eating more, taking food supplements, taking legal PES, taking anabolic steroids or other illegal PES, exercise, and weight lifting; the dependent variables were working memory task score, short-term verbal memory scores, or long-term verbal memory scores. Covariates included gender, age, Wave I general verbal ability, and BMI. These covariates were entered in Step 1 and the individual MODBs were entered simultaneously in Step 2. The covariates were included to test whether participation in specific MODBs would be associated with cognitive performance over and above the demographics. Entering the MODBs simultaneously allows us to ascertain whether certain behaviors are unique predictors of cognitive function variables compared to others. However, simultaneous entry of related items can lead to multicollinearity. Thus, multicollinearity was assessed by examining the variance inflation factors (VIF). No VIF values were greater than 10; therefore, the simultaneous models were retained (Bowerman & O′Connell, 1990). All analyzes were conducted using SPSS version 25.

3. Results

3.1. Preliminary aim 1 analyses – weight-goal groups comparison

When examining differences in cognitive function variables across the four weight goal groups, it was found that there were significant group differences on measures of immediate word recall scores, F(3, 12,819) = 3.845, p = .009, and delayed word recall scores, F (3, 12,807) = 5.933, p < .001, after adjusting for covariates (see Fig. 2 for means). Number recall scores were not significantly different across groups. Post-hoc analyses were conducted to examine specific group differences across cognitive tests using estimated marginal means. Those in the Weight Gain group had significantly lower immediate word recall scores than those in the Weight Loss group (Mdiff = − 0.084, p = .005), Weight Maintenance group (Mdiff = − 0.101, p = .002), and Weight Neutral group (Mdiff = − 0.056, p = .041). Similarly, the Weight Gain group had significantly lower delayed word recall scores than the Weight Loss (Mdiff = − 0.098, p = .001), Weight Maintenance (Mdiff = − 0.131, p < .001), and Weight Neutral groups (Mdiff = − 0.080, p = .003). No additional significant mean differences were observed.

Detailed demographic and descriptive data for the Weight Gain subsample can be found in Table 1. Bivariate correlations between MODBs and Wave IV cognitive scores in this subsample can be found in Table 2.

Table 1.

Characteristics of participants.

Total sample (Max N = 1976) Males
(Max n = 1620)		 Females
(Max n = 356)
Demographic Factors/Covariates
  Age (Wave IV) 28.8 ± 1.8 28.8 ± 1.8 28.4 ± 1.7*
  BMI (Wave IV) 25.3 ± 4.6 25.9 ± 4.6 22.5 ± 4.1*
  Picture Vocabulary Test (Wave I) 99.7 ± 13.9 102.0 ± 13.8 96.2 ± 13.5*
  Highest Education Level (Wave IV)
    Less than High School Diploma 190 (0.1) 157 (9.7) 33 (9.3)
    High School Diploma/GED 420 (21.3) 375 (23.1) 63 (17.7)
    Some College/Vocational Training 1323 (66.9) 1072 (66.2) 251 (70.5)
    Bachelor’s Degree and Beyond 43 (0.0) 34 (2.1) 9 (2.5)
MODBs (Wave III)
    Eat Different Foods (% yes) 563 (28.5) 435 (27.0) 128 (36.5)*
    Exercise (% yes) 1014 (51.3) 930 (57.4) 84 (23.7)*
    Lift Weights (% yes) 1029 (52.1) 965 (59.6) 64 (18.1)*
    Take Food Supplements (% yes) 433 (21.9) 383 (23.6) 50 (14.1)*
    Eat More (% yes) 1042 (52.7) 816 (50.4) 226 (63.8)*
    Use Creatine (past 12 mo.; % yes) 439 (22.2) 427 (26.8) 12 (3.4)*
    Use AAS (past 12 mo.; % yes) 62 (3.1) 56 (3.5) 6 (1.7)
Cognitive Factors (standardized)
    Immediate Word Recall −0.12 ± 0.9 −0.15 ± 0.9 −0.01 ± 1.0*
    Delayed Word Recall −0.15 ± 1.0 −0.19 ± 1.0 0.02 ± 1.0*
    Number Recall 0.01 ± 1.0 0.02 ± 1.0 0.00 ± 1.0
Open in a new tab

Note. Data is from the National Longitudinal Study of Adolescent to Adult Health (Add Health). Wave I = 1994–1995. Wave III = 2001–2002. Wave IV = 2008. BMI = body mass index. MODBS = Muscularity-oriented disordered behaviors. Continuous variables represented with mean ± SD. Categorical variables represented with N (%).

*

p < .05 for independent t-test or chi-square test comparing males and females.

Table 2.

Bivariate correlations between baseline MODBs and 7-year follow-up cognition scores.

Variables 1 2 3 4 5 6 7 8 9 10
MODBs (Wave III)
1. Eat Different Foods --
2. Exercise −.01 --
3. Lift Weights .09** .41** --
4. Take Food Supplements .06* .14** .21** --
5. Eat More .07** −.17** −.23** .04 --
6. Legal PES-Use −.01 .17** .26** .36** −.04 --
7. Illegal PES-Use −.02 .02 .07** .13** −.03 .25** --
Cognitive Function (Wave IV)
8. Immediate Recall −.04 −.05* −.01 .00 −.03 −.08** −.01 --
9. Delayed Recall −.02 .09** −.01 .01 −.02 .02 −.03 .65** --
10. Number Recall −.01 −.01 .08** .01 .02 .09** .00 .25** .27** --
Open in a new tab

Note. Data are from the National Longitudinal Study of Adolescent to Adult Health (Add Health) Waves III and IV. Bivariate correlations included Pearson’s and point-biserial depending on the nature of the variables. MODBS = Muscularity-oriented disordered behaviors. PES = performance-enhancing substances. Light grey and *p < .05, and dark grey and **p < .01.

3.2. Primary aim 2 results – wave IV cognition associations

To follow-up initial descriptive analyses, the primary study analyses consisted of a series of hierarchical linear regressions to evaluate the relationship of MODBs in young adulthood (Wave III) on different measures of cognitive function later in adulthood (Wave IV) adjusting for key covariates. Covariates were entered in Step 1 and included age, gender, BMI, highest education level, and Wave I general verbal ability. Individual MODB scores were entered simultaneously in Step 2. Associations between all covariates and cognitive outcomes, as well as associations between MODBs and cognitive outcomes, are displayed in Table 3. In the next section, we describe the specific results across the different cognitive variables.

Table 3.

Prospective associations between Wave IV covariates, Wave III MODBs and Wave IV standardized cognitive outcomes in adults 24–32 years of age.

Immediate word recall
(short-term memory)
 Delayed word recall
(long-term memory)
 Number recall
(working memory)
R 2 ΔR 2 R 2 ΔR 2 R 2 ΔR 2
Step 1a .079 .082 .098
Step 2 .086 .008* .084 .002 .106 .008*
βa (95 % CI) βa (95 % CI) βa (95 % CI)
Age −0.05 (−0.05 to −0.01)** −0.04 (−0.06 to −0.01)** −0.03 (−0.05 to 0.01)
Biological Sex 0.08 (0.07–0.32)** 0.09 (0.10–0.35)*** 0.01 (−0.12 to 0.14)
Body Mass Index (BMI) 0.00 (−0.01 to 0.01) −0.04 (−0.02 to 0.00) −0.04 (−0.02 to 0.00)
Highest Education 0.16 (0.05–0.10)*** 0.15 (0.05–0.09)*** 0.15 (0.05–0.09)**
Peabody Picture Vocabulary Test 0.15 (0.01–0.01)*** 0.17 (0.01–0.02)*** 0.21 (0.01–0.02)**
Eat Different Foods −0.05 (−0.20 to −0.01)* −0.03 (−0.16 to 0.04) −0.00 (−0.12 to 0.14)
Exercise 0.03 (−0.03 to 0.17) 0.01 (−0.09 to 0.11) 0.06 (0.03–0.23)*
Lift Weights −0.06 (−0.22 to −0.01)* −0.03 (−0.16 to 0.01) −0.06 (−0.23 to −0.02)*
Take Food Supplements −0.02 (−0.16 to 0.07) 0.01 (−0.09 to 0.14) −0.03 (−0.18 to 0.05)
Eat More −0.03 (−0.15 to 0.04) −0.02 (−0.13 to 0.01) 0.02 (−0.05 to 0.14)
Legal PES-Use 0.06 (0.03–0.527)* 0.01 (−0.01 to 0.14) 0.06 (0.03–0.27)*
Illegal PES-Use −0.02 (−0.34 to 0.17) −0.02 (−0.38 to 0.13) 0.00 (−0.25 to 0.27)
Open in a new tab

Note. Data is from the National Longitudinal Study of Adolescent to Adult Health (Add Health) Waves III & IV. MODBS = Muscularity-oriented disordered behaviors. PES = performance enhancing substances. Covariates included age, biological sex, BMI, education, and Wave I cognition.

*

p < .05

**

p < .01

***

p < .001

a

Only covariates were entered in Step 1. For parsimony, specific beta coefficients of covariates are only presented for Step 2.

3.2.1. Immediate word recall

Hierarchical linear regression results demonstrated that both eating different foods than usual, β = − 0.046, p = .043, and lifting weights to gain weight β = − 0.057, p = .033, were negatively associated with immediate word recall scores (Table 3). Additionally, legal PES-use positively predicted immediate word recall scores, β = 0.063, p = .015. Other individual MODBs did not account for a significant amount of variance in immediate word recall scores. The overall regression model was significant, F(12, 1807) = 14.188, p < .001, and specific MODBs explained 0.8 % of the variance.

3.2.2. Delayed word recall

Results of the hierarchical linear regression revealed that no specific MODBs were significantly associated with delayed word recall scores (Table 3). The regression model was significant, F(12, 1804) = 13.803, p < .001, but specific MODBs explained 0.0 % of the variance.

3.2.3. Number recall

Results of the hierarchical linear regression of number recall scores on specific MODBs demonstrated that both exercising, β = 0.062, p = .014, and using legal PESs, β = 0.061, p = .017, were positively associated with number recall scores (Table 3). Lifting weights negatively predicted number recall scores, β = 0.–059, p = .025 (Table 3). Other individual MODBs did not account for a significant amount in the variance of number recall scores. The overall regression model was significant, F(12, 1810) = 17.828, p < .001, and specific MODBs explained 0.8 % of the variance.

4. Discussion

The purpose of this study was to determine if the drive for muscularity and/or associated muscularity-oriented disordered behaviors (MODBs) were associated with cognitive function later in adulthood. Preliminary Aim 1 analyses revealed that − as a group − those with a drive for muscularity (as defined by a desire to “gain weight/bulk up”) had lower scores on short-term verbal memory and long-term verbal memory than those with other weight change goals. Working memory scores were similar across all groups. Primary Aim 2 analyses showed that within individuals with a drive for muscularity, engaging in specific MODBs (e.g., legal PES-use, weight-lifting) displayed small associations with working memory and short-term verbal memory, but no association with long-term verbal memory. Specifically, legal PES-use and exercise were positively associated with follow-up cognition scores, and weight-lifting and eating different foods than usual were negatively associated with follow-up cognition scores. Though all observed effects were small (ΔR2 = 0.002–0.008), these data provide preliminary evidence on potential relationships between MODBs and cognitive function and can be used to inform further research in this area.

Many of the current findings are consistent with past work. Specifically, the finding that legal PES-use is associated with better cognitive performance is consistent with previous literature on the beneficial cognitive impact of creatine, a popular legal PES (e.g., McMorris et al., 2007; Rae et al., 2003). Similar legal PESs, such as dehydroepiandrosterone (DHEA; an Andro supplement), have also been shown to improve cognitive function (i.e., episodic memory) (Alhaj et al., 2006). However, other psychosocial factors (e.g., income) may also contribute to the observed relationship between endorsement of legal PES-use and better cognitive performance and should be considered in future studies.

Furthermore, our findings that weight-lifting is associated with poorer short-term verbal and working memories may be in line with the frontal aging hypothesis (Greenwood, 2000; West, 1996). Both of these cognitive functions are more related to speeded or executive attention abilities, and the frontal aging hypothesis predicts that deficits in speeded and executive control abilities decline at a faster rate than memory for words or general knowledge due to the greater vulnerability of the frontal lobe and its neural circuitry to age, stress, and disease. It is possible that even sub-clinical disease development is occurring in younger populations (i.e., the current sample) as more evidence emerges that markers of cardiometabolic disruption are detectable prior to overt disease (Castorani et al., 2020; Oren et al., 2003). Additionally, it is possible that those who reported weight lifting to gain muscle may have done so excessively, as a drive for muscularity has been shown to be positively predictive of exercising to the point of injury (e.g., lifting extremely heavy amounts of weight, lifting weights for a long period of time, or both) (Karazsia & Crowther, 2010). Excessive weight lifting and high-intensity exercise have been shown to be detrimental to cognition scores (Kobus et al., 2010), which would be consistent with the current findings.

Increased cognitive performance on the working memory task was also observed in those in our sample who endorsed “exercise” to gain weight or bulk up. While it is unclear which forms of exercise are being engaged in by our participants, a substantial body of literature links higher physical activity levels with cognitive benefits across a wide array of populations (Donnelly et al., 2016; Kemoun et al., 2010; Tseng et al., 2011). Thus, our finding that exercise showed protective relationships with working memory scores is expected. It is important to note that given the lack of nuance in the exercise item prompt (e.g., no information about whether the exercise was compulsive or excessive), the exercise item may not qualify as a MODB, thereby potentially limiting its usefulness as a MODB predictor of cognitive function.

Eating different foods than usual, a dietary change-MODB, was associated with poorer cognition scores on the short-term verbal memory task in the current study. Our findings align with a previous study that observed cognitive dysfunction in those who participated in MODE (Griffiths et al., 2016). However, these findings are preliminary and should continue to be explored given the lack of specificity of this measure in the current study.

4.1. Limitations and future directions

Various limitations should be considered when interpreting the findings of the current study. First, measures assessing participation in MODBs were gathered via self-report and may have been vulnerable to reporting bias. In particular, AAS-use is illegal in the United States and is associated with various negative attitudes, which may lead to under-reporting (Griffiths et al., 2016; Pope et al., 2017). Additionally, questions regarding taking food supplements, eating different foods, or legal PES-use did not specify the dietary changes made (e.g., limited carbohydrate consumption, increased protein consumption) or the type of product used (e.g., protein shakes, pre-workout, creatine, aminos). Thus, it is not clear if there is a specific product that may be more related to cognitive performance than others or whether a third variable (e.g., higher socio-economic status) related to the ability to access these products is driving cognitive improvement. Similarly, other details regarding the severity (frequency and degree) of MODB participation, such as exercise and weight lifting, were not measured. Given the lack of specificity among these measures, it is possible that other confounding variables (e.g., anxiety, depression, and stigma) may have also had an effect on cognition, and thus, weakening our ability to clearly determine which specific aspects of MODBs, their correlates, or confounders may be implicated in cognitive health. Future research should investigate the specific types and amount of supplements and food consumed to gain weight, as well as the amount and intensity of exercise engaged in to assess for a dose-response relationship between specific MODBs and cognitive performance as well as probe critical features such as level of preoccupation and compulsivity. Together, such issues call for the development of psychometrically-validated measures of MODBs for future investigations.

Next, the Add Health survey did not include questions regarding body image-related concerns and drives, or MODB behavior participation in later waves. This information may be important given the shift in body image ideals in recent years towards a more muscular, lean, and toned body in both men and women (Homan, 2010; Leit et al., 2001; Pope Jr et al., 1999). Therefore, it is unknown whether participants continued or began engaging in these behaviors into adulthood, prohibiting causal conclusions to be drawn. In addition, cognitive testing was also not administered in earlier waves, such as Wave III. While the current study controlled for general verbal ability at Wave I, participants’ cognitive ability was not assessed at the same time as their endorsed MODB engagement. Consequently, we are unable to determine changes in cognition from young adulthood into older adulthood, or account for developmental processes that may have occurred. The lack of baseline cognitive assessment also poses an issue of potential bidirectionality between study variables. Future work should include cognitive measures across all time points to consider developmental processes and strengthen the utility of these findings. In addition, the cognitive testing administered by Add Health also only included a small number of tests that do not evaluate a diverse range of cognitive abilities (e.g., inhibitory control, processing speed, or cognitive flexibility). Future research should include a more diverse cognitive battery. Further, follow-up investigations would benefit from evaluating cognitive performance and body image concerns/behaviors at all time points to assess for cross-sectional and longitudinal effects of varying levels of different drives and their associated behaviors at multiple levels (i.e., length of use, frequency of use) on cognitive function.

4.2. Conclusions

Overall, the current study revealed that – at the group level – individuals with a drive for muscularity scored lower on multiple domains of cognitive function compared to groups with other weight goals (e.g., stay the same weight) – though effects are small. Further investigation in this subsample showed that specific MODBs (i.e., legal PES-use and weight-lifting) demonstrated small associations with working memory and short-term verbal memory later in life. While previous findings regarding the effects of legal PES-use and weight-lifting on cognitive performance have been mixed, our study shows potential small, positive associations with legal PES use and negative associations with lifting weights. These preliminary results should be built upon in future, more robust studies on the potential cognitive, emotional, and physical impacts of various body image related behaviors, including MODBs. Future investigations will benefit from more precise and detailed measurements of muscularity-disordered behaviors and cognition – as well as their covariates and confounders – over time in order to better elucidate these potential relationships.

Funding

Harley Layman, M.S. is supported via the Graduate Research Training Initiative for Students (G-RISE) from the National Institute of General Medical Sciences (NIGMS) of the National Institutes of Health (1T32GM140953-01). Natalie Keirns, M.S., is supported via the Ruth L. Kirschstein Predoctoral Individual National Research Service Award (F31HL152620). Dr. Misty Hawkins is supported via pilot funding through the Center for Integrated Research on Child Adversity (CIRCA), and the NIGMS of the National Institutes of Health (P20GM109097). Dr. Jason Nagata is supported via funding through the NHLBI of the National Institutes of Health (K08HL159350).

Footnotes

CRediT authorship contribution statement

Harley M. Layman: Conceptualization, Methodology, Formal analysis, Writing – original draft, Project administration. Natalie G. Keirns: Writing – review & editing, Visualization. Misty A.W. Hawkins: Supervision, Funding acquisition, Data curation. Jason M. Nagata: Supervision, Validation, Presentation, Visualization.

Declarations of interests

The authors declared no conflict of interest.

References

  1. Alhaj HA, Massey AE, & McAllister-Williams RH (2006). Effects of DHEA administration on episodic memory, cortisol and mood in healthy young men: A double-blind, placebo-controlled study. Psychopharmacology, 188(4), 541–551. 10.1007/s00213-005-0136-y [DOI] [PubMed] [Google Scholar]
  2. Anstey K, Cherbuin N, Budge M, & Young J (2011). Body mass index in midlife and late-life as a risk factor for dementia: A meta-analysis of prospective studies. Obesity Reviews, 12(5), e426–e437. 10.1111/j.1467-789X.2010.00825.x [DOI] [PubMed] [Google Scholar]
  3. Avgerinos KI, Spyrou N, Bougioukas KI, & Kapogiannis D (2018). Effects of creatine supplementation on cognitive function of healthy individuals: A systematic review of randomized controlled trials. Experimental Gerontology, 108, 166–173. 10.1016/j.exger.2018.04.013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bjørnebekk A, Westlye LT, Walhovd KB, Jørstad ML, Sundseth ØØ, & Fjell AM (2019). Cognitive performance and structural brain correlates in long-term anabolic-androgenic steroid exposed and nonexposed weightlifters. Neuropsychology, 33(4), 547. 10.1037/neu0000537 [DOI] [PubMed] [Google Scholar]
  5. Bowerman BL, & O’connell RT (1990). Linear statistical models: An applied approach. Brooks & Cole. [Google Scholar]
  6. Bozsik F, Whisenhunt BL, Hudson DL, Bennett B, & Lundgren JD (2018). Thin is in? Think again: The rising importance of muscularity in the thin ideal female body. Sex Roles, 79(9), 609–615. 10.1007/s11199-017-0886-0 [DOI] [Google Scholar]
  7. Cassilhas RC, Viana VA, Grassmann V, Santos RT, Santos RF, Tufik S, & Mello MT (2007). The impact of resistance exercise on the cognitive function of the elderly. Medicine and Science in Sports and Exercise, 39(8), 1401. 10.1249/mss.0b013e318060111f [DOI] [PubMed] [Google Scholar]
  8. Castorani V, Polidori N, Giannini C, Blasetti A, & Chiarelli F (2020). Insulin resistance and type 2 diabetes in children. Annals of Pediatric Endocrinology & Metabolism, 25(4), 217. 10.6065/apem.2040090.045 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Donnelly JE, Hillman CH, Castelli D, Etnier JL, Lee S, Tomporowski P, … Szabo-Reed AN (2016). Physical activity, fitness, cognitive function, and academic achievement in children: A systematic review. Medicine and Science in Sports and Exercise, 48(6), 1197. 10.1249/MSS.0000000000000901 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Eisenberg ME, Wall M, & Neumark-Sztainer D (2012). Muscle-enhancing behaviors among adolescent girls and boys. Pediatrics, 130(6), 1019–1026. 10.1542/peds.2012-0095 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Greenwood PM (2000). The frontal aging hypothesis evaluated. Journal of the International Neuropsychological Society, 6(6), 705–726. 10.1017/S1355617700666092 [DOI] [PubMed] [Google Scholar]
  12. Griffiths S, Murray SB, & Mond JM (2016). The stigma of anabolic steroid use. Journal of Drug Issues, 46(4), 446–456. 10.1177/0022042616661837 [DOI] [Google Scholar]
  13. Guardabassi V, & Tomasetto C (2018). Does weight stigma reduce working memory? Evidence of stereotype threat susceptibility in adults with obesity. International Journal of Obesity, 42(8), 1500–1507. 10.1038/s41366-018-0121-2 [DOI] [PubMed] [Google Scholar]
  14. Harris KM, Halpern CT, Whitsel E, Hussey J, Tabor J, Entzel P, & Udry J (2016). The national longitudinal study of adolescent to adult health: Research design. 2009. http://www.cpc.unc.edu/projects/addhealth/design. (Accessed 5). [Google Scholar]
  15. Homan K (2010). Athletic-ideal and thin-ideal internalization as prospective predictors of body dissatisfaction, dieting, and compulsive exercise. Body Image, 7(3), 240–245. 10.1016/j.bodyim.2010.02.004 [DOI] [PubMed] [Google Scholar]
  16. Jakobsen LH, Kondrup J, Zellner M, Tetens I, & Roth E (2011). Effect of a high protein meat diet on muscle and cognitive functions: A randomised controlled dietary intervention trial in healthy men. Clinical Nutrition, 30(3), 303–311. 10.1016/j.clnu.2010.12.010 [DOI] [PubMed] [Google Scholar]
  17. Kanayama G, Kean J, Hudson JI, & Pope HG Jr. (2013). Cognitive deficits in long-term anabolic-androgenic steroid users. Drug and Alcohol Dependence, 130(1–3), 208–214. 10.1016/j.drugalcdep.2012.11.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Karazsia BT, & Crowther JH (2010). Sociocultural and psychological links to men’s engagement in risky body change behaviors. Sex Roles, 63(9–10), 747–756. 10.1007/s11199-010-9802-6 [DOI] [Google Scholar]
  19. Karazsia BT, Murnen SK, & Tylka TL (2017). Is body dissatisfaction changing across time? A cross-temporal meta-analysis. Psychological Bulletin, 143(3), 293. 10.1037/bul0000081 [DOI] [PubMed] [Google Scholar]
  20. Kemoun G, Thibaud M, Roumagne N, Carette P, Albinet C, Toussaint L, … Dugué B (2010). Effects of a physical training programme on cognitive function and walking efficiency in elderly persons with dementia. Dementia and Geriatric Cognitive Disorders, 29(2), 109–114. 10.1159/000272435 [DOI] [PubMed] [Google Scholar]
  21. Kita M, Kobayashi K, Obara K, Koikeda T, Umeda S, & Ano Y (2019). Supplementation with whey peptide rich in β-lactolin improves cognitive performance in healthy older adults: A randomized, double-blind, placebo-controlled study. Frontiers in Neuroscience, 13, 399. 10.3389/fnins.2019.00399 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Kobus DA, Brown CM, Wu L, Robusto K, & Bartlett J (2010). Cognitive performance and physiological changes under heavy load carriage. ⟨ 10.21236/ADA551468⟩. [DOI] [Google Scholar]
  23. Kyrejto JW, Mosewich AD, Kowalski KC, Mack DE, & Crocker PR (2008). Men’s and women’s drive for muscularity: Gender differences and cognitive and behavioral correlates. International Journal of Sport and Exercise Psychology, 6(1), 69–84. 10.1080/1612197X.2008.9671855 [DOI] [Google Scholar]
  24. Lachman ME, Neupert SD, Bertrand R, & Jette AM (2006). The effects of strength training on memory in older adults. Journal of Aging and Physical Activity, 14(1), 59–73. 10.1123/japa.14.1.59 [DOI] [PubMed] [Google Scholar]
  25. Leit RA, Pope HG Jr., & Gray JJ (2001). Cultural expectations of muscularity in men: The evolution of Playgirl centerfolds. International Journal of Eating Disorders, 29(1), 90–93. [DOI] [PubMed] [Google Scholar]
  26. McMorris T, Mielcarz G, Harris RC, Swain JP, & Howard A (2007). Creatine supplementation and cognitive performance in elderly individuals. Aging, Neuropsychology, and Cognition, 14(5), 517–528. 10.1080/13825580600788100 [DOI] [PubMed] [Google Scholar]
  27. McVey G, Tweed S, & Blackmore E (2005). Correlates of weight loss and muscle-gaining behavior in 10- to 14-year-old males and females. Preventive Medicine, 40(1), 1–9. 10.1016/j.ypmed.2004.04.043 [DOI] [PubMed] [Google Scholar]
  28. Murray SB, Nagata JM, Griffiths S, Calzo JP, Brown TA, Mitchison D, Mond JM, et al. (2017). The enigma of male eating disorders: A critical review and synthesis. Clinical Psychology Review, 57, 1–11. 10.1016/j.cpr.2017.08.001 [DOI] [PubMed] [Google Scholar]
  29. Nagata JM, Murray SB, Bibbins-Domingo K, Garber AK, Mitchison D, & Griffiths S (2019). Predictors of muscularity-oriented disordered eating behaviors in US young adults: A prospective cohort study. International Journal of Eating Disorders, 52(12), 1380–1388. 10.1002/eat.23094 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Nikendei C, Funiok C, Pfüller U, Zastrow A, Aschenbrenner S, Weisbrod M, … Friederich H-C (2011). Memory performance in acute and weight-restored anorexia nervosa patients. Psychological Medicine, 41(4), 829–838. 10.1017/S0033291710001121 [DOI] [PubMed] [Google Scholar]
  31. Oren A, Vos LE, Uiterwaal CS, Grobbee DE, & Bots ML (2003). Cardiovascular risk factors and increased carotid intima-media thickness in healthy young adults: The Atherosclerosis Risk in Young Adults (ARYA) Study. Archives of Internal Medicine, 163(15), 1787–1792. 10.1001/archinte.163.15.1787 [DOI] [PubMed] [Google Scholar]
  32. Pila E, Mond JM, Griffiths S, Mitchison D, & Murray SB (2017). A thematic content analysis of# cheatmeal images on social media: Characterizing an emerging dietary trend. International Journal of Eating Disorders, 50(6), 698–706. 10.1002/eat.22671 [DOI] [PubMed] [Google Scholar]
  33. Pope HG, Khalsa JH, & Bhasin S (2017). Body image disorders and abuse of anabolic-androgenic steroids among men. JAMA, 317(1), 23–24. 10.1001/jama.2016.17441 [DOI] [PubMed] [Google Scholar]
  34. Pope HG Jr., Olivardia R, Gruber A, & Borowiecki J (1999). Evolving ideals of male body image as seen through action toys. International Journal of Eating Disorders, 26(1), 65–72. [DOI] [PubMed] [Google Scholar]
  35. Rae C, Digney AL, McEwan SR, & Bates TC (2003). Oral creatine monohydrate supplementation improves brain performance: A double–blind, placebo–controlled, cross–over trial. Proceedings of the Royal Society of London. Series B: Biological Sciences, vol. 270(issue 1529) (pp. 2147–2150). ⟨ 10.1098/rspb.2003.2492⟩. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Smith KE, Mason TB, Johnson JS, Lavender JM, & Wonderlich SA (2018). A systematic review of reviews of neurocognitive functioning in eating disorders: The state-of-the-literature and future directions. International Journal of Eating Disorders, 51(8), 798–821. 10.1002/eat.22929 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Smolak L, Murnen SK, & Thompson JK (2005). Sociocultural influences and muscle building in adolescent boys. Psychology of Men & Masculinity, 6(4), 227. 10.1037/1524-9220.6.4.227 [DOI] [Google Scholar]
  38. Tseng C-N, Gau B-S, & Lou M-F (2011). The effectiveness of exercise on improving cognitive function in older people: A systematic review. Journal of Nursing Research, 19(2), 119–131. 10.1097/JNR.0b013e3182198837 [DOI] [PubMed] [Google Scholar]
  39. Weinstein A, & Weinstein Y (2014). Exercise addiction-diagnosis, bio-psychological mechanisms and treatment issues. Current Pharmaceutical Design, 20(25), 4062–4069. 10.2174/13816128113199990614 [DOI] [PubMed] [Google Scholar]
  40. West RL (1996). An application of prefrontal cortex function theory to cognitive aging. Psychological Bulletin, 120(2), 272. 10.1037/0033-2909.120.2.272 [DOI] [PubMed] [Google Scholar]
  41. Yang Y, Shields GS, Guo C, & Liu Y (2018). Executive function performance in obesity and overweight individuals: A meta-analysis and review. Neuroscience & Biobehavioral Reviews, 84, 225–244. 10.1016/j.neubiorev.2017.11.020 [DOI] [PubMed] [Google Scholar]

RESOURCES