HPA Activity Mediates the Link Between Trait Impulsivity and Boredom

Boredom, a complex emotional state with implications for mental health and well-being, has garnered attention across disciplines, yet remains relatively understudied in psychiatric research. Here, we explored the intricate relationship between trait-impulsivity, stress, and boredom across two studies. Participants completed self-report measures of trait-impulsivity and boredom and boredom-inducing tasks. Study 1, involving 80 participants (42 women and 38 men, aged 20-63), replicates previous findings, by demonstrating that impulsive individuals report greater boredom following a boring task. Study 2 then extends this, using 20 participants (9 women and 12 men, aged 18-24), to show that hypothalamic-pituitary-adrenal (HPA) axis activity, specifically heightened salivary cortisol responses, mediate the link between impulsivity and boredom following a boring task. Collectively, these results demonstrate that HPA axis activity may underline the relationship between trait-impulsivity and boredom by extending previous work and offering a novel insight into potential mechanisms. These findings offer promise for personalised interventions, designed for high impulsivity individuals, to alleviate the negative impacts of boredom and potentially break the identified feedback loop.


Introduction
Boredom has been studied across several disciplines, from psychology and neuroscience [1,2] to sociology [3,4] and philosophy [5,6].It is a complex emotional and psychological state which has been defined as disengagement in a particular task or surroundings, with failed attempts to get involved, despite the motivation to do so [7].Boredom is not merely the "absence of something to do"; it has significant implications for mental health, cognitive functioning, and overall well-being [8][9][10], but remains strikingly understudied in psychiatric research [11].Therefore, understanding the complexity of boredom may inform effective interventions and coping mechanisms for psychiatric disorders.
Impulsivity is a tendency to act on a whim, displaying behaviour characterised by little or no forethought, reflection, or consideration of the consequences [12].Often regarded as a multifaceted construct, impulsivity can be broken down into different aspects or dimensions, including (for example) lack of premeditation, lack of perseverance, sensation seeking, and urgency [13].In fact, impulsivity subtypes are diverse to the point that they often do not intercorrelate, prompting some to reject impulsivity as a psychological concept altogether [14].Impulsivity is a feature of several psychiatric disorders, including Attention-Deficit/Hyperactivity Disorder (ADHD), Borderline Personality Disorder, and Substance Use Disorders [15].As such, treatment often involves both pharmacological and behavioural interventions which can focus on improving impulse control, such as cognitive-behavioural therapy (CBT) [16,17] or pharmacological interventions that act on relevant neurotransmitter systems (e.g., psychostimulants or noradrenergic reuptake inhibitors for ADHD) [18].
Impulsivity can be exacerbated with intense emotional events, such as stressful events [19].
People with high trait-impulsivity have greater boredom propensity [20][21][22][23].There is also evidence which suggests that experiencing boredom leads to increased state-impulsivity [24].In addition to increasing impulsivity, boredom causes a psychophysiological stress responses (e.g., rising heart rate and cortisol levels) [11].Furthermore, people become more impulsive under stress [25].Collectively, this is suggestive of a positive feedback loop between impulsivity, stress and boredom.However, this idea has not been tested substantively.Thus, the present study tests the hypothesis that high trait-impulsivity results in increased levels of boredom due to heightened psychophysiological (i.e., cortisol) response.
Our first aim here was to conceptually replicate previous studies (e.g., Watt & Vodanovich, 1992) that have shown that impulsive individuals are more prone to boredom (Study 1).Meanwhile, our second goal was to test the hypothesis 1 that the relationship between trait impulsivity and state boredom would be mediated by hypothalamic-pituitaryadrenal (HPA) axis activity (Study 2).Specifically, we hypothesised that: (1) individuals with higher levels of impulsivity would report greater levels of subjective boredom following a boring task; and (2) the positive relationship between trait-impulsivity and self-report boredom, following a boring task, would be mediated by heightened HPA axis (i.e., cortisol) responses.

Participants
Participants were 80 adults (42 women, 38 men) ranging in age from 20 to 63 years old (M = 32.34,SD = 10.12) who were recruited using Prolific Academic (https://www.prolific.co)and reimbursed at a rate of £6/hour.Participants were required to be aged 18 or older and have English as a first language or an equivalent level of fluency.Detailed demographic statistics are presented in Table S1 (Supplementary Material).The study was approved by the University of Portsmouth Science and Health Faculty Ethics Committee 1 These analyses were not pre-registered and should be treated as exploratory, rather than confirmatory.

The Shortened UPPS-P Impulsive Behaviour Scale (S-UPPSP) (Cyders, 2015)
The S-UPPSP was used to assess negative urgency (i.e. the tendency to act rashly under extreme negative emotions), lack of premeditation (i.e. the tendency to act without thinking), lack of perseverance (i.e. the inability to remain focused on a task), sensation seeking (i.e. the tendency to seek out novel and thrilling experiences), and positive urgency (i.e. the tendency to act rashly under extreme positive emotions).When completing the S-UPPSP, participants rate several items related to their impulsive behaviour using a four-point Likert-type scale (1 = Agree strongly, 2 = Agree some, 3 = Disagree some, 4 = Disagree strongly).The maximum score on each subscale is 16 as each subscale comprises four items.Higher scores reflect greater levels of impulsivity.In this study, Cronbach's α ranged from 0.72 to 0.90.

The Barratt Impulsiveness Scale (BIS-11) (Patton et al., 1995)
The BIS-11 is one of the most widely used measures designed to assess impulsivity (Stanford et al., 2009).When completing the BIS-11, participants rate 30-items related to the way that they think or act on a four-point Likert-type scale (1 = Rarely / Never, 2 = Occasionally, 3 = Often, 4 = Almost always / Always).The BIS-11 is made up of six first-order factors (attention, cognitive instability, motor impulsivity, perseverance, self-control, cognitive complexity) and three second-order factors (attentional impulsivity, motor impulsivity, nonplanning).Either the total score or subscale scores can be used to assess impulsivity, whereby higher scores reflect greater impulsivity.In this study, Cronbach's α ranged from 0.72 to 0.88.

Boredom Task
The 'peg turning task' has shown to reliably induce boredom (Markey et al., 2014).During the peg-turning task, participants repeatedly clicked on icons of pegs that were arranged in two rows of four.Each mouse click rotated the peg 45° clockwise.Participants were only able to click and rotate a peg when it was highlighted in yellow.The highlighted peg was randomly chosen by the computer.Once highlighted, participants had 15 seconds to click and rotate the specified peg and a countdown timer was presented for the final three seconds of each trial.Participants were unable to move to the next trial until the full 15 seconds had elapsed and the task lasted approximately five minutes.

Multidimensional State Boredom Scale (MSBS) (Fahlman et al., 2013)
The MSBS was used to measure boredom.The MSBS is a 29-item scale with good psychometric properties which can be used to assess boredom using either the total score or across five subscales: disengagement, high arousal, low arousal, inattention, and time perception.Participants complete the MSBS using a seven-point Likert scale (1 = Strongly disagree, 2 = Disagree, 3 = Somewhat disagree, 4 = Neutral, 5 = Somewhat agree, 6 = Agree, 7 = Strongly agree).Higher scores reflect greater levels of boredom with a maximum score of 203.In this study, the Cronbach's α ranged from 0.86 to 0.92.

Procedure
This study was carried out online.After informed consent was confirmed, 500 participants completed the SUPPS-P.An attention check 2 was placed between items 11 and 12 of the SUPPS-P and any participant who failed the attention check was excluded at this stage (n = 4).We then split the data by sex before ranking based on the SUPPS-P total scores.Subsequently, we invited 100 participants at the extreme ends of the distribution (25 highest men, 25 lowest men, 25 highest women, 25 lowest women) to continue participating.Of these, 80 participants completed the full study protocol.They were exposed to the 'pegturning task' before completing the MSBS.Finally, the participants were asked to complete the BIS-11 at a later date.

Analysis
Descriptive statistics (means, standard deviations, and the proportion of missing data) were calculated and bivariate associations (Spearman's rank values were calculated as they require less assumptions to be made about the data are required) were explored for key study variables.There were no missing data.A series of independent (Student's and Welch's3 ) ttests were used to: (a) check whether those in the high impulsivity group had higher scores for each S-UPPSP and BIS-11 facet compared to the low impulsivity group; and (b) to test the hypothesis that higher trait impulsivity would be related to higher level of boredom following the boredom induction task.Results were considered significant when p < .05 or when the 95% confidence intervals (CIs) did not overlap zero.Study 1 analyses were conducted in Stata IC 16.1.
Table 2 shows the sample-wise descriptive statistics (means and standard deviations) and intercorrelations (Spearman's rank values) for key study variables.All measures of selfreport impulsivity were positively intercorrelated (r s = .28-.87, all p s < .05),except for sensation seeking and perseverance, sensation seeking and premeditation, sensation seeking and BIS-11 attentional, and sensation seeking and BIS-11 non-planning (all p s > .05).
Importantly, BIS-11 total score was positively correlated with all measures of boredom (r s = .29-.56, all p s < .01).Therefore, as hypothesised, both the t-test and correlation results suggest that those who are more impulsive are more likely to report greater levels of boredom following a task designed to induce boredom.
As we found moderate-to-strong correlations between measures of self-report impulsivity boredom, whereby high impulsivity individuals reported greater levels of boredom compared to low-impulsive participants, we then investigated a potential mechanism underlying this relationship: HPA axis activity.
Highly impulsive people have been shown to have altered cortisol responses.For example, individuals who are dependent on drugs, such as alcohol, show hypercortisolism and a blunted stress response (for reviews, see: Blaine & Sinha, 2017; Sinha, 2001) [26,27].
Boredom, on the other hand, can be viewed as a form of psychological stress, a state of restless dissatisfaction that often motivates individuals to seek stimulation or novelty [28].
For those with heightened impulsivity, the experience of boredom might trigger a more pronounced cortisol release due to their inherent heightened reactivity to stressors [29,30].
The underlying mechanism behind the impulsivity-boredom relationship, therefore, could be that the HPA axis.In other words, those high in impulsivity may find the "stress" of boredom to be more adverse, leading to greater level of cortisol in response to boredom.Thus, the HPA axis may serve as the physiological "bridge" connecting impulsivity and the intensified experience of boredom.

Study 2
Our hypothesis for Study 2 posits that the HPA axis activity, specifically cortisol responses, mediate the relationship between impulsivity and experienced boredom.Therefore, in Study 2, we aimed to elucidate this potential mediation by incorporating salivary cortisol change into our analyses.Here, we used a 20-minute vowel cancellation task (VCT) [31] instead of the 5-minute peg turning task.Both the peg turning task [32] and repetitive cognitive tasks (such as the VCT) [33] have been shown to effectively induce boredom.In Study 1, we opted for a shorter task as we were concerned that online participantswithout the presence of an experimenterwould lose interest, resulting in either poor task adherence or high levels of attrition.On the other hand, in Study 2, we required a longer task to ensure that enough time had elapsed to induce a cortisol response.As such, the shorter peg turning task not deemed to be suitable.

Participants
Participants were 20 adults (9 women and 12 men) ranging in age from 18 to 24 (M = 20.80,SD = 1.51) who were students studying at the University of Portsmouth.Participants were required to be aged 18 or older; have English as a first language or an equivalent level of fluency; to not currently be undergoing treatment for anxiety, depression or any other stressrelated disorder.Due to the effects on salivary cortisol, women could not be pregnant, breastfeeding or currently taking oestrogen-and progesterone-based contraception.For all participants, participants could not take any of the following medications within the past 24 h: barbiturates, phenytoin, carbamazepine, meprobamate, glutethimide, alpha-methyldopa, corticosteroids, non-steroidal anti-inflammatory agents (e.g.aspirin, ibuprofen]), codeine, propranolol, beta-adrenergic agonists, cyproheptadine and psychotropic medications (including sedative hypnotics).This was verified via a pre-screening questionnaire.The study was approved by the University of Portsmouth Science and Health Faculty Ethics Committee.

Cortisol
Participants provided two (pre-and post-boredom initiation) saliva samples using a passive drool method according to the sampling schedule described below.During the passive drool procedure, participants sit with their head tilted slightly forward, allowing saliva to pool in the mouth.They then spit the saliva into a collection tube [34].All samples were collected between 12:00 and 15:00 to limit the effect of diurnal cortisol fluctuation on our data [35].
Following sample collections, the saliva samples were placed on ice until the end of each study session.Once the study session was complete, the saliva samples were rendered acellular and kept in a freezer (-20 °C) until they were assayed in duplicate using highsensitivity cortisol ELISA kits (Salimetrics, Carlsbad, CA) according to the manufacturer instructions.The average intraassay coefficient of variability was 3.97%.
Briefly, during the VCT, participants were given three pages of printed text (taken from Arifin, 1984) [36] and instructed to use any method they wish to count the total number of vowels in the text.They were given up to 20-min to carry out the task.

Procedure
Study 2 was carried out in a laboratory (there was no one else present except the experimenter and the participant).Once informed consent was confirmed, participants completed the BIS-11 (Cronbach's α = 0.92) and provided their pre-boredom saliva sample.Subsequently, participants completed, the VCT, before immediately completing the MSBS (Cronbach's α = 0.98).Finally, participants provided their post-boredom saliva sample after approximately ten minutes had elapsed following the completion of the rest of the study.This allowed adequate time for changes in cortisol levels to be detected in saliva [37].

Analysis
Descriptive statistics (means, standard deviations, and the proportion of missing data) were calculated and bivariate associations were explored for key study variables.There were no missing data.Cortisol variables were log transformed prior to analysis and a change score (postpre) was calculated.We used t-tests to check whether our manipulation worked as expected.Our primary hypothesis was tested through mediation analysis using PROCESS model 4 (version 4.1, SPSS 28).Bias-corrected bootstrapped (n = 10,000) 95% confidence intervals (CIs) were calculated for PROCESS models.We also calculated standardized and unstandardized model coefficients.Results were considered significant when p < .05 or when the 95% CIs did not overlap zero.

Relations between impulsivity, boredom, and salivary cortisol [INSERT FIGURE 1 ABOUT HERE]
The results of the mediation analysis are summarised in Figure 1 and Table S2 (Supplementary Material).Impulsivity (BIS-11) positively predicted an increased cortisol

General Discussion
The aim of this work was to: (a) replicate previous studies which have demonstrated that high-impulsivity individuals are more likely to experience greater levels of self-report boredom following a boredom inducing task; and (b) test the hypothesis that the relation between trait-impulsivity and boredom would be mediated by increased HPA (i.e., cortisol) activity.In two studies, we have demonstrated that people high in trait-impulsivity show higher self-reported boredom after completing a boredom-inducing task, and in Study 2, that this link is mediated, at least in part, by the HPA axis activity.

Implications of the relationship between impulsivity and boredom
Previous work has shown that those with reduced impulse control tend to have greater boredom proneness (Isacescu et al., 2017;Struk et al., 2016).For instance, electroencephalographic (EEG) evidence suggests higher levels of task-induced boredom are associated with reduced P3 amplitudes [39].Similarly, reduced P3 amplitudes are observed in several externalising disorders, such as ADHD, substance use disorders, and obesity [40,41], and P3 amplitude reduction is directly related to diminished impulse control [42].Therefore, this suggests that a shared neurophysiological link between impulsivity, boredom, and externalising disorders.
Given the established links between boredom and various negative outcomes (e.g., substance abuse), understanding this impulsivity-boredom nexus requires further study as these links may be paramount in understanding externalising behaviour.Therefore, questions about how interventions might be designed to help high impulsivity individuals manage or mitigate the effects of boredom in their daily lives rise from this work.

The role of the HPA axis
HPA axis hyperactivity (i.e., hypercortisolism) has been ostensibly linked to several psychopathologies such as depression and suicidality [43,44], but critically to externalising and impulse control disorders, such as substance use (for reviews, see: Blaine & Sinha, 2017; Sinha, 2001), and ADHD [46].Elevated cortisol can mediate the release of glutamatethe brain's primary excitatory neurotransmitterleading to potential neurotoxicity, changes in synaptic plasticity, and altered glutamate receptor sensitivity [47].Additionally, chronic cortisol elevation reduces P3 amplitudes [48], suggesting a possible mechanism for observed alterations in cognitive function or attention following chronic stress.The exact means by which cortisol affects P3 remains unclear, but the links between impulsivity, boredom, and P3 amplitude, along with our findings here on the role of the HPA in mediating impulsivity and boredom, are intriguing and could suggest a mechanism for future exploration.Furthermore, reduced P3 aptitudes, heightened impulsivity, and stress are also linked to emotional dysregulation [49,50], which is also a key component in several disorders, such as post-traumatic stress disorder [51] and alcohol use disorder [52].For instance, impulsivity strengthens the link between emotional dysregulation and externalising behaviours such as alcohol misuse [19].
Therefore, there is a clearly an overlapping set of risk-factors and consequences of psychopathology which may potentially all related to the same P3 event related potential.

Limitations
We acknowledge several study limitations.First, memory inaccuracies, response biases and demand characteristics can leave self-report measures susceptible to increased levels of measurement error.For example, self-report impulsivity measures may reflect self-identified behaviours rather than directly tapping into the construct [53].However, we selected and used common and psychometrically valid tools for this study to mitigate.Second, though theory driven, neither the study nor analysis plan was pre-registered on a publicly available platform, thus results should be considered exploratory.In this vein, we did not apply a correction for multiple comparisons, choosing instead to retain sensitivity in detecting trends for future research.Third, our final sample sizes were relatively small as they were limited by resource constraints.It could therefore be argued that our studies were underpowered, which can lead to inflated effect sizes [54], and that our samples are less generalisable.However, we present 95% CIs as an expression of our confidence in our mediation results, and by showing that even the lower bounds of the interval for the standardised indirect effect is relatively high (0.13), suggesting practical importance, we therefore argue that the true population effect is likely to also be relatively robust.The low sample size could also affect the distribution approximation when trying to estimate the true population distribution.Despite this, a replication of Study 2 with a large representative sample is warranted.Finally, in this study, we measured state boredom, not trait boredom (ie boredom proneness).Impulsivity and boredom proneness are highly correlated, but distinct constructs [55].In future studies, it would be worth including measures of boredom proneness to control for the unique effects of this construct.

Conclusion
We demonstrate that HPA axis activity mediates the link between trait-impulsivity and boredom.Recognising that the HPA axis may underline this relationship extends previous work and offers a novel insight into the potential mechanisms.Moreover, the well-established nature of reduced P3 amplitudes, heightened impulsivity, stress, and emotional dysregulation highlights a need for further exploration, especially given their relevance across numerous psychiatric disorders.Future studies could focus on how interventions can be designed for high impulsivity individuals to alleviate the negative impacts of boredom and potentially break the identified feedback loop.Table 1 Descriptive statistics (M and SD) across high (n = 41) versus low (n = 39) impulsivity groups.

Table 3
Descriptive statistics (M an SD) intercorrelations (Spearman's rank values) for key study variables.We have no known conflict of interest to disclose.

1
Figures Levene's test was significant (p < .05)suggesting a violation of the assumption of equal variances. a

Table 2
Descriptive statistics (M an SD) intercorrelations (Spearman's rank values) for key study variables.