Snacking accounts for 15–20% daily intake of food and drinks besides main meals and mostly occurs during work (Chaplin & Smith, 2011; Liu et al., 2015). Given that snacking is not typically triggered by hunger, it contributes to excessive energy intake and may therefore lead to weight gain and obesity, and consequently, detrimental health consequences like high blood pressure, diabetes and heart diseases (Chaplin & Smith, 2011; World Health Organization, 2021). Considering that currently, more than a third of the world’s population is overweight or obese (World Health Organization, 2021), shedding light onto factors that may result in weight gain is crucial both from a theoretical and practical perspective.

Nevertheless, research on snacking has only recently emerged in organizational literature with recent reviews stressing the paucity of existing work (e.g., Clohessy et al., 2019). Although preliminary research has provided important first insights into predictors of work-related snacking, it has focused on a limited range of factors that represent conditions of high work stress, such as high workload interpersonal conflict, and self-control demands (Liu et al., 2017; Sonnentag et al., 2017; Vasiljeva et al., 2023). Yet, preliminary work on job strain and arousal suggests that not only over-stimulating job demands (i.e., high arousal) but especially under-stimulating job demands (i.e., low arousal) may negatively affect employee’ eating behaviors during work (Payne et al., 2005, 2013). Surprisingly, however, the role of specific under-stimulating work conditions, such as job monotony, as well as their mechanisms and boundary conditions have not been considered in the work-related snacking literature yet.

The goal of the present study is therefore to broaden theory on work-related snacking by examining the role of low arousal work demands. Drawing from optimal arousal theory, which states that there is an optimal level of arousal for desirable work outcomes (Hebb, 1955), we develop and test a conceptual model explaining how job monotony leads to unhealthy snacking during work through the experience of boredom. As snacking is a behavior that varies from day-to-day, we adopt a within-person perspective to investigate these relationships in our model. Although traditionally, job monotony has been viewed as a rather constant job characteristic with some jobs considered as monotonous while others not (Loukidou et al., 2009), we argue that employees may experience various levels of monotony in their daily work. Such a within-person approach allows us to identify more proximal predictors of work-related snacking, thus, considering that unhealthy snacking may be prevalent in any type of job. Furthermore, since it is well established that eating behaviors differ between people (Greeno & Wing, 1994), we propose that trait mindfulness may be an important boundary condition between daily work-related boredom and unhealthy snacking as mindfulness increases one’s emotion regulation ability and decreases impulsive reactions, which may arise from boredom as a state of dissatisfaction (Brown et al., 2007). For an overview of the model, see Fig. 1

Our study makes several important contributions to organizational literature. First, we extend prior research on work-related factors that may contribute to unhealthy snacking choices of employees by investigating two previously overlooked factors: job monotony and work-related boredom. While prior studies have shown that high arousal work stressors and associated negative emotions may increase unhealthy snacking (Liu et al., 2017; Sonnentag et al., 2017), we direct attention to the snacking-inducing potential of low arousal psychological experiences, specifically boredom. Research suggests that 79 to 87% of employees report being bored at least sometimes during their work (Fisher, 1993). Surprisingly however, boredom has received little attention from contemporary organizational researchers in general (van Hooft & van Hooff, 2018) and no attention in the work-related snacking literature. Yet, exploratory research outside the organization sciences points to the possibility that boredom might be a better predictor of eating behaviors than previously used negative affect measures (Koball et al., 2012). Additionally, prior studies show that boredom is qualitatively distinct from and generally more frequently reported than other negative affective states, such as sadness, anger, frustration, anxiety, and hopelessness (Pekrun et al., 2010; van Tilburg & Igou, 2012). It is therefore imperative to further investigate the role of monotony and work-related boredom for employee unhealthy snacking behaviors during work.

We directly address this need by building on first empirical evidence outside the organizational sciences, which show that completing a repetitive task (i.e., writing the letters “cd” over and over again; Abramson & Stinson, 1977; or watching a short clip on repeat for 60 min; Havermans et al., 2015), is perceived as boring and leads to higher consumption of crackers and chocolate. While these imperative findings demonstrate a causal link between induced boredom and consumption of certain foods, they lack ecological validity and generalizability, as these studies have been conducted in the laboratory with students and the performed tasks are not directly applicable to real-world settings, especially the work context. However, opposed to exclusively boring and interesting conditions created in the lab, real-world jobs may include varied levels of monotony that fluctuate within and across workdays, thus, stressing the need for field studies that account for such daily fluctuations in the work context.

Surprisingly, prior studies in the organizational literature have almost exclusively focused on between-person effects of job monotony, examining differences in average levels of such phenomena and, thus, neglecting the day-to-day fluctuations within individuals (Loukidou et al., 2009; for an exception see Scharp et al., 2021). In particular, prior literature has viewed job monotony and work-related boredom as predominantly stable work characteristics, commonly co-occurring in repetitive, simple jobs, such as manufacturing (van Hooff & van Hooft, 2017). Yet, the potential variability of job monotony and boredom across days within a given occupation or individual has not been taken into account yet. For instance, a researchers’ job is usually quite varied (i.e., non-monotonous by nature), including multiple tasks during the workday (e.g., attending meetings, lecturing, analyzing data), however, some days may well be monotonous and boring (e.g., predominantly grading exams). Therefore, while the average level of monotony may be low in such a job, researchers might still be negatively influenced by monotony and boredom on days when their tasks lack variety, which would be overlooked by designs that solely focus on between-person differences. Our within-person approach allows us to account for such daily fluctuations, considering that the effects of monotony and boredom are not fixed for every individual across occupations.

Lastly, we contribute to a more comprehensive understanding of protective factors surrounding snacking in the work context and thereby respond to a recent call highlighting the need for more in-depth understanding of factors that affect work-related eating behaviors (Hill et al., 2021). By investigating trait mindfulness as a boundary condition, we focus on factors that may limit work-related snacking. This does not only expand our theoretical understanding of workplace snacking but also informs future workplace interventions that aim to prevent or reduce such behavior. Specifically, inter-individual differences in trait mindfulness may explain why some individuals eat more when feeling bored during work while others do not, further allowing to tailor interventions to individual needs. This is especially relevant as mindfulness can be increased through training (Brown et al., 2007; Good et al., 2016).

Fig. 1
figure 1

Conceptual model and path estimates. Note For the indirect effects, 95% CI is indicated. * p <.05. ** p <.01. *** p <.001 (two-tailed)

Full size image

Monotonous Work is Boring

Job monotony describes one’s tasks or job as repetitive or lacking of variety (Game, 2007), thereby potentially having a substantial negative impact on employees. Historically, research originating from the early 1900s has linked job monotony to various physical and mental-health related consequences that result from performing repetitive manual labor (e.g., increased risk of heart attack, asthma and hand tremor in telegraphists performing high precision and concentration tasks; Ferguson, 1973; increased anxiety, irritability and depression in blue collar workers; Melamed et al., 1995) as well as its overall negative impact on work outcomes (e.g., lower job satisfaction and increased sick leave; Melamed et al., 1995). Furthermore, job monotony in such work environments has been frequently linked with the experience of boredom (Loukidou et al., 2009; O’Hanlon, 1981).

Traditionally, monotony and boredom have been assumed to co-occur, with job monotony considered as the main cause of boredom (O’Hanlon, 1981; Smith, 1981). Indeed, the definition of boredom itself: “a state of relatively low arousal and dissatisfaction which is attributed to an inadequately stimulating environment” (Mikulas & Vodanovich, 1993, p. 3), suggests that boredom is a product of monotonous work settings. In line with this definition, early work on work-related boredom has applied optimal arousal theory to explain the relation between job monotony and boredom in terms of low external stimulation that leads to low internal arousal of individuals (Geiwitz, 1966; Hebb, 1955). Therefore, when faced with repetitive work, low arousal may be expressed as paying little attention to a task, daydreaming and sleepiness – all typical indicators of boredom (Damrad-Frye & Laird, 1989). As a result, attention and consequent performance is impaired (Loukidou et al., 2009). Furthermore, monotonous work may also influence perception of time in a way that time seems to drag or go by slowly (Mikulas & Vodanovich, 1993; Pekrun et al., 2010). This could also intensify the experienced boredom as employees may feel that they have to continue doing their monotonous work for a longer time. In line with this, we propose that daily job monotony makes employees bored when working.

Hypothesis 1

Daily job monotony is positively related to daily work-related boredom.

Unhealthy Snacking When Bored

Optimal arousal theory posits that effective behavior requires an optimal level of arousal that is not too low or too high (Hebb, 1955). Thus, individuals strive to maintain such an optimal level of arousal, meaning that if the experienced level of arousal is below optimal, individuals will seek stimulation to increase their arousal (Hebb, 1955). Accordingly, theoretical work on boredom has applied optimal arousal theory to define boredom as an unpleasant, low arousal state (Mikulas & Vodanovich, 1993) that consequently encourages individuals to seek less boring activities in order to increase their arousal level. In line with this, research on work-related boredom has shown that in order to cope with boredom, individuals may look for additional stimulation and variety by engaging in other non-work related activities, such as reading, talking, listening to music, playing games, and smoking (Fisher, 1993; Kishida, 1977 as cited in Game, 2007).

Linking optimal arousal theory (Hebb, 1955) with laboratory studies in the non-work-related eating literature (Abramson & Stinson, 1977; Havermans et al., 2015), we argue that another form of compensatory behavior employees may revert to when being bored is unhealthy snacking. Since “boredom is commonly associated with a motivation to cognitively or physically change or escape the situation” (van Hooft & van Hooff, 2018, p. 932), snacking may serve as a distraction from the boring task. Indeed, an experimental study showed that completing a boring (versus interesting) puzzle task was linked to a higher desire to snack (Moynihan et al., 2015). Furthermore, as boredom encourages people to seek sensation (van Tilburg & Igou, 2012), especially unhealthy foods may be preferred in such an endeavor, as they are often viewed as more exciting and stimulating than healthier options (Moynihan et al., 2015). Accordingly, we propose that being bored during work may facilitate unhealthy snacking.

Hypothesis 2

Daily work-related boredom is positively related to daily unhealthy snacking during work.

Given that job monotony may promote boredom during work and boredom may in turn increase intake of unhealthy snacks, we propose that daily work-related boredom mediates the relationship between daily job monotony and unhealthy snacking. Indeed, first experimental evidence from laboratory studies with students in non-work contexts has shown that completing a short, repetitive task was perceived as boring and led to increased intake of crackers and chocolate (Abramson & Stinson, 1977; Havermans et al., 2015).

Hypothesis 3

The positive relationship between daily job monotony and daily unhealthy snacking during work is mediated by daily work-related boredom.

Trait Mindfulness as a Buffer for Boredom-Related Unhealthy Snacking

Considering that individual differences play an important role for eating behaviors (Greeno & Wing, 1994), we propose that trait mindfulness may have a buffering effect in the relationship between work-related boredom and unhealthy snacking. Mindfulness can be described as the “enhanced attention to and awareness of current experience or present reality” (p. 822), with a particular focus on an open and receptive attitude towards these present experiences (Brown & Ryan, 2003). Trait mindfulness refers to the natural, rather stable tendency to be mindful therefore differing among individuals irrespective of any prior training (Brown & Ryan, 2003).

Mindfulness allows individuals to process information experientially, attending the situation at hand as it is - with neutral evaluation and without habitual referencing (Good et al., 2016). This means that mindful individuals are able to observe current experiences more objectively by providing psychological distance between the stimulus and its appraisal, allowing for more effective behavioral regulation (Good et al., 2016). Linking the mindfulness literature with optimal arousal theory, we therefore suggest that having a high trait mindfulness can increase the tolerance of and willingness to experience aversive low-arousal states, such as boredom, rather than actively trying to avoid or escape such experiences (Brown et al., 2007; Galla et al., 2020). Consequently, the experience of boredom may be encountered with curiosity and openness, thereby reducing compensatory impulsive reactions that seek to increase stimulation and arousal but instead facilitating more flexible and adaptive behavioral reactions to such experiences (Brown et al., 2007; Galla et al., 2020). In doing so, individuals with high trait mindfulness levels can acknowledge that they are bored without automatically acting on the impulse to compensate for such an unpleasant feeling and create a more optimal state of arousal by engaging in unhealthy snacking.

In line with this, prior research has shown that while boredom is associated with higher impulsivity (Moynihan et al., 2017), mindfulness can improve emotion regulation and decrease impulsive reactions to negative emotions, such as boredom (e.g., Watford et al., 2019). In particular, cross-sectional studies on individuals with overweight and obesity have shown that having high trait mindfulness may enable individuals to engage in less emotional eating and decrease food-related impulsivity (Levin et al., 2014; Watford et al., 2019). Accordingly, we propose that trait mindfulness may act as a buffer in the relationship between work-related boredom and unhealthy snacking during work.

Hypothesis 4

The positive relationship between daily work-related boredom and daily unhealthy snacking during work is moderated by trait mindfulness such that this relationship is weaker for employees high as opposed to low in trait mindfulness.

Method

Sample and Procedure

This study was approved by the ethical review board of the authors’ home institution (ERCPN-OZL_207_05_04_2019_S2). Data collection took place between April and July 2021. Altogether, 142 individuals agreed to participate in our study but 15 did not finish the baseline measurement, 5 did not work at the time of the study or indicated being a student as their profession and 17 did not participate in daily surveys. Therefore, the final sample included 105 participants who had completed the baseline measurement and at least one daily survey (73.9% of those signing up). Our sample included 67.6% females and the average age was 36.54 (SD = 12.55). Most of our participants (79.0%) had completed a university degree (3.8% Doctoral degree, 45.7% Master’s degree and 29.5% Bachelor’s degree) while 8.6% had finished general higher education, 2.9% advanced technical college, 8.6% high school and 1.0% were without school certificate. Our participants represented various organizational sectors, such as education and research (18.1%), healthcare (16.2%) consultancy and human resources (11.4%) and finance and sales (6.7%). These organizations were located in multiple countries, including Germany (53.3%), Canada (30.5%), United States of America (6.7%) and others (9.5%). Since the majority of our participants were recruited in Germany, questionnaires were provided in both German and English. The average organizational tenure was 6.57 years (SD = 8.42; ranging from 1 month to 40 years). Additionally, as our data collection took place during the global Covid-19 pandemic, we assessed whether people worked at their usual workplace or from home. In our sample, 41.0% participants worked from home every day, 20.0% worked from home on some days of the week and 39.0% worked at their usual workplace.Footnote 1 The average participant had a normal weight at the time of the study, according to the body mass index (BMI; M = 23.63; SD = 4.03; World Health Organization, 2010).

Participants for this study were recruited through convenience sampling by a team of research assistants (e.g., via personal networks, advertising on social media and flyers distributed in public spaces). Respondents had to be at least 18 years old and work for a minimum of 20 h a week on all or most days of a workweek (at least 5 h per day during regular working hours, with no night shifts). After registration, providing their consent and filling in the baseline questionnaire, employees responded to a daily end of work survey for two consecutive workweeks in the online software Qualtrics (available between 4:00 p.m and 7:00 p.m.). Participants received an initial e-mail invitation and a reminder 1 h before the expiration of the survey (at 6:00 p.m.). At the start of each survey, participants were asked whether the respective day was an official workday – if this question was negated, the survey stopped and no further measurements were recorded. In total, participants filled in 592 surveys, averaging to 5.6 surveys per person (including 111 surveys filled in on days that people did not work; containing no data). Following Arend and Schäfer (2019), given on average 5.6 repeated measures per person on Level 1, medium ICC values, and a sample size of N = 105 on Level 2, our data allowed us to detect small direct effects (minimum detectable effect size of 0.15 to 0.16).

Measures

Measures used in this study were adjusted to the daily context from previously validated scales. For the German version of our surveys, we used previously validated translations of existing scales or applied a translation and back-translation procedure in consultation with two bilingual speakers who were proficient in both German and English (Brislin, 1970).

The baseline measurement included questions about employees’ demographics and a trait mindfulness scale. The daily survey assessed daily job monotony, work-related boredom and unhealthy snacking. All items (except unhealthy snacking) were rated on a scale from 1 (strongly disagree) to 5 (strongly agree).

Trait Mindfulness

Trait mindfulness was measured with the Mindfulness@Work Scale (M@Work; Hülsheger & Alberts, 2021). The M@Work scale encompasses different sub-facets of mindfulness, thereby allowing to capture mindfulness more comprehensively than scales targeting only attentional aspects of mindfulness (e.g., Mindful Attention Awareness Scale; Brown & Ryan, 2003). These sub-facets are: acting with awareness (being attentive and fully engaged in present activity), non-reactivity to inner experiences (accepting thoughts and feelings without reacting or getting caught up), non-judging (experiencing thoughts and feelings without judging or criticizing oneself) and describing (being able to find the words to describe the present experience; Hülsheger & Alberts, 2021). The M@Work scale includes 22 items (e.g., “When negative things happen at work, I have immediate intense reactions.”). Cronbach’s alpha for this scale was .82.

Daily Job Monotony

Daily job monotony was assessed with three items from the subjective monotony measurement by Melamed et al. (1995) suited to measure daily job monotony: (“Today at work/while working at home, my work was…”). These items were “routine”, “monotonous” and “not varied enough”.Footnote 2 Cronbach’s alpha (averaged across days) was .77 (SD = 0.07, ranging from 0.69 to 0.88).

Daily Work-Related Boredom

Daily work-related boredom was assessed with three items from van Hooff and van Hooft (2017) adapted from Lee (1986) and suited to daily measurement (e.g., “Today at work/while working at home, I found my job boring.”). Cronbach’s alpha (averaged across days) was .84 (SD = 0.06, ranging from 0.74 to 0.92).

Daily Unhealthy Snacking

We built on established measures in the wider eating literature (Brown et al., 2008; Liu et al., 2017; Mouchacca et al., 2013; Powell et al., 2017) to develop a measure suited for daily assessment of unhealthy snacking. We combined items from these measures into an adapted scale that includes all unhealthy snacking items in a shortened version. For example, we clustered categories like pizza, deep fried food and fast food in order to reduce the burden on participants (Fisher & To, 2012). The unhealthy snacking items measured in our study were “Salty or savory snacks (e.g., potato chips, piece of cheese)”, “Sweet snacks (e.g., chocolate, candy, pastry)”, “Fast food (e.g., takeaway, deep fried food, pizza)” and “Sodas or sugary drinks (e.g., cola, syrup, iced tea)”. Participants were asked to indicate how many portions of each snacking item they had consumed during work/while working at home that day (excluding what they ate for their regular meals). The instructions explained that one portion refers to one piece of food item (e.g., one pastry) or the recommended portion size as indicated on the product packaging (e.g., 30 g of chocolate, 30 g of chips). Similarly to prior studies, the number of consumed portions per each item were summed from all four items in order to calculate an overall number of consumed unhealthy snacks. Due to the nature of this variable (i.e., a sum), we did not calculate a reliability estimate for this scale.

Control Variables

Our supplementary analyses included two control variables: workload and negative affect, which were measured in our daily survey. Daily Workload was assessed with three items based on the Job Content Questionnaire (JCQ; Karasek, 1985) previously used in Bakker et al. (2010). An example item is “Today at work/while working at home, I worked under time pressure.” Cronbach’s alpha (averaged across days) was .80 (SD = 0.05, ranging from 0.68 to 0.86).

Daily Negative Affect was measured with three items suited to measurement of work-related negative affect from Bono et al. (2007). Participants indicated how anxious, angry and irritated they felt today at work/while working at home. Cronbach’s alpha (averaged across days) was .84 (SD = 0.03, ranging from 0.79 to 0.88).

Statistical Analysis

Following recommendations in the literature to not exclude participants with missingness (Hox, 2002; Singer & Willett, 2003), all participants who responded to at least one daily survey were included in the analyses (see also Griep et al., 2021; Hülsheger et al., 2021). In order to address missing data, we used maximum likelihood estimation (Newman, 2014; Peugh & Enders, 2004). For analyzing our multilevel data (daily measurements were nested within people), we used multilevel Poisson regression modelling in Mplus 8.8 (Muthén & Muthén, 2017). The use of a Poisson distribution was necessary as our outcome was a count variable (sum of the four unhealthy snacking items).

The intra-class correlation coefficient (ICC; the proportion of variance that can be attributed to differences between individuals) ranged from 41 to 65% in our daily variables (see Table 1). In our analyses, we person-mean centered Level 1 predictors (daily job monotony and work-related boredom) and grand-mean centered the Level 2 variable (trait mindfulness). We modelled our predictors using fixed slopes for direct and indirect effects in Hypotheses 1–3 and a random slope (daily work-related boredom to unhealthy snacking) to test the cross-level interaction in Hypothesis 4. For estimation of the indirect effect, we used the model constraint function in Mplus. Following the recommendations of Preacher et al. (2010), we tested the indirect effects at the within-person level with a parametric bootstrap procedure to compute 95% confidence intervals that account for the asymmetric nature of the sampling distribution.

It has to be noted that all daily variables in our model were measured concurrently at the end of the workday, referring to the experiences during work on that day. While a time-separated design comes with a number of advantages (e.g., to reduce common method bias; Gabriel et al., 2019), from a theoretical and practical perspective, such separation may provide an incomplete and inaccurate representation of rather short-lasting effects (Beal, 2015), as is the case in our study. Importantly, prior literature has confirmed that boredom is activity-related, meaning that it instantly occurs while performing the boredom-inducing activity, and disappears soon after (e.g., Pekrun et al., 2010; van Hooft & van Hooff, 2018). Thus, assessing our daily variables at different instances (e.g., in the morning and in the afternoon) was not sensible, as boredom occurs during the completion of a monotonous task and consequences of boredom, such as unhealthy snacking, follow within short time periods. Therefore, a concurrent measurement of predictors and outcomes of boredom is in line with established conceptual and empirically supported ideas about boredom.

Table 1 Means, standard deviations, reliabilities, and correlations among variables
Full size table

Results

The means, standard deviations, reliabilities, ICCs and inter-correlations between study variables are presented in Table 1. First, multilevel confirmatory factor analysis (CFA) was conducted to examine the construct validity of job monotony and work-related boredom. The one factor model did not provide an adequate fit: χ2 = 97.31, p <.001, comparative fit index (CFI) = 0.89, Tucker–Lewis index (TLI) = 0.82, root-mean-square error of approximation (RMSEA) = 0.10, standardized root-mean-square residual (SRMR) = 0.06 at the within-person level and 0.07 at the between-person level. The two factor model showed a good and significantly better fit: χ2 = 35.49, p <.01, CFI = 0.97, TLI = 0.95, RMSEA = 0.05, SRMR = 0.03 at the within-person level and 0.06 at the between-person level, chi-square difference = 64.82, df = 2, p <.001. Therefore, the two constructs, job monotony and work-related boredom, were confirmed as distinct factors.

Results from multilevel Poisson regression are summarized in Table 2 (direct and indirect effects; Hypotheses 1–3) and Table 3 (cross-level interaction; Hypothesis 4). As Table 2 shows, daily job monotony was a significant predictor of daily work-related boredom (γ = 0.54, p <.001), confirming Hypothesis 1. In line with Hypothesis 2, daily work-related boredom was significantly positively related to daily unhealthy snacking (γ = 0.21, p =.02).Footnote 3 This means that a one-unit increase in daily work-related boredom is associated with an increase in unhealthy snacking by a factor of 1.23 (e.21 = 1.23) or 23% more unhealthy snacking. Furthermore, the mediation effect of daily work-related boredom in the relationship between daily job monotony and daily unhealthy snacking was confirmed (γ = 0.11, 95% CI [0.02, 0.20]), supporting Hypothesis 3. This finding suggests that a one-unit increase in daily job monotony is related to an increase in unhealthy snacking through work-related boredom by a factor of 1.12 (e.11 = 1.12) or 12% more unhealthy snacking. Lastly, as Table 3 shows, the interaction effect of trait mindfulness on the relationship between daily work-related boredom and daily unhealthy snacking was not significant (γ = − 0.05, p =.81), therefore not confirming Hypothesis 4.

Table 2 Multilevel poisson model for direct and indirect effects
Full size table
Table 3 Multilevel poisson model for cross-level interaction
Full size table

Supplementary Analyses

To provide further empirical evidence for the robustness of our findings and conduct a more conservative test of our model, we investigated the effect of job monotony on work-related boredom and unhealthy snacking above and beyond an alternative predictor and mediator.

What is the Role of Over-Stimulating Work Demands?

To provide a more complete picture of employees’ daily work life and put the role of job monotony into perspective, we considered high workload as an over-stimulating work demand. Thus, we suggest that low arousal and high arousal characteristics may co-exist, as one’s work tasks may be monotonous but still need to be completed in a short amount of time, indicating a high workload. The results showed that the relationships in our model did not change when including daily workload as an additional predictor. Daily job monotony remained a strong predictor of daily work-related boredom (γ = 0.48, p <.001), while daily workload had a substantially weaker, negative relationship with daily work-related boredom (γ = − 0.27, p <.001), see Fig. 2. Both daily job monotony (γ = − 0.15, p =.10) and daily workload (γ = 0.06, p =.42) were not directly related to daily unhealthy snacking. However, as in our hypothesis test, daily job monotony was indirectly related to increased daily unhealthy snacking via daily work-related boredom (γ = 0.11, 95% CI [0.02, 0.20]) while daily workload was indirectly related to decreased daily unhealthy snacking via daily work-related boredom (γ = − 0.06, 95% CI [-0.11, − 0.01]).

Fig. 2
figure 2

Path Estimates for Supplementary Analysis with Workload as an Alternative Predictor. Note Both pathways were tested in the same model. For the indirect effects, 95% CI is indicated. * p <.05. ** p <.01. *** p <.001 (two-tailed)

Full size image

Is Boredom a Unique Explanatory Mechanism?

While recent diary studies have found general negative affect to explain the relationship between (high-arousal) job demands and unhealthy eating (Liu et al., 2017; Sonnentag et al., 2017), research outside the organizational sciences posits that boredom as a distinct emotion may be a better predictor of eating behaviors than other negative affective states (Koball et al., 2012). Thus, to provide further support for the unique explanatory role of work-related boredom in the relationship between job monotony and unhealthy snacking, we tested negative affect as an alternative mediator in our model. The results showed that although daily negative affect was significantly positively related to daily unhealthy snacking (γ = 0.26, p <.01), daily job monotony was not related to daily negative affect (γ = 0.02, p =.64). Thus, only daily work-related boredom (γ = 0.10, 95% CI [0.01, 0.20]) but not daily negative affect (γ = 0.01, 95% CI [-0.02, 0.03]) acted as a mediator between daily job monotony and daily unhealthy snacking, see Fig. 3.

Fig. 3
figure 3

Path Estimates for Supplementary Analysis with Negative Affect as an Alternative Mediator. Note Both pathways were tested in the same model. For the indirect effects, 95% CI is indicated. * p <.05. *** p <.001 (two-tailed)

Full size image

Discussion

The current study provides valuable insights into why employees consume unhealthy snacks during work. Drawing from optimal arousal theory (Hebb, 1955), we investigated the role of low arousal work demands for work-related snacking behaviors. Specifically, we proposed a conceptual model in which daily job monotony predicts unhealthy snacking through work-related boredom. Addressing the need to learn more about the conditions affecting employee eating behaviors (Hill et al., 2021), we investigated trait mindfulness as an inter-individual buffer between daily work-related boredom and unhealthy snacking during work.

Our findings support our proposition that low arousal work demands, specifically job monotony, promote unhealthy snacking behaviors during work due to the experienced work-related boredom. In other words, on days when employees have to perform uninteresting, repetitive tasks, they tend to lack stimulation, resulting in boredom (Fisher, 1993; Loukidou et al., 2009). This, in turn, facilitates the search for variety through other activities in order to alleviate the unpleasant feeling of boredom, encouraging compensatory behaviors like unhealthy snacking. Given that unhealthy snacking may not only be detrimental for employee health (e.g., weight gain and obesity; World Health Organization, 2021) but also work performance (i.e., decreased organizational citizenship behaviours and increased employee withdrawal; Cho & Kim, 2022), our findings offer important theoretical implications.

First, although more diary studies on predictors of work-related unhealthy behaviors have been published in recent years, most of this work has focused on delayed effects, indicating that job demands affect eating behaviors after work (i.e., Headrick & Park, 2022; Liu et al., 2017; for an exception see Vasiljeva et al., 2023). However, since unhealthy snacking is more common at work than at home (Liu et al., 2015), it is of particular relevance to identify predictors of unhealthy snacking during work. Therefore, our findings extend the current work-related snacking literature by identifying monotony as a low-arousal work characteristic that immediately elicits unhealthy snacking in the work context, which has been neglected so far. In particular, our study shows that, similar to over-stimulating work demands (e.g., self-control demands during work; Sonnentag et al., 2017), under-stimulating work conditions may also facilitate unhealthy snacking. Moreover, in our supplementary analyses, we found that these effects persisted when controlling for an additional high-arousal work characteristic: workload, and alternative psychological mechanism: negative affect. Furthermore, only work-related boredom but not negative affect explained the relationship between job monotony and unhealthy snacking. Taken together, our results show that the effects of job monotony and work-related boredom are unique and influence unhealthy snacking above and beyond alternative predictors and mediators.

Second, while job monotony has been consistently linked to various important health and work outcomes (e.g., Fisher, 1993; Melamed et al., 1995), extant research has almost exclusively focused on chronic between-person differences in job monotony, assuming that it is stable within individuals due to the design of their job (for an exception see Scharp et al., 2021). We argued and found that monotony not only varies systematically between jobs and their incumbents but also displays substantial variability from day-to-day and therefore has a dynamic component. In fact, 59% of variance in job monotony was attributable to within-person differences and this was the highest within-person variance percentage among all study variables (similarly to Scharp et al., 2021, who reported 57.1% within-person variability in job monotony). This finding suggests that, on average, job monotony varies more between days than between individuals, indicating that most employees may experience different levels of job monotony and work-related boredom across the workweek. Research on job monotony but also on job characteristics more generally may therefore benefit from adopting a within- rather than a pure between-person perspective. Doing so will allow capturing the micro-level processes occurring on a day-to-day level that drive daily job monotony and will help uncover the short-term consequences of job monotony that might be overlooked when studying job monotony on an aggregate between-person level as a chronic disposition. We took a first step in that direction by showing that job monotony drives immediate experiences of boredom, which, in turn, result in unhealthy behavior during work in terms of snacking.

Regarding our moderation hypothesis, we were not able to confirm the protective role of trait mindfulness for boredom-related snacking. Although a recent cross-sectional study with obese participants found a direct association between high trait mindfulness and eating as a reaction to boredom (Watford et al., 2019), other studies with normal-weight samples investigating the buffering role of mindfulness for boredom-induced outcomes have been mixed. While some studies confirm a buffering effect of trait mindfulness (e.g., between boredom proneness and problematic smartphone use; Regan et al., 2020), others do not (e.g., between boredom proneness and negative affect; Lee & Zelman, 2019). Thus, more research on the buffering potential of trait mindfulness is needed.

Limitations and Suggestions for Future Research

Despite the strengths of our study, it also has several limitations. First, our study used self-report measures, raising the issue of common method bias. To address this, we group-mean centered our Level 1 predictors. In doing so, we removed the between-person variance, ensuring that the relationships between our within-person constructs are not influenced by between-person factors, such as individual differences, response tendencies and social desirability, thereby obtaining unbiased estimates of the within-person relationships in our model (Gabriel et al., 2019). To further remedy common-method bias, future studies could employ multi-source measurements by, for instance, asking supervisors or peers to evaluate the monotony of employee’s work. Additionally, researchers have raised awareness for potential priming effects depending on sequencing of measures in surveys (Podsakoff et al., 2003). To minimize sequencing effects, our predictor and mediator variables were presented in a random order to avoid any systematic biases. Unhealthy snacking as our outcome variable was assessed after the predictor and mediator variables. Yet, while this in principle might bias reporting, we believe that such concerns are mitigated by the fact that the unhealthy snacking scale requires people to report concrete behavior that has either already taken place or not (i.e., someone either consumed certain snacks or not during the workday).

Second, our sample is highly homogenous in terms of educational level: almost 80% of participants had obtained at least a Bachelor’s degree from a university. Prior research has shown that generally, individuals with a university degree are much less likely to engage in unhealthy snacking than those with a vocational or high school background (Liu et al., 2015). Hence, it is plausible that the effects in our study might be different in samples with lower educational levels. Similarly, given that our sample consisted predominantly of individuals with normal weight (66%; according to the BMI classification), effects found in our study might be different in samples with overweight and obesity. Particularly, prior studies have shown that individuals with obesity are less able to resist eating high-caloric snacks in response to negative mood (Udo et al., 2013). Future research should therefore test and compare our results in various sampling populations (i.e., individuals with normal weight and individuals with overweight and obesity). Additionally, given that practical recommendations to reduce unhealthy snacking behaviors often suggest substituting unhealthy snacks with more healthy options (Franja et al., 2021; Lloyd-Williams et al., 2009), future studies could also investigate the effect of under-stimulating work characteristics on healthy snacking behaviors.

Third, other moderators may play a role for the relationships examined in our study. For instance, boredom proneness (a personality trait reflecting the general tendency to experience boredom across situations) might increase the positive relationship between work-related boredom and unhealthy snacking, as it has been linked to increased impulsivity and poor impulse control (Vodanovich, 2003) as well as eating in response to boredom (Crockett et al., 2015). In contrast, trait self-control may act as a buffer between boredom and unhealthy snacking. As trait self-control is “the ability to withstand impulses for immediate gratification of one’s needs in view of one’s long term-goals” (Adriaanse et al., 2014; p. 1), individuals with high trait self-control are less likely to engage in impulsive eating (e.g., when their self-regulatory resources are depleted; Wang et al., 2015). Future studies should therefore test other potential boundary conditions.

Practical Implications

Prior research has shown that reducing unhealthy snacking by just one portion per day could have major implications for health outcomes and weight gain. For example, replacing one unhealthy snack with one healthy snack per day was estimated to prevent as many as 6000 yearly deaths from cardiovascular diseases in the UK (Lloyd-Williams et al., 2009). Furthermore, reducing daily energy intake by as little as 100 calories - less than one portion of most snacks – has been estimated to potentially be sufficient to prevent weight gain in the majority of the population (Hill et al., 2003). Our study shows that chances of engaging in unhealthy snacking are significantly higher on days when employees experience monotony and boredom when working, thereby highlighting the detrimental role of under-stimulating low arousal work characteristics for employees’ health.

Furthermore, our study emphasizes the unique role of daily job monotony and work-related boredom for unhealthy snacking behaviors, supporting the notion that job monotony and boredom should be addressed and treated separately rather than confounding the two constructs (van Hooff & van Hooft, 2016). Although we did not find support for our hypothesized buffering effect of trait mindfulness, it is important to discuss what organizations and employees can do to limit the detrimental effects of job monotony. This is particularly important considering that intra-individual variance of job monotony (i.e., day-to-day fluctuations) is larger than differences between individuals, thus, emphasizing its prevalence and dynamic character in any type of job. By separating the measurement of job monotony and work-related boredom in our study, we help identifying what type of interventions may prove more or less successful, based on the specific construct (i.e., the job itself or the subjective perception) they target. For instance, organizational measures may address job monotony by redesigning work tasks to have more variety through job enrichment procedures. Targeting the experience of boredom may also be possible but more challenging as individually tailored interventions may be better suited here. For example, employees may engage in job crafting and reorganize their tasks and work environment to suit their personal abilities and needs, such as learning new things and starting new projects, in order to decrease feelings of boredom while working (van Hooff & van Hooft, 2014). Such job crafting activities may improve one’s working environment by making work tasks more interesting, challenging and satisfactory, therefore making employees less likely to engage in compensatory behaviors, such as unhealthy snacking (van Hooff & van Hooft, 2014).

Conclusion

The present study advances the work-related snacking literature by providing evidence for the detrimental effect of daily job monotony on work-related boredom and unhealthy snacking during work through boredom. Importantly, these effects persist above and beyond over-stimulating work stressors (i.e., daily workload) and alternative underlying mechanisms (i.e., daily negative affect). Furthermore, our findings emphasize the dynamic nature of daily job monotony and work-related boredom, challenging the common conceptualization of jobs as either monotonous or not.