Understanding the Link Between Physical Exercise, Autonomous Motivation, Exercise Dependence, and Adult ADHD Symptoms: A Cross-Sectional Study

doi:10.21203/rs.3.rs-4324428/v1

Download PDF

Research Article

Understanding the Link Between Physical Exercise, Autonomous Motivation, Exercise Dependence, and Adult ADHD Symptoms: A Cross-Sectional Study

https://doi.org/10.21203/rs.3.rs-4324428/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

Physical Activity (PA) and Physical Exercise (PE) tends to reduce ADHD symptoms in children; however, the relationship might be more complicated within adults, and differ dependent on ADHD-symptom subtypes. Given the higher prevalence of addiction in individuals with ADHD, investigating the relationship between Exercise Dependence (ED), ADHD, and intrinsic/extrinsic motivations is crucial when considering PE as a potential intervention for ADHD. This cross-sectional study investigated the relationship between PE, ADHD symptoms, intrinsic/extrinsic motivation, and ED.

Methods

152 adult participants (77.63% female) completed an online survey measuring: PE level (IPAQ-L); ADHD symptomatology (ADHD-RS-IV); subclinical ADHD diagnosis (ADHD-RS-IV); ED (EDS-R), and autonomous exercise motivation (BREQ-3).

Results

Positive correlations were found between ADHD symptomatology (total and hyperactive) with ED, and negative correlations were found between ADHD symptomatology (total and hyperactive) with autonomous exercise motivation. Additionally, a significant positive correlation was observed between ED and autonomous exercise motivation.

Conclusions

Promoting intrinsic exercise motivation in adults with ADHD could potentially lead to increased PE engagement. However, caution is advised, as intrinsic motivation was also positively associated with ED. Further research is needed to explore strategies for optimising the positive effects of motivation-related interventions, while mitigating potential risks.

Attention Deficit Hyperactivity Disorder

Physical Exercise

Exercise Dependence

Exercise Motivation

Autonomous Motivation

Adult Population

Attention Deficit/Hyperactive disorder (ADHD) is primarily characterised by a persistent pattern of inattention and/or hyperactivity/impulsivity, along with poor attention and memory, which impede development or functioning [1]. Prevalence rates vary, with estimates typically ranging between 2.2%[2] and 6–7%[3] for children and adolescents, and 2.8%[2] to 5%[3] for adults.

Due to the limitations associated with both pharmacological (e.g., reduced efficacy over time, side effects) [4, 5] and non-pharmacological treatments (e.g., cost-effectiveness) [6] for ADHD, Physical Activity (PA) and/or Physical Exercise (PE) (a subset of PA) has been considered as an alternative. Indeed, recent evidence suggests that PA/PE may help alleviate cognitive deficits (i.e., attention, information processing), reduce socio-emotional difficulties, and improve physical/motor symptoms [7, 8]. In male adults with ADHD (or those expressing elevated ADHD symptoms, a 20-minute bout of exercise was found to improve motivation for cognitive tasks and increase feelings of energy [9], and a positive association between PA and reduced ADHD symptoms has also been reported [10].

However, an overlooked issue is the potential for PA-based interventions in individuals with ADHD, particularly adults, to increase the risk of Exercise Dependence (ED) or Exercise Addiction (EA). ED refers to a state in which a person exercises excessively/obsessively to a point of dependence, whereas EA is defined as dysfunctional behaviour marked by exaggerated training, loss of control over exercise behaviour, and negative impacts on daily life [11]. Whilst the rate of EA among general exercisers has been found to be around 8.1% [12], rates may be higher in individuals with ADHD. Indeed, there is an established association between ADHD and risk of addictions, including behavioural ones [13], and Colledge et al., [14] found that individuals deemed “at-risk” of exercise addiction reported significantly higher symptoms of ADHD. Similarly, Popat et al., [15] found that medicated individuals with ADHD reported higher levels of exercise withdrawal and fixation levels, although ED total scores did not differ significantly from healthy controls. Further, regardless of medication status, those with ADHD were more likely to be symptomatic non-dependant (showing some symptoms but not meeting the diagnostic threshold for ED) than healthy controls, suggesting increased risk even if diagnostic thresholds are not met. Nevertheless, owing to limited research, the underlying mechanisms and nature of the relationship between ADHD and ED, remain unclear, warranting further investigation.

However, understanding the motivation behind exercise behaviour may shed light on the relationship between ADHD and ED, where different types of motivation have been linked with both increased PA [16, 17] and ED [18, 19]. For example, more autonomous motivations (self-determined motivations/freely initiated by the individual) have been linked with greater exercise frequency, duration, intensity [16], and adherence [17]. Motivation is also recognised as critical factor for ADHD interventions, enhancing engagement and intervention effectiveness [20, 21], and intrinsic motivation has been identified as a critical component in enhancing intervention effectiveness in general [22]. Notably, individuals with ADHD typically have lower intrinsic motivation in academic related work [23], but the relationship between intrinsic motivation and PA within ADHD remains unexplored. Regarding ED, introjected motivation (a form of extrinsic motivation) has been found to positively correlate with, and predict, levels of ED [18, 19] Thus, understanding such associations in the context of ADHD can inform both the ED-ADHD relationship and potential exercise interventions for ADHD.

Such research is also particularly important for adults with ADHD, where motivation plays a key role in their engagement with interventions [21]. Unlike children, adults with ADHD may lack extrinsic motivations and structural systems such as parental or school oversight, necessitating greater reliance on intrinsic motivations to sustain engagement and participation. Additionally, adulthood has been identified as a critical stage for ED risk to manifest [24], yet the relationship between PA and ADHD in adults remains underexplored [8]. For example, Den Heijer er al. [7] reviewed 25 studies but found that only four [9, 10, 25, 26] focussed on adult participants.

Therefore, the aim of this study was to investigate the relationships between ADHD symptomatology, motivation, and ED within an adult population. Based on previous literature, it was hypothesised: 1) there would be a significant positive correlation between ADHD symptomatology and ED; 2) there would be a significant correlation between ADHD symptomatology and autonomous exercise motivation; 3) there would be a significant correlation between ED and autonomous exercise motivation.

Participants

Participants were recruited online via social media adverts, email invitations, Survey Circle [27], and a University Psychology participant pool system. The study was advertised as “Physical activity, motivation and behaviour in adults”, and participants were not offered any reimbursement or financial compensation for taking part. To be eligible to take part, participants had to be at least 18 years of age and resident in the United Kingdom. G-Power calculations indicated that for 95% power, with a rejection criterion of p < .05, and medium effect size (r = .3), a minimum sample size of 138 participants was required.

Of the 207 participants who accessed the survey, 55 were subsequently excluded for the following reasons: not being a resident of the UK (n = 8); not completing the study in full (n = 17); not correctly answering attentional control checks (n = 13); having duplicate responses to another respondent with same IP address (likely same participant responding twice; n = 3); scoring above the maximum limit for the IPAQ-L (n = 10); and counting as extreme outliers for the IPAQ-L Leisure domain score (defined as exceeding the 3rd quartile of TFE score + 3*IQR; n = 4).

This left 152 participants, of whom 122 (77.63%) were female. Ages ranged from 18 to 62 (M = 21.59, SD = 5.74) years, with further demographic information presented in Table 1. Options for the Gender and Ethnicity question were based on the categories recommended by the UK government [28], with gender referring to the socially constructed roles, behaviours, and identities of female, male and gender-diverse people.

Table 1

*Demographic Information of the Study Sample (*N *= 152)*
Demographics	n (%)
Gender
Male	22 (14.47)
Female	118 (77.63)
Other	9 (5.92)
Prefer not to say	3 (1.97)
Ethnicity
White/English/Scottish/Northern Irish/British	109 (71.71)
Arab	3 (1.97)
Any other white background	14 (9.21)
White and Black African	1 (0.66)
White and Asian	5 (3.29)
Any other Mixed/Multiple ethnic background	3 (1.97)
Indian	5 (3.29)
Pakistani	3 (1.97)
Bangladeshi	3 (1.97)
Chinese	3 (1.97)
African	3 (1.97)
Highest Level of education
Secondary/Highschool (e.g. GCSE’s) or equivalent	3 (1.97)
College/Sixth form (e.g. BTEC, A-Levels) or equivalent	82 (53.95)
University (e.g. BSc, BA, Degree)	53 (34.87)
Masters (e.g. MSC, MA)	10 (6.58)
PhD/Doctorate	2 (1.32)
Other	2 (1.32)
Employment Status
Full-time employment	12 (7.89)
Part-time employment	19 (12.5)
Unemployed as unable to work	1 (0.66)
Currently unemployed but looking for work	4 (2.63)
Full-time student	107 (70.39)
Part-time student	6 (3.95)
Unpaid voluntary work	1 (0.66)
Other	2 (1.32)

Materials

The survey was administered on Qualtrics [29]. Participants supplied standard demographic details (e.g., age, gender, ethnicity) and responded to questions about comorbid conditions (e.g., “Have you ever been diagnosed with any of the following conditions/disorders” [10 options, such as: none; mood disorder; sleep disorders; other; etc], “If so do you take any prescribed medication?”). Standardised questionnaires were also included:

The ADHD Rating Scale – IV with Adult prompts (ADHD-RS-IV) [30]: Based on DSM-IV criteria and used as a measure of ADHD symptomology, the ADHD-RS-IV is an 18-item self-report questionnaire assessing the frequency and severity of ADHD in adults. Each item consists of prompts (e.g. “do you rush through work or activities”) corresponding to specific ADHD symptoms (e.g. carelessness). Respondents rate each item based on the extent to which the prompts affect them on a scale of 0–3 (0 = None; 1 = Mild; 2 = Moderate; 3 = Severe). The score for each item is determined by the highest score generated by the prompts in that item (e.g., if one prompt in an item generates 3, but the rest generate 2, the item score is 3). The scale has two subscales: inattentive (ADHD-I; 9 items) and hyperactive (ADHD-H; 9 items) ADHD symptoms. Responses from each subscale can be summed to yield scores ranging from 0–27, with higher scores indicating greater severity of each ADHD subtype. Combining subscale scores provides a composite ADHD score (ADHD-T) ranging from 0–54, with higher scores indicating greater severity of ADHD symptoms. Within the current sample, the ADHD-RS-IV demonstrated good to excellent internal reliability, with Cronbach’s Alpha coefficients of .88 for inattentive (items one-nine), .85 for hyperactive (items 10–18), and .9 for ADHD-T (items 1–18).

The International Physical Activity Questionnaire – Long Format, last seven days, self-administered (IPAQ-L); [31]: A 27-item questionnaire measuring PA over the last seven days. Data can be converted to Metabolic Equivalents (MET), representing the amount of oxygen consumed by the body while sitting at rest. Accompanying guidelines recommend a maximum weekly exercise duration of 16 hours per day, resulting in METs range of 0 to 53,760. Test-retest reliability clusters around .80, with a criterion validity correlation of about 0.3, consistent with other self-report scales [31]. This questionnaire was chosen due to its established use in related research, its measurement of METs (a commonly used outcome variable facilitating cross-study comparisons), and its comprehensive assessment of various domains (Work, Travel, Domestic and Leisure). Total IPAQ-L score (sum of all four domain sub scores) measured PA level, while the Leisure domain indexed PE level.

Behavioural Regulation in Exercise Questionnaire – version 3 (BREQ-3) [32]: A 24 item self-report questionnaire assessing external, introjected, identified, and intrinsic forms of regulation of exercise behaviour and motivation. Participants rate statements about their exercise attitudes (e.g. “I feel guilty when I don’t exercise”) on a four-point scale (0 = not true of me, 4 = very true of me). Mean scores for set items generate scores for six subscales: Amotivation (lacking any intention to engage in exercise); External Regulation (exercise is only done to comply with external pressures); Introjected Regulation (external pressures are internalised); Identified Regulation (conscious acceptance of exercise as important for personally valued outcomes); Integrated Regulation (internalisation of Identified regulation, so exercise is confluent with one’s sense of self); and Intrinsic Regulation (fully self-determined, exercise is appreciated for the experience and extrinsic outcomes are not concerned). Each subscale score can then be weighted to generate the Relative Autonomy Index (RAI), indicating the degree of self-determination. Subscale scores range from 0–4, and RAI ranges from − 24 to 24. The BREQ-3 has adequate internal consistency (McDonald’s Ω ranging between .73 to .91) [33], with good to excellent levels in the current sample also (Cronbach α range .84 to .94).

Exercise Dependence Scale – Revised (EDS-R) [34]: A 21-item self-report questionnaire assessing the degree of exercise dependence based on DSM-IV criteria for substance dependence. Participants rate statements regarding their beliefs and behaviours with exercise (e.g., “I exercise longer than I expect”) over the past three months on a six-point Likert scale (1 = never, 6 = always). Ratings provide a total ED score (EDS-T), which comprises the sum of ratings for seven subscales: Withdrawal (manifested by withdrawal symptoms for exercise); Continuance (Exercise continues even though a problem is likely to have been caused/exacerbated by exercise); Tolerance (need for increased amounts of exercise to achieve desired effect, or experiencing diminished effects with same amount of exercise); Lack of Control (desire/unsuccessful effort to reduce/control exercise); Reductions in Other Activities (ROA) (reduction or abandonment of activities due to exercise); Time (amount of time engaging in exercise); and Intention effects (exercising in larger amounts/over a longer period than intended). Subscale scores range from 3–18 and 21–126 for EDS-T. The scale has excellent test-retest reliability (R = .95), and most subscales show adequate to good internal consistency (Cronbach α ranging between .78 to .92) [34]. Within the current study, internal consistency was excellent (Cronbach α = .96).

The Adult ADHD Self Report Scale – v1.2 (ASRS-5) [35]: A 6-item tool rated on a 5-item Likert (0 = never to 4 = very often), with a total score threshold of 14 or above indicating subclinical ADHD. Although included in the survey, data from the ASRS-5 was not utilised in the analysis as the ADHD-RS-IV generated usable subclinical ADHD subgroups, including those based on subtype.

Design

The study was a cross-sectional observational study. The primary variables of interest were ED (EDS-T); motivation (RAI); ADHD symptomatology (ADHD-T, ADHD-I, ADHD-H) and PE Level.

Procedure

The research was approved by the Swansea University Psychology Department Research Ethics Committee (Approval Number: 1 2023 6008 5223). Participants accessed the survey via Qualtrics and were presented with a detailed information page, followed by an electronic consent page (Yes/No checkbox). Non-consenting participants were directed to the end of the survey and were excluded from data analysis. This ensured that all participants included within data analysis were 18 years old or over and had provided informed consent. Consenting participants then completed demographic questions, followed by questions on comorbidities, and then the ASRS-5. They were then presented with the ADHD-RS-IV; BREQ-3; EDS-R; and IPAQ-L in a randomised order. Finally, participants were presented with a debrief page. All survey sections required a response (except for the questions in the IPAQ-L a participant would have skipped given responses), ensuring complete participation. There was no time limit, and the average completion time, excluding outliers, was 15 minutes and 34 seconds.

Data Processing and Statistical Analysis

Statistical analyses were carried out using IBM SPSS Statistics version 28 and version 29 [36]. JASP 0.18.3 [37] was used to generate a correlation matrix.

Attentional check items were included in the ASRS-5, BREQ-3, and EDS-R (e.g., “Please select option “4””). Participants who failed these simple attentional check items were excluded from further analysis.

IPAQ-L data underwent cleaning following guidelines set by the IPAQ group. Participants were also grouped into non-ADHD or subclinical ADHD participants based on their ADHD-RS-IV scores: participants responding 2–3 on 6 or more inattentive prompts but 5 or less hyperactive prompts were categorised as meeting the subclinical threshold for inattentive ADHD diagnosis (Sub-Inattentive). The reverse was applied for hyperactive ADHD (Sub-Hyperactive). Participants meeting both thresholds were categorised as combined ADHD (Sub-Combined), while those meeting neither were categorised as Non-ADHD. A final group comprised participants who met any subclinical ADHD threshold (Sub-ADHD).

Most study variables (ADHD-T, ADHD-I, ADHD-H, RAI) were normally distributed (skewness, histograms). However, EDS-T and PA level required transformation (EDS-T by logarithmic transformation; PA level by square root transformation) to meet normal distribution assumptions. Despite transformation, PE; BREQ-3 and EDS-R remained non-normally distributed. Therefore, analyses involving these variables utilised nonparametric tests with the untransformed variables. Additionally, demographic factors and comorbidities did not show significant correlations/differences with multiple key variables, so were not used as control variables.

Pearson’s correlations were performed to assess relationships between normally distributed variables. Correlations between ED/Motivation and ADHD symptomatology were then controlled for by PE level (semi-partial correlations), to ensure any correlations ED/Motivation and ADHD symptomatology were not a result of PE level. For correlations involving non-normally distributed data or controlling for PE, non-parametric semi-partial correlations were utilised. This involved generating residuals of independent variables from PA level and then performing Kendall Tau’s correlations between the residual independent and dependent variables. Additionally, to account for multiple comparisons, Bonferroni-Holm corrections were applied to correlations between ADHD symptomatology and both ED (seven) and Motivation (six) subscales.

Descriptive Results

Descriptives are presented in Table 2. Significant correlations were observed between most key variables, including significant positive correlations between: PA level and ADHD-H, RAI, and EDS-T; between PE level and ADHD symptomatology (Total and Hyperactive), RAI, and EDS-T; between RAI and EDS-T; and between EDS-T and ADHD symptomatology (Total and Hyperactive). Additionally, a significant negative correlation was found between RAI and Inattentive ADHD symptomatology.

Table 2

*Descriptive Statistics and Correlation Coefficients (Pearson’s* r) for Study Variables (N *= 152)*
Variable	M	SD	1	2	3	4	5^a	6
1. ADHD-T	22.84	10.61	-	-	-	-	-	-
2. ADHD-I	12.27	6.05	.89**	-	-	-	-	-
3. ADHD-H	10.57	5.87	.89**	.59**	-	-	-	-
4. PA level	65.33 (3702.5)^b	37.79 (7490.63)^b	.14	> .01	.24**	-	-	-
5. PE level^a	(594)^b	(1334.5)^b	.19*	.05	.27**	.71**	-	-
6. RAI	7.71	7.75	− .11	− .19*	− .01	.18*	.34**	-
7. EDS-T	1.67 (47)^b	0.18 (31)^b	.23**	.1	.31**	.34**	.47**	.59**
Note. Statistical significance: p < .05; *p < .01.

^a Correlation coefficients represent Pearson’s r, except for coefficients with PE level which represent Spearman’s rho.

^b Brackets represent the median and IQR of non-normally distributed variables/untransformed variables below the mean and SD of the transformed variable.

Table 3

*Distribution of Subclinical Diagnosis of ADHD in Relation to Gender* (N = *152)*
Gender	Non-ADHD	Sub-ADHD	Sub-Inattentive	Sub-Hyperactive	Sub-Combined
	n (%)	n (%)	n (%)	n (%)	n (%)
Male	12 (54.55)	10 (45.45)	7 (31.82)	0 (0)	3 (13.64)
Female	79 (73.75)	39 (33.05)	17 (14.41)	8 (6.78)	14 (11.86)
Other	4 (44.44)	5 (55.56)	1 (11.11)	1 (11.11)	3 (33.33)
Prefer not to say	0 (0)	3 (100)	2 (66.67)	0 (0)	1 (33.33)
Total	95 (62.5)	57 (37.5)	27 (17.76)	9 (5.92)	25 (13.82)

As shown in Table 3, a slightly higher proportion of males (45.45%) had subclinical ADHD compared to females (33.05%), although this difference was not statistically significant, χ2(1) = 1.25, p = .263 (12 participants who selected ‘other’ or ‘prefer not to say’ were excluded from this analysis). Additionally, sub-inattentive subtype had the highest prevalence (17.76%), followed by Sub-combined (13.82%), and then Sub-hyperactive (5.92%).

ADHD Links with Exercise Dependence

A semi-partial correlation was conducted to determine the relationship between EDS-T (dependant variable) and ADHD symptomatology (independent variable), whilst controlling for PE. There was a significant positive semi-partial correlation between EDS-T and ADHD-T (τ_b [149] = .1, p = .038); a non-significant positive semi-partial correlation between EDS-T and ADHD-I (τ_b [149] = .05, p = .189); and a significant positive semi-partial correlation between EDS-T and ADHD-H (τ_b [149] = .14, p = .005).

Table 4

*Semi-Partial Correlations (Kendall’s Tau) Between ED, ADHD Symptomatology and RAI While Controlling for PE (*N *= 152)*
Variable	ADHD-T	ADHD-I	ADHD-H	RAI
EDS-T	.1*	.05	.14**	. 37**
Withdrawal	.15*	.08	.2**	.32**
Continuance	.19**	.14*	.22**	.18**
Tolerance	− .02	− .07	.06	.42**
Lack of Control	.13*	.12	.14*	.24**
ROA	.14*	.1	.16*	.29**
Time	.02	− .03	.08	.41**
Intention Effects	.1	.05	.14*	.27**
Note. Statistical significance: p < .05; *p < .01 (after corrections have been applied as mentioned in the Methods).

Table 4 presents the semi-partial correlations between ED (and its subscales) and ADHD symptomatology, while controlling for PE. Notably, ADHD-H was consistently more positively correlated with all ED subscales compared to ADHD-I. Among the subscales, Tolerance and Time were the only subscales not significantly positively correlated with ADHD-H, and ADHD-T was also not significantly positively correlated with Intention Effects. Conversely, ADHD-I was only significantly positively correlated with Continuance.

ADHD Links with Motivation

A semi-partial correlation was conducted to determine the relationship between RAI (dependant variable) and ADHD symptomatology (independent variable), whilst controlling for PE. Significant negative semi-partial correlations were found between RAI and both ADHD-T (τ_b [149] = − .13, p = .023) and ADHD-I (τ_b [149] = − .15, p = .008). A non-significant negative correlation was found between EDS-T and ADHD-H (τ_b [149] = − .06, p = .275).

Table 5

*Semi-partial Correlations (Kendall’s Tau) Between Motivation, ADHD Symptomatology and EDS-T While Controlling for PE (*N *= 152)*
Variable	ADHD-T	ADHD-I	ADHD-H	EDS-T
RAI	− .13*	− .15**	− .06	.37**
Amotivation	.1	.11	.07	− .09
External Regulation	.16*	.14	.15	.08
Introjected Regulation	.08	.07	.05	.28**
Identified Regulation	.01	− .02	.04	.45**
Integrated Regulation	− .08	− .12	<-.01	.46**
Intrinsic Regulation	− .08	− .09	− .02	.38**
Note. Statistical significance: p < .05; *p < .01 (this is after corrections have been applied as mentioned in the Methods).

Table 5 shows the partial correlations between motivation (and subscales) and ADHD symptomatology, while being controlled for by PE. After controlling for PE, the negative correlation between ADHD-T and RAI became significant, contrasting to the correlation without control (Table 2). ADHD-H was consistently more positively correlated with motivation subscales positively contributing to relative autonomy (Identified, Integrated and Intrinsic Regulation) than ADHD-I was, while being generally more negatively correlated with subscales that negatively contribute to relative autonomy (Introjected Regulation and Amotivation). However, External Regulation was the only subscale significantly correlated with ADHD symptomatology, with a significant positive correlation with ADHD-T but not ADHD-I or ADHD-H.

Exercise Dependence Links with Motivation

A semi-partial correlation was conducted to determine the relationship between EDS-T (dependant variable) and RAI (independent variable) whilst controlling for PE. There was a significant positive partial correlation between EDS-T and RAI (τ_b [149] = .37, p < .001). Further, all motivation subscales were significantly positively correlated with EDS-T, except for Amotivation and External Motivation (Table 5). Additionally, all ED subscales were significantly positively correlated with RAI (Table 4).

The aim of this study was to investigate the relationships between ADHD symptomatology with motivation and ED. Based on previous literature [14, 15], it was hypothesised that there would be a significant positive correlation between ADHD symptomatology and ED. The current findings support this, as evidenced by the significant positive correlation between EDS-T and both ADHD-H and ADHD-T (along with a non-significant positive correlation with ADHD-I). Like Popat et al., [15] a significant positive correlation was found between ADHD symptoms and Withdrawal. However, in our study, we also found significant positive correlations between ADHD-T and total ED, Continuance, Lack of control, and ROA, as well as intention effects with ADHD-H specifically. This suggests that measuring ADHD symptoms on a continuous scale provides greater insight compared to simply testing mean differences between groups. Additionally, subdividing ADHD symptomatology by subtype enabled a more nuanced analysis, revealing a particularly strong positive correlation between ED and hyperactive ADHD symptoms over inattentive symptoms, something not previously explored. This finding cannot solely be attributed to hyperactive symptoms being linked to higher exercise level (as PE was controlled for), suggesting that hyperactive symptoms may be more commonly linked with addictive behaviours [38] and different motivations.

In support of hypothesis two and aligning with prior work (Smith & Langberg, 2018), significant negative correlations were observed between RAI and both ADHD-T and ADHD-I symptoms. Further analysis of subscales suggests that this may be largely attributable to a significant positive correlation between the External Regulation subscale (an extrinsic motivation) and total ADHD symptomatology. Regarding ADHD symptom presentation, autonomous/intrinsic exercise motivations generally demonstrated negative correlations with inattentive ADHD symptoms, while displaying slightly more positive/less negative correlations with hyperactive ADHD symptoms. Conversely, extrinsic motivations (except External Regulation) showed a reverse pattern. This could also partially explain hyperactive symptoms having a more positive relationship with ED, as RAI was also found to have positive correlations with ED.

Regarding hypothesis three, ED was positively associated with autonomous exercise motivations, with significant positive correlations observed between EDS-T and RAI, as well as most motivation subscales. This contrasts previous work [18, 19] reporting that greater introjected motivation specifically was the main or only predictor of ED, which would then also potentially suggest a negative correlation between ED and autonomous motivation (e.g., RAI; due to Introjected Regulation counting as an extrinsic motivation, that is negatively weighted in RAI). Additionally, while Introjected Regulation was positively correlated with ED, it was less positively correlated than all three intrinsic motivation forms (Identified, Integrated and Intrinsic Regulation). However, Hamer et al., [18] previously noted Identified Regulation (an intrinsic motivation) as the second biggest predictor of ED after Introjected, potentially balancing out the correlation between ED and RAI. Indeed, in this study, Identified Regulation had a strong correlation with ED, although the strongest correlation (albeit marginally) was observed for Integrated.

One possible explanation for the discrepancy in findings, could be differences in sample demographics. Unlike Hamer et al. [18] who recruited endurance athletes, participants from the general population were sampled here. It is plausible that endurance athletes had reached a “ceiling effect” with their engagement and motivations, where positive associations become harder to detect. Additionally, the older version of the BREQ [39] was used which did not include Integrated Regulation, precluding comparison with data from the current study. Regarding Edmunds et al. [19], their research also treated ED as a categorical rather than a continuous variable and lacked enough participants categorised as “at risk” to perform analyses on all three groups (at risk, nondependent-symptomatic, nondependent-asymptomatic). Additionally, neither study controlled for the total amount of PE performed, meaning any positive relationships found between ED and motivation forms might be due to the form of motivation being more likely to increase PE level (acting as a mediator), which could then be the direct factor leading to increased ED risk (and vice versa). This means that an analysis of the different motivation forms when PE level is held constant cannot be performed.

Contrasting prior work [10], an unexpected finding was the positive correlation between PE level and ADHD symptomatology, which did not change significantly when the sample was split by sub-group (See Additional File 1). However, correlations between PE and ADHD symptomatology were only significant within the non-ADHD group; within the Subclinical-ADHD group, they were less positive and lost significance. This discrepancy with previous research may partially stem from differences in recruitment strategy: Abramovitch et al., [10] recruited individuals with a pre-existing diagnosis of ADHD, whereas the current study recruited from the general population, potentially resulting in a more positive correlation due to the inclusion of non-ADHD participants. However, even within the subclinical ADHD group of the current study, only ADHD-I had a negative correlation with PE, which was non-significant. This suggests that recruitment strategy alone cannot fully explain the findings, highlighting the need for further research to more effectively understand these associations.

In contrast to previous studies [3, 40], we found no significant gender difference in subclinical ADHD diagnosis rates, supporting the notion of potential underdiagnosis of ADHD in females [41] Further, whilst the prevalence rate of ADHD subgroups in our sample, particularly hyperactive presentation, generally aligns with previous research, we found a slightly higher prevalence of the Sub-Inattentive (17.76%) than Sub-Combined (13.82%), which is typically reported to be the most prevalent [42]. This finding may support the notion that the inattentive presentation is more likely to be underdiagnosed due to its less disruptive nature [41]. Regarding gender differences, females had a lower proportion of Sub-Inattentive compared to males, contrary to previous findings suggesting that girls are more likely than boys to have the inattentive presentation [3, 43]. Nevertheless, research also indicates that disparities in presentation rates between males and females diminish after adolescence [43] consistent with our study where all participants were at least 18 years of age.

The findings of this study have significant implications, particularly regarding the potential risks associated with using exercise as an intervention for ADHD. Higher levels of ADHD symptoms in our adult sample were found to be linked with higher risks of ED. Therefore, any potential implementation of exercise as an intervention for ADHD, particularly hyperactive, should carefully weigh the potential benefits against the risks of ED and associated harms. While the potential risk appears to be comparatively lower for inattentive ADHD, combined or hyperactive presentations warrant additional consideration, as ADHD-H was positively correlated with all subscales but Tolerance and Time. Such findings underscore the importance of regularly monitoring such interventions to offset the risk of unhealthy behaviours developing. Additionally, considering an individual’s background and motivations is also important, as these factors may also influence ED risk.

The findings regarding the relationship between RAI and ADHD symptomatology suggest that the lower intrinsic motivation observed in children with ADHD, as seen in academic settings [23], may also extend to other contexts (i.e., exercise) and across different age groups. As intrinsic motivation appears to be lower within ADHD subjects (and that it has a positive link with PE), efforts focused around increasing intrinsic exercise motivations for individuals with ADHD could effectively increase exercise engagement. This suggests that intervention techniques suggested by Smith and Langberg [23] to enhance intrinsic motivation for academic tasks, should also be applied exercise interventions for adults, particularly those with the inattentive presentation.

However, caution is needed as while our study found that RAI was positively correlated with higher PE level, it was also positively correlated with EDS-T, even after controlling for PE. This suggests that while increasing intrinsic motivations might lead to greater engagement with any exercise intervention, it might also elevate risk of ED development. Therefore, if exercise interventions seek to enhance engagement by improving autonomous motivation, caution should be exercised concerning ED risk, with careful monitoring and more thorough analysis to discern their suitability for prospective individuals, taking into consideration relevant risk factors [44].

However, the viability of any potential exercise intervention primarily depends on whether exercise reduces ADHD symptoms and yields positive effects. The positive correlation between ADHD symptomatology and PE level could imply the opposite, that PA and PE might increase ADHD symptoms. Although, due to the correlational nature of this study, causation cannot be determined. Furthermore, the possibility that PE may exacerbate symptoms of ADHD contradicts much of the literature [7, 8]. Such an effect would also contrast most existing literature where there is aetiological evidence supporting the beneficial, rather than detrimental, effects of PA/PE on ADHD symptoms [45]. An alternative explanation could be that individuals with greater ADHD symptoms are more likely to engage in PA, and while this contrasts with findings from previous literature [10], there is less conflicting evidence, and this could be a partial result of post-pandemic changes. While this would mean that the positive effect of PE on ADHD symptomatology would still be present, it could mean that individuals with ADHD are already exercising more than the general population, making any possible exercise intervention redundant, or elevating the potential risk of overexercising and ED. However, without further analysis and research on the topic post-pandemic, it is hard to determine such implications with confidence.

Additionally, the current study is not without limitations. Data was gathered from self-report measures, meaning that results might not reflect true behaviours/feelings – particularly for motivation and ED. Individuals with ED might not acknowledge this and play down any maladaptive behaviours/feelings they have when answering the EDS-R. Similarly, individuals who have higher extrinsic motivations may internalise them and portray them as internalised motivations to feel better.

The sample composition, skewed towards females, students, and younger ages, also means that findings may be less generalisable to the general population. For example, as students are generally found to have higher rates of ADHD [46], the prevalence of adult ADHD symptoms in this study may not accurately represent those of the general population. Moreover, the proportionally smaller sample size of males in the current sample made comparison between males and females harder to conduct.

The use of using RAI as a continuum scale for self-determination has also been criticised [47] with theoretical and statistical limitations regarding the reduction of different dimensions into a single scale. Therefore, analysis of the individual motivation subscales would have provided greater insight and accuracy. Although, analysis of individual the BREQ-3 subscales largely support the overall findings and implications of the current study. There was a significant negative correlation between External Regulation and total ADHD symptomatology, while all three intrinsic forms of motivation were positively correlated with ED (Table 5) and PE (See Additional File 2). However, the positive correlation between External Regulation and ADHD-I was not significant (in contrast to RAI which had a significant negative correlation with both ADHD-T and ADHD-I), suggesting that differences between inattentive and hyperactive symptoms might be less pronounced when solely considering RAI. Additionally Introjected Regulation (an extrinsic motivation) was also significantly correlated with ED and PE, albeit weaker compared to the three intrinsic forms, underscoring the risk of treating all extrinsic motivations as equivalent, as Extrinsic and Introjected Regulation might have different relationships and effects.

Finally, and as referred to previously, the correlational nature of the study means that causation between study variables cannot be inferred. Whilst significant associations were identified, the nature of these relationships, as well as the potential mechanisms underlying each, remain speculative. Consequently, the implications of the current study would benefit from further empirical validation via more experimental designs. As requisite examples, future research endeavours could focus on developing interventions tailored to improve/develop intrinsic exercise motivation, possibly via motivational interviewing, as previously explored with individuals with ADHD [48]. Such interventions could be designed to monitor their impact on both PA level and ED. Research could also investigate ways in which intrinsic exercise motivation could be encouraged/improved, without increasing the risk of ED in individuals with ADHD (and possibly individuals with hyperactive ADHD particularly). To maximise the efficacy and safety of such interventions, it would be beneficial to consider and address potential risk factors (e.g., body dissatisfaction, drive for thinness, bulimia), while supporting and increasing protective factors (e.g., self-esteem) [44].

In conclusion, the findings of this study suggest that ADHD symptomatology in adults is linked with reduced intrinsic exercise motivation and higher risks of ED. Additionally, while intrinsic exercise motivation may be linked with increased PE level, it is also linked with increased ED, suggesting that caution should be applied to any intervention focused around increasing intrinsic exercise motivation. Finally, the relationship between ADHD and PE may be more complex than previously understood, especially considering the potential impact of post-Covid-19 effects. Overall, while there is potential to effectively increase exercise engagement within ADHD by focusing on intrinsic exercise motivation, any such intervention needs to be carefully considered and monitored to reduce risks and optimise effectiveness.

ADHD: Attention Deficit Hyperactive Disorder

PA: Physical Activity

PE: Physical Exercise

EA: Exercise Addiction

ED: Exercise Dependence

RAI: Relative Autonomy Index

ROA: Reduction in Other Activities

Ethics approval and consent to participate

Consent for publication

Not applicable.

Availability of data and materials

The (anonymous) datasets generated and analysed during the current study are available in the OSF repository, https://osf.io/8vxq2/?view_only=08a3b17735044671b19265518e1c23e8

Competing interests

The authors declare that they have no competing interests.

Funding

Not applicable.

Authors' contributions

All authors contributed to conceptualisation, design, methodology, project administration, and analysis and interpretation of data. RT oversaw data collection and curation, while CW and PR provided supervision. RT drafted the original manuscript, and all authors made critical revisions. All authors read and approved the final manuscript.

Acknowledgements

Not applicable.

Use of Generative AI and AI-assisted Technologies

During the preparation of the manuscript, the authors used ChatGPT/OpenAI to assist with succinctness of written expression. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for its contents.

American Psychiatric Association. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders Fifth Edition. Arlington. 2013.
Fayyad J, Sampson NA, Hwang I, Adamowski T, Aguilar-Gaxiola S, Al-Hamzawi A, et al. The descriptive epidemiology of DSM-IV Adult ADHD in the World Health Organization World Mental Health Surveys. ADHD Attention Deficit and Hyperactivity Disorders. 2017;9(1).
Willcutt EG. The Prevalence of DSM-IV Attention-Deficit/Hyperactivity Disorder: A Meta-Analytic Review. Neurotherapeutics. 2012 Jul 15;9(3):490–9.
Goode AP, Coeytaux RR, Maslow GR, Davis N, Hill S, Namdari B, et al. Nonpharmacologic treatments for attention-deficit/hyperactivity disorder: A systematic review. Pediatrics. 2018 Jun 1;141(6).
Parker J, Wales G, Chalhoub N, Harpin V. The long-term outcomes of interventions for the management of attention-deficit hyperactivity disorder in children and adolescents: A systematic review of randomized controlled trials. Psychol Res Behav Manag. 2013 Sep 16;6:87–99.
Wu EQ, Hodgkins P, Ben-Hamadi R, Setyawan J, Xie J, Sikirica V, et al. Cost effectiveness of pharmacotherapies for attention-deficit hyperactivity disorder: A systematic literature review. CNS Drugs. 2012;26(7):581–600.
Den Heijer AE, Groen Y, Tucha L, Fuermaier ABM, Koerts J, Lange KW, et al. Sweat it out? The effects of physical exercise on cognition and behavior in children and adults with ADHD: a systematic literature review. J Neural Transm. 2017;124:3–26.
Vysniauske R, Verburgh L, Oosterlaan J, Molendijk ML. The Effects of Physical Exercise on Functional Outcomes in the Treatment of ADHD: A Meta-Analysis. J Atten Disord. 2020 Mar 1;24(5):644–54.
Fritz KM, O’Connor PJ. Acute exercise improves mood and motivation in young men with ADHD symptoms. Med Sci Sports Exerc. 2016 Jun 1;48(6):1153–60.
Abramovitch A, Goldzweig G, Schweiger A. Correlates of physical activity with intrusive thoughts, worry and impulsivity in adults with attention deficit/hyperactivity disorder: A cross-sectional pilot study. Isr J Psychiatry. 2013;50(1):47–54.
Miller PM. Principles of Addiction: Comprehensive Addictive Behaviors and Disorders, Volume 1. Academic Press; 2013.
Trott M, Jackson SE, Firth J, Fisher A, Johnstone J, Mistry A, et al. Exercise Addiction Prevalence and Correlates in the Absence of Eating Disorder Symptomology: A Systematic Review and Meta-analysis. J Addict Med. 2020;14(6).
Starcevic V, Khazaal Y. Relationships between behavioural addictions and psychiatric disorders: What is known and what is yet to be learned? Front Psychiatry. 2017;8(53).
Colledge F, Buchner U, Schmidt A, Wiesbeck G, Lang U, Pühse U, et al. Individuals at Risk of Exercise Addiction Have Higher Scores for Depression, ADHD, and Childhood Trauma. Front Sports Act Living. 2022;3.
Popat P, Dinu LM, Runswick O, Findon JL, Dommett EJ. Investigating the Relationship Between Attention-Deficit Hyperactivity Disorder, Obligatory Exercise and Exercise Addiction. Int J Ment Health Addict. 2021;21:1365–77.
Duncan LR, Hall CR, Wilson PM, Jenny O. Exercise motivation: A cross-sectional analysis examining its relationships with frequency, intensity, and duration of exercise. Int J Behav Nutr Phys Act. 2010 Jan 26;7(1):1–9.
Teixeira PJ, Carraça E V., Markland D, Silva MN, Ryan RM. Exercise, physical activity, and self-determination theory: A systematic review. Int J Behav Nutr Phys Act. 2012;9.
Hamer M, Karageorghis CI, Vlachopoulos SP. Motives for exercise participation as predictors of exercise dependence among endurance athletes. J Sports Med Phys Fitness 2002;42(2).
Edmunds J, Ntoumanis N, Duda JL. Examining exercise dependence symptomatology from a self-determination perspective. J Health Psychol. 2006;11(6).
van der Oord S, Tripp G. How to Improve Behavioral Parent and Teacher Training for Children with ADHD: Integrating Empirical Research on Learning and Motivation into Treatment. Clin Child Fam Psychol Rev. 2020;23(4).
Weiss M, Safren SA, Solanto M V., Hechtman L, Rostain AL, Ramsay JR, et al. Research forum on psychological treatment of adults with ADHD. J Atten Disord. 2008;11(6).
Medalia A, Saperstein A. The role of motivation for treatment success. Schizophr Bull. 2011;37(SUPPL. 2).
Smith ZR, Langberg JM. Review of the Evidence for Motivation Deficits in Youth with ADHD and Their Association with Functional Outcomes. Clin Child Fam Psychol Rev. 2018;21(4).
Costa S, Hausenblas HA, Oliva P, Cuzzocrea F, Larcan R. The role of age, gender, mood states and exercise frequency on exercise dependence. J Behav Addict. 2013;2(4).
Fuermaier ABM, Tucha L, Koerts J, Van Heuvelen MJG, Van Der Zee EA, Lange KW, et al. Good vibrations - Effects of whole body vibration on attention in healthy individuals and individuals with ADHD. PLoS One. 2014 Feb 28;9(2):e90747.
Fuermaier ABM, Tucha L, Koerts J, van den Bos M, Regterschot GRH, Zeinstra EB, et al. Whole-body vibration improves cognitive functions of an adult with ADHD. ADHD Attention Deficit and Hyperactivity Disorders. 2014;6(3):211–20.
SurveyCircle. Research website SurveyCircle [Internet]. 2016 [cited 2024 Jan 20]. Available from: https://www.surveycircle.com
Race Disparity Unit. Writing about ethnicity - GOV.UK [Internet]. Ethnicity facts and figures. 2021 [cited 2022 Oct 26]. Available from: https://www.ethnicity-facts-figures.service.gov.uk/style-guide/writing-about-ethnicity
Qualtrics [Internet]. Provo, Utah, USA: Qualtrics; 2005. Available from: https://www.qualtrics.com
Adler LA, Spencer TJ, Biederman J, Kollins S, Weisler R, Goodman D, et al. The internal consistency and validity of the Attention-Deficit/Hyperactivity Disorder Rating Scale (ADHD-RS) with adult ADHD prompts as assessed during a clinical treatment trial. J ADHD Relat Disord. 2009;1(1):14–24.
Craig CL, Marshall AL, Sjöström M, Bauman AE, Booth ML, Ainsworth BE, et al. International physical activity questionnaire: 12-Country reliability and validity. Med Sci Sports Exerc. 2003;35(8):1381–95.
Wilson PM, Rodgers WM, Loitz CC, Scime G. “It’s Who I Am … Really!’ The Importance of Integrated Regulation in Exercise Contexts1. J Appl Biobehav Res. 2006;11(2).
Rocchi MA, Wilson PM, Sylvester BD, Guerin E, Shi Z, Sweet SN. Toward Brief Tools Assessing Motivation for Exercise: Rationale and Development of a Six and 12-Item Version of the Behavioral Regulation in Exercise 3 (BREQ3) Questionnaire. Sport Exerc Perform Psychol. 2023;12(3):205–7.
Downs DS, Hausenblas HA, Nigg CR. Factorial validity and psychometric examination of the exercise dependence scale-revised. Meas Phys Educ Exerc Sci. 2004;8(4).
Ustun B, Adler LA, Rudin C, Faraone S V., Spencer TJ, Berglund P, et al. The world health organization adult attention-deficit/hyperactivity disorder self-report screening scale for DSM-5. JAMA Psychiatry. 2017;74(5).
IBM Corp. IBM SPSS Statistics for Windows. Version 29.0. 2019. Armonk, NY: IBM Corp; 2023.
JASP Team. JASP. Version 0.17.1. JASP - Free and User-Friendly Statistical Software. 2023.
De Alwis D, Lynskey MT, Reiersen AM, Agrawal A. Attention-deficit/hyperactivity disorder subtypes and substance use and use disorders in NESARC. Addictive Behaviors. 2014;39(8).
Mullan E, Markland D, Ingledew DK. A graded conceptualisation of self-determination in the regulation of exercise behaviour: Development of a measure using confirmatory factor analytic procedures. Pers Individ Dif. 1997;23(5).
Nøvik TS, Hervas A, Ralston SJ, Dalsgaard S, Rodrigues Pereira R, Lorenzo MJ, et al. Influence of gender on attention-deficit/hyperactivity disorder in Europe - ADORE. Eur Child Adolesc Psychiatry. 2006;15(SUPPL. 1).
Quinn PO, Madhoo M. A review of attention-deficit/hyperactivity disorder in women and girls: Uncovering this hidden diagnosis. Prim Care Companion J Clin Psychiatry. 2014;16(3).
Wilens TE, Biederman J, Faraone S V., Martelon M, Westerberg D, Spencer TJ. Presenting ADHD symptoms, subtypes, and comorbid disorders in clinically referred adults with ADHD. J Clin Psychiatry. 2009 Nov;70(11):1557–62.
Biederman J, Faraone S V., Monuteaux MC, Bober M, Cadogen E. Gender effects on attention-deficit/hyperactivity disorder in adults, revisited. Biol Psychiatry. 2004;55(7):692–700.
Gori A, Topino E, Griffiths MD. Protective and risk factors in exercise addiction: A series of moderated mediation analyses. Int J Environ Res Public Health. 2021;18(18).
Ding Q, Vaynman S, Akhavan M, Ying Z, Gomez-Pinilla F. Insulin-like growth factor I interfaces with brain-derived neurotrophic factor-mediated synaptic plasticity to modulate aspects of exercise-induced cognitive function. Neuroscience. 2006;140(3):823–33.
Gray S, Woltering S, Mawjee K, Tannock R. The adult ADHD self-report scale (ASRS): Utility in college students with attention-deficit/hyperactivity disorder. PeerJ. 2014 Mar 25;2014(1):e324.
Chemolli E, Gagné M. Evidence against the continuum structure underlying motivation measures derived from self-determination theory. Psychol Assess. 2014;26(2).
Sibley MH, Graziano PA, Kuriyan AB, Coxe S, Pelham WE, Rodriguez L, et al. Parent-Teen Behavior Therapy + Motivational Interviewing for Adolescents with ADHD. J Consult Clin Psychol. 2016;84(8).

No competing interests reported.

SupplementaryTable1.docx
Additional File 1.docx. Supplementary Table 1. Descriptive Statistics and Correlation Coefficients for Study Variables split by subclinical ADHD diagnosis.
SupplementaryTable2.docx
Additional File 2.docx. Supplementary Table 2. Correlation Coefficients between PE level and Motivation subscales.

Download PDF

Editorial decision: Revision requested
29 Oct, 2024
Reviews received at journal
25 Oct, 2024
Reviewers agreed at journal
14 Oct, 2024
Reviews received at journal
23 Aug, 2024
Reviewers agreed at journal
16 Aug, 2024
Reviewers invited by journal
12 Aug, 2024
Submission checks completed at journal
06 May, 2024
Editor assigned by journal
06 May, 2024
First submitted to journal
25 Apr, 2024

You are reading this latest preprint version

Understanding the Link Between Physical Exercise, Autonomous Motivation, Exercise Dependence, and Adult ADHD Symptoms: A Cross-Sectional Study

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Background

Methods

Participants

Materials

Design

Procedure

Data Processing and Statistical Analysis

Results

Descriptive Results

ADHD Links with Exercise Dependence

ADHD Links with Motivation

Exercise Dependence Links with Motivation

Discussion

Conclusions

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1