Introduction

Attention is considered a limited cognitive resource (for reviews, see Thomson et al., 2015, 2016). The ability to sustain attention on a task is influenced by individual and contextual factors (Stenfors et al., 2019; Taylor et al., 2016). Momentary lapses in attention can result in minor inconvenience and embarrassment in everyday life, such as losing one’s train of thought in a meeting, but also potentially detrimental outcomes, such as traffic accidents or medical malpractice (Broadbent et al., 1982; Carriere et al., 2008). The ability to sustain attention is associated with better academic (e.g., Unsworth et al., 2012; Isbell et al., 2018) and work outcomes (Tan & Thamarapani, 2018). The advent of smart devices and social media has made sustaining attention increasingly valuable yet challenging (e.g., Stenfors et al., 2019). Learning effective ways to increase and apply attentional resources is essential for improving task performance as well as long-term emotional well-being (Livingstone & Isaacowitz, 2017; Sohlberg & Mateer, 2001). Attention and boredom have been theorized to be intertwined (Carriere et al., 2008; Cheyne et al., 2006; Damrad-Frye & Laird, 1989; Danckert & Merrifield, 2018; Hunter & Eastwood, 2018; Tam et al., 2021a, b), with some evidence suggesting a causal relationship between them (Westgate & Wilson, 2018). In this study, we developed and tested an intervention to improve sustained attention and reduce boredom.

Attention regulation and sustained attention

Attention can be categorized into two distinct processes: bottom-up attention, and top-down attention (Katsuki & Constantinidis, 2014; Posner & DiGirolamo, 1998; Theeuwes, 1991). Bottom-up attention is known as an involuntary, automatic process that is driven by external stimuli. This process could also be referred to as the default-mode network (DMN), and empirical findings suggest this as the state of mind wandering that allows us to actively respond to external stimuli in a timely manner to ensure safety while at rest. However, such responses may not be entirely relevant to a person’s will and desires. Hyperconnectivity with the DMN was shown to be related to rumination and proneness for depression (Chou et al., 2023; Kringelbach & Berridge, 2009; Stawarczyk et al., 2012) and negatively correlated to subjective well-being (Shi et al., 2018).

Top-down attention is the voluntary and goal-directed process that allows individuals to selectively attend, focus, and suppress attention to unwanted stimuli based on an individual’s existing knowledge, goals, and intentions. This process involves cognitive control, including attentional control and executive functioning (Diamond, 2013; Hopfinger et al., 2000; Hopfinger & Slotnick, 2020). Attentional control is commonly known as the cognitive processes that involve exerting control on focused attention, sustained attention, selective attention, alternating attention, and divided attention to regulate information processing that facilitates goal-directed behaviors despite distractions. In particular, sustained attention is a crucial attentional function that entails an array of cognitive processes and cognitive capacities that actively select and process specific information while tuning away from other stimuli (Engle, 2018; Esterman & Rothlein, 2019; Sohlberg & Mateer, 2001).

The recruitment of effective cognitive control requires effort (Thomson et al., 2015). Mackworth (1948) first characterized the psychological construct of sustained attention (or vigilance) and developed the assessment of the state of readiness to detect and respond to rare targets among non-targets over prolonged periods of time. His studies observed a vigilance decrement and performance decline over time. Such findings promoted the theory that sustained attention is a limited resource (Thomson et al., 2015; Warm et al., 2008). Beyond understanding the capacity for sustained attention from the vigilance decrement perspective, research suggests that baseline attentional resource fluctuates throughout the day, depending on one’s arousal state associated with homeostatic balance and circadian rhythm (Esterman et al., 2013; Kucyi et al., 2017; Rosenberg et al., 2016). Mind wandering is found to be associated with hypoarousal, whereas distractibility to the external environment is associated with hyperarousal (Unsworth & Robison, 2018). Studies show that sustained attention can be improved by manipulating arousal levels (Grillon et al., 2016; O’Connell et al., 2008).

Factors that compromise attention also include the level of cognitive demands of a task and one’s motivation. Studies showed an increased rate of mind wandering during easy and difficult tasks (Seli et al., 2016a, b; Smallwood & Andrews-Hanna, 2013; Thomson et al., 2013). More specifically, studies showed that an increase in intentional mind wandering was found to be associated with easy tasks, while unintentional mind wandering was found to increase with task difficulty (Seli et al., 2016a, b). These findings highlight the notion of intentionality associated with task-unrelated thoughts, suggesting a role of cognitive control in the allocation of attentional resources. The opportunity cost model suggests that the cognitive effort to sustain attention is distributed based on the perceived reward, value, and motivation of the tasks at hand, compared to the value of other possible mental activities available (Kurzban et al., 2013). Findings suggest that manipulation of sustaining one’s motivation helps reduce vigilance decrements and attention lapses (Esterman et al., 2016). These processes are believed to involve the recruitment of executive control, which generally entails inhibitory control, working memory, and cognitive flexibility. These functions are essential to support higher-level executive functioning in tasks such as reasoning, planning, and problem-solving. Attentional control and executive functioning are distinct constructs (Deodhar & Bertenthal, 2023; Friedman & Miyake, 2017) that are crucial in facilitating the adequate allocation of attentional resources that is essential for efficient functioning (Diamond, 2013; Hopfinger et al., 2000; Hopfinger & Slotnick, 2020).

Attention and boredom

Boredom is described as an aversive and unpleasant experience when an individual desires but fails to engage in a satisfying activity from their environment (Eastwood et al., 2012). Although boredom shares some elements of other negative affective states, it is a distinct emotion with its own set of feelings, cognitions, motivations, action tendencies, and expressions (Smith et al., 2009; van Tilburg & Igou, 2012). Boredom is found to be correlated with both high arousal states, such as anger, anxiety, and frustrations, and low arousal states, such as fatigue and loneliness (Fahlman et al., 2013; Havermans et al., 2015; Perkins & Hill, 1985; Tam & Chan, 2019; van Tilburg et al., 2019). Frequent and intensive experiences of boredom have shown to be associated with various potentially detrimental outcomes, such as aggression (Pfattheicher et al., 2021), school dropout (Bridgeland et al., 2006; Tvedt et al., 2021), and emotional eating (Crockett et al., 2015), but also desirable outcomes, such as meaning making (Coughlan et al., 2017; van Tilburg & Igou, 2011; Yang et al., 2019). Furthermore, particularly among adolescents and college students, boredom (Weybright et al., 2020) and chronic boredom (Gu et al., 2023) appear to be on the rise. Given the potential adverse behavioral and mental health implications of boredom, especially chronic boredom, it is imperative to identify constructive means to cope with it.

The experience of boredom activates similar regions of the DMN (Danckert & Merrifield, 2018), which is known as the state of mind-wandering that could happen during attentional lapses (Carriere et al., 2008; Cheyne et al., 2006). This helps to support the notion that boredom represents a disengaged attentional state. Theorists have made various attempts to explain the underlying mechanism and association between attention and boredom. Some propose that attention failures occur when there is a drop in cortical arousal after prolonged exposure to monotonous tasks, resulting in an attention shift from task-at-hand towards more rewarding stimuli, leading to the experience of boredom (e.g., O’Hanlon, 1981). Others suggest that sustained attention failure amplifies the feelings of mental effort, or perception of non-optimal internal or external stimulation, which results in the experience of state boredom (Fisher, 1998; Leary et al., 1986; Martin et al., 2006; Merrifield & Danckert, 2014). Damrad-Frye and Laird (1989) highlight the role of self-perception and suggest that feelings of boredom derive from a self-perception of actions. Particularly, they suggest self-awareness of attentional experiences to be one important contributing factor to boredom experiences. Empirical findings also reveal that boredom experience is associated with self-awareness of attention lapse, whilst the feelings of boredom and performance errors remain as separate consequences of lapses in attention (Carriere et al., 2008; Cheyne et al., 2006). Eastwood et al. (2012) further postulate boredom experience as an aversive experience during a state of attentional disengagement that occurs when there is an awareness of a discrepancy between an individual’s intention to attend and actual attentional activities. The Meaning and Attentional Components (MAC; Westgate & Wilson, 2018) Model proposes boredom experiences as a result of a mismatch of attentional resources, as well as a mismatch between one’s values and activity, and posits attention and meaning as two orthogonal predictors of boredom. These studies highlight the role of conscious awareness on the lapse of attention in engagement with the world and maintaining motivation, and boredom experience could be a consequence.

The functional account of state boredom suggests boredom, like other emotions, has a unique influence on thoughts and behaviors. Boredom may potentially serve to motivate us to pursue alternative goals out of a boredom-inducing situation (Bench & Lench, 2013), prompts us to regulate and redirect our attention and resources to seek rewarding opportunities and meaningful pursuits (van Tilburg & Igou, 2012; van Tilburg et al., 2013), and yields us to opportunities for the continual pursuit of a meaningful life and facilitates personal growth (Elpidorou, 2018). These accounts suggest that the discomfort of boredom experience could function as a signal of cognitive dissonance, or a form of internal conflict, that is resolved by pursuing consistencies between one’s values and behaviors.

Tam et al. (2021a, b) offer the Boredom Feedback Model to help explain the mechanism between attention and boredom experience. The model suggests that state boredom may arise when there is inadequate attentional engagement (IAE), which is characterized by the discrepancies between one’s desired and actual levels of attentional engagement. IAE can occur in situations when a person is obligated but fails to attend to a task or when the person struggles to find something to do despite a strong desire for attentional engagement. The model stipulates that, upon IAE, the person’s attention will either shift externally, internally, or back to the original boredom-inducing task. This feedback loop continues until the person’s attentional resource is optimally engaged–that is, the gap between the desired and actual levels of attentional engagement closes by lowering the former, increasing the latter, or both. If IAE continues after several iterations of the feedback loop, the experience of boredom may persist and be amplified through operant and classical conditioning and develop into chronic boredom associated with other negative emotions and problematic behaviors in the long run (Tam et al., 2021a, b).

While the exact relationship between IAE and boredom warrants further theorizing and empirical tests, in this study, we extrapolate the constructivist view of emotions (e.g., Barrett, 2017) to understand boredom, like other emotions, as a concept that the mind uses to describe sensory inputs based on past experiences and the current context. IAE is plausibly one form of interoception that the mind conceptualizes as “boredom,” which prompts the exploration of alternative, more rewarding targets for spending attentional resources.

It is important to identify strategies that help individuals effectively cope with and overcome boredom. Attention training has been suggested as a possible way to reduce the experience of boredom (Hunter et al., 2016; Kenah et al., 2018; Macklem, 2015). To date, however, studies on interventions for boredom are lacking. The current study intended to develop an intervention that improves one’s ability to engage their attention by closing the gap between the desired and actual levels of attentional engagement (i.e., reducing IAE) so as to enhance their capability to break the boredom feedback loop.

Attentional training

The area of attentional training has predominantly focused on treatment and rehabilitation for those with disordered attention (e.g., those formally diagnosed with ADHD or with brain injuries) (Bartfai et al., 2022; Park & Ingles, 2001; Sohlberg & Mateer, 1987, for a review, see Peng & Miller, 2016). Few studies, however, emphasize the generalizability of attentional training to the non-clinical population. Sohlberg and Mateer (2001) adapt rehabilitation techniques to improve the attention of adults with ADHD. Many of the strategies could potentially be applied in daily life situations by the general public. Five components that are commonly used by clinicians to address attention difficulties are (i) environmental modifications and support, (ii) attention training, (iii) self-regulatory strategies, (iv) use of external aids, and (v) psychosocial support (Sohlberg & Mateer, 2001). These strategies, which we elaborate on below, were integrated into the intervention program evaluated in the current study.

Environmental modifications and support

This module directly alters the physical or digital workspace to reduce external distractions (e.g., noise and visual clutter). Considering attention is a limited resource, a suitable environment is crucial for minimizing cognitive effort in managing attentional difficulties. Stenfors and colleagues (2019) found that cognitive processing performance declines in the presence of various distractions from urban environments, such as other people and advertisements. As such, equipping individuals with the ability to assess and construct a suitable environment for work might be helpful in improving attention. This can include learning to identify the difficult and helpful environmental factors for mental activities, strategies to reduce distractions, and modifying the environment for optimal cognitive processing (Sohlberg, 2000).

Attention control training

The attention training component includes specific tasks and exercises designed to strengthen the participant’s core attentional capacities, such as focusing exercises or practices that increase awareness of attention shifts. The training material was initially designed to facilitate changes in cognitive capacity for those who suffered from acquired brain injuries through repeated activation and stimulation of the attentional system (Niemann et al., 1990; Park, 1999; Sohlberg & Mateer, 1987; Sohlberg et al., 2000). The training involves the practice of a series of repetitive exercises that target various types of attention in a hierarchical order, with increasing attentional demands. Post-training improvements in attention levels, memory, learning, and overall levels of independent living were found among individuals with acquired brain injuries (Park & Ingles, 2001).

Self-regulation strategies

Self-regulation strategies are a range of strategies designed for individuals with brain injuries to modify their engagement with attentional resources. An example of this is the use of self-instructions and orienting procedures that help individuals to combat distractions and to remain on task (Winograd & Hare, 1988). Strategies such as pacing help individuals learn to set realistic expectations of productivity and overcome fatigue. This involves learning the skills of working in short intervals, and scheduling breaks to maintain an optimal arousal level and minimize fluctuation of attentional states throughout the day. Skills to manage interruptions from the environment (e.g., phone calls or text messages) and internal interruptions (e.g., a thought unrelated to the current task) to help individuals remain on task were also considered helpful in expanding one’s capacity for cognitive control and allocation of attentional resources.

Use of external aids

This module encourages the direct use of tools to manage tasks and reduce the cognitive load associated with remembering and organizing tasks. The use of external devices such as calendars and tracking applications could help relieve the cognitive resources and the reliance on attentional resources it takes for planning and remembering extensive information (e.g., Risko & Dunn, 2015).

Psychosocial support

This training component aims to provide emotional and social support to participants, which can potentially improve mental health and indirectly support better attention through reduced stress and anxiety. Sohlberg and Mateer (2001) emphasize the importance of emotional support in cognitive training interventions concerning the well-established connection between emotional states and cognitive functioning (Tyng et al., 2017). This includes stress management and social support.

Current study

The current study developed and tested a four-week attention training program that is informed by Boredom Feedback Model and is based on the aforementioned five approaches. The aim of the training was to enhance the participant’s ability to resolve the discrepancies between the desired and actual attentional levels during a potentially boring task or when having difficulty choosing what to do. We investigated (i) the effectiveness of the attention enhancement program outlined above and (ii) whether improvement in sustained attention leads to a reduction of boredom experience.

The five components theoretically improve attention through both direct (primary) and indirect (secondary) means. Primary interventions, including the modules on environmental modifications and support, attention training, and self-regulatory strategies, directly target attention enhancement and distraction reduction. These are techniques and practices designed to immediately improve the ability to focus and sustain attention on tasks.

Secondary interventions are strategies that indirectly contribute to improving attention by enhancing overall psychological well-being. These interventions, including the use of external aids and psychosocial support, address the more nuanced aspects of attention management, including motivational and emotional factors. While primary interventions directly enhance the cognitive aspects of attention, secondary interventions support the underlying motivational and emotional states necessary for sustained attention and engagement.

In terms of their potential impact on boredom, primary interventions directly impact attention by reducing distractions and enhancing cognitive control, leading to reduced boredom by increasing attentional engagement with tasks at hand. Secondary interventions improve psychological well-being and motivation, indirectly supporting sustained attention by increasing perceived resources and desire for engaging attention. This enhanced attentional engagement with tasks reduces the likelihood of experiencing boredom.

A summary of the content of the attention training is presented in Table 1.

Table 1 Summary of the content of attention training
Full size table

Previous studies raise concerns about studying attention and boredom experiences under an artificial experimental setting (e.g., requiring participants to watch an excruciatingly boring video or a computerized task that is understimulating as a means to manipulate attention), which could be ecologically invalid and potentially ungeneralizable to other contexts (Hunter & Eastwood, 2018; Kida et al., 2017; Westgate & Wilson, 2018). This study is interested in the level of attention and state boredom associated with various activities in real-world settings. Participants’ daily experiences of attention and state boredom, as well as the activities they engaged in, were studied using ecological momentary assessment (EMA).

Even among tasks specifically designed to induce boredom, there can be considerable variation in the degree to which they elicit this emotion. A study that explored different boredom-inducing tasks found variability in the extent to which tasks induce different levels of intensity in boredom experiences as well as other emotional experiences (Markey et al., 2014). As such, task differences in affecting boredom experience and attention level were also accounted for.

Given the relationship between boredom experience and attention, the current study intended to influence an individual’s experience of boredom in natural settings by enhancing factors that are believed to have contributed to one’s day-to-day attentional resources. In this study, attentional resources were seen as a constant negotiation between the external demands (e.g., the nature and level of complexity of the task at hand), the individual’s motivation to engage with such demands, and the individual’s internal capacity to meet the demands. The current study also postulated that by minimizing the discrepancies between these factors, one could maximize their attention resources to each particular everyday life situation. Thus, the attention training focused on training individual of the normal population with skills in (i) preserving and enhancing internal capacity, this involved cognitive training, mood and stress management, sleep habits, environmental modification, and use of external devices; (ii) matching their resources with external demands, which involves planning and self-organizational skills; and (iii) self-regulatory skills that involve engaging with one’s higher level thinking process to reassess and align one’s everyday life choices and their values, as to maximize ones’ attentional resources and overall sense of satisfaction in daily life experiences.

To the best of our knowledge, the present study is the first to examine the impact of attention training on boredom experience and boredom proneness.

Hypotheses

  1. 1.

    Compared to the control group, participants in the intervention group experienced greater improvements in the level of attention in everyday life situations throughout the four-week attention training.

  2. 2.

    The reduction of the level of state boredom is followed by improved attention in everyday situations.

As an exploratory aim, we also compared the boredom proneness of the two groups to assess the influence of the intervention on trait-level boredom.

Method

This was a two-arm between-subject quasi-experiment. Data was collected from a four-week experience sampling and questionnaire at baseline and post-intervention.

Participants

Participants of the study were recruited via posters, advertisements, and social media. Eligible participants were at least 18 years old and were permanent residents of Hong Kong. Exclusion criteria included known neurological conditions, severe head trauma, psychosis, or learning disabilities.

Recruitment began three weeks before the attention training. Participants who indicated that they were able to attend the training on the designated dates were included in the intervention group. Participants who signed up after the official recruitment period and those who indicated that they were unable to attend the training sessions were allocated to the control group. Participants in the control group joined the study knowing that no intervention would be provided. A control group instead of waitlist control was used as previous studies suggested that a waitlist control group could exaggerate the treatment effects of research studies due to the nocebo effect (e.g., Furukawa et al., 2014).

To the best of our knowledge, there are no intervention studies that target boredom, let alone with an EMA design. As such, no a priori power analysis was conducted. The study sample consisted of 73 healthy adults (63% female). The age of the participants ranged from 18 to 35, with a mean of 25.5 (SD = 4.0); 40% of the participants were employed, and the rest were full-time students. All participants have received at least some tertiary education. They all self-identified as ethnically Chinese. Thirty-two participants were assigned to the intervention group; all of them attended all four attention training sessions. Forty-two participants were assigned to the control group. The intervention group was younger and comprised of more students than the control group (Table 2). Among the attention training group, no attrition was observed post-intervention. Among the control group, 58.5% did not respond to the follow-up questionnaire; however, their EMA data (submitted before the follow-up questionnaire) were included in the analyses.

Table 2 Demographic information
Full size table

Procedure

Participants from both groups received an email link to fill out the consent forms and an online baseline questionnaire (T1). After completion, they received an informational email with instructions on the installation of the smartphone application PACO for the four-week EMA. Participants in both groups received notifications thrice per day to respond to questions regarding their current activities, attention level, and boredom for 28 consecutive days. Across this period, participants from the intervention group received four weekly group sessions of attention training. The weekly sessions were conducted online via Zoom for four consecutive weeks. The duration of each session was 50 min. Data collection for four-week EMA for the intervention group began on the day of the first training session and ended one-week post-training (T2). Data collection of four-week experience sampling for the control group began after they completed the baseline survey. Participants from both groups were invited to fill out a follow-up questionnaire upon completion of the four-week EMA. Participants from both groups were rewarded HKD $100 (~ USD $12.7) as a token of appreciation for completing the 28-day study, including the two online questionnaires. Across the 28 days, the average number of days with entry was 26.3 (2.49) and 24.0 (5.31) among the intervention and control groups, respectively. The ethics approval was obtained by the Departmental Research Ethics Committee of the Department of Psychology at The University of Hong Kong prior to the commencement of the study. Prior to data collection, participants provided informed consent.

Attention training

The training focuses on enhancing the capacity to exert control on the top-down attentional process, which involves both enhancing attentional control and elements of executive functioning. It also targeted the bottom-up attentional process by including modules that aimed to modify one’s immediate environment. The outline of the training program is reported in Table 1; it is largely based on an existing program designed for patients with ADHD (Sohlberg & Mateer, 2001), integrated with other empirically supported treatment components commonly found in concepts of metacognitive therapy for attention and refocusing techniques, as well as cognitive behavioral therapy for mood, anxiety, and stress management (Curtiss et al., 2021; Pozuelos et al., 2019; Sturmey, 2009; Uphoff et al., 2019; Vogel et al., 2016). This training was delivered by the first author who, at the time of the study, was a trainee in clinical psychology. The training material used in this study is available upon request.

In the first week, participants were given psychoeducation on how sustained attention works and were taught skills such as the use of orienting procedures (Sohlberg & Mateer, 2001; Winograd & Hare, 1988) to improve their sustained attention. Participants were assigned to practice sustained attention on a specific task for up to 25 min each day for a week. In the second week, participants were equipped with skills to preserve their mental energy whilst maximizing their sustained attention by working in successions of short intervals with frequent short breaks. The Pomodoro technique (Cirillo, 2006) was introduced as a self-regulatory strategy to preserve mental effort and to increase a sense of self-control through practicing pacing and task-switching. Participants were also taught to schedule their work according to their energy level for the day. In the third week, participants were guided to reflect on their values (Lundgren et al., 2012), followed by an introduction to the use of external aids, such as making use of explicit reminders of their values and allocating time and dates for important tasks that are value- or goal-congruent. In the final week, participants were provided with psychoeducation on the maintenance of emotional well-being and strategies to work under stress and emotional distress, such as behavioral activation and scheduling of worry time (Jacobson et al., 2001; Kelly, 2002, 2003).

Measures

Ecological momentary assessment

Attention was measured using a single question, “How well can you focus on this task?”. The question was rated on a five-point scale where 1 = “no focus on the task,” 3 = “50% of your focus on the task”, and 5 = “100% focus on your task”. A higher average score on the scale denotes a higher level of attention during the tasks at hand. We also asked the participants to report whether they were engaging in one of the six life domains: (i) work (e.g., assignments, revision, lectures, report writing), (ii) leisure (e.g., having a conversation, watching a movie, reading a book, handcrafts), (iii) commute (e.g., public transportation), (iv) personal chores and errands (e.g., preparing meals, cleaning up), (v) not doing anything, and (vi) others. These six domains were selected from the nine domains in the Everyday Life Attention Scale (ELAS; Groen et al., 2019) based on their relevance to our targeted population.

Boredom. Experience of state boredom was measured using a single-item question on a five-point scale that ranged from 1 = “not bored at all” to 5 = “extremely bored.”

Baseline and post-study measurement

Boredom Proneness (T1 & T2). The Short Boredom Proneness Scale (SBPS; Struk et al., 2017) was used to measure one’s tendency to experience boredom. Participants’ boredom proneness was measured in the first and last weeks of the study. The SBPS consists of eight items rated on a 7-point Likert-type scale ranging from 1 = “strongly disagree” to 7 = “strongly agree”. Higher scores indicate a higher tendency to experience boredom. Cronbach’s alpha at baseline and post-study was α = 0.83 and α = 0.86, respectively.

Demographic Information. At T1, participants indicated their age, educational level, occupation, and gender. They were also asked to indicate whether they had ever been diagnosed with developmental or mental disorders.

Statistical analyses

Multilevel modeling (MLM) was used for linear growth model analysis. Prior to the analysis, the Intra-Class Correlations (ICC) were performed to determine the suitability of proceeding with multilevel modeling analysis. A linear mixed model was conducted with lme4 and lmerTest packages in R statistical software version 4.1.0 (R Core Team, 2021). A fixed effect linear mixed model was conducted to investigate the change of slope in attention and boredom experience interacted with intervention effects and activities over the four weeks, followed by examining the random effect (i.e., within-person variation) in the level of attention and boredom on each day varied from the person’s intercept (level 1) and the random effect capturing between-person variation throughout the four weeks (level 2). Data collected after 28 days and up to 40 days were included. The full dataset for a four-week between-group comparison would have consisted of up to 73 (Participants) * 28 (Days) * 3 (Timepoints) = 6,216 observations. The potential moderating effect of types of activities was tested using 3-way interactions (i.e., time by group by activity). Data and R codes are available on OSF: https://osf.io/5tks3/?view_only=429517e71e1b4c8589bde3b0de474101.

Results

The overall descriptive statistics of the key variables are reported in Table 3.

Table 3 Descriptive statistics of the key variables
Full size table

Multilevel analysis

In total, the dataset consisted of 4,215 observations across 73 participants. ICC indicated that 36% of the variance in the level of attention and 28% of the variance in the level of state boredom stemmed from individual differences. Therefore, the use of multilevel analyses was justified.

The summary of the model of fit of the linear growth models in the level of attention is reported in Table 4. The chi-square difference test revealed a significant difference between Model (Days) and Model (Groups), χ² (1) = 11.113, p < .001. There was also a significant difference between Model (Groups) and Model (Days*Groups) χ² (1) = 100.606, p < .001, indicating a significant difference in growth in the level of attention between the intervention and control groups over the four weeks.

Table 4 Model fit indices for multilevel linear growth model in level of attention
Full size table

The summary of the model of fit of the linear growth models in the level of boredom is presented in Table 5. Significant chi-square changes between Model (Days) and Model (Groups), χ² (1) = 9.898, p < .001, and between Model (Groups) and Model (Days*Groups) χ² (1) = 23.202, p < .001 were found. The results revealed a significant difference in growth in the level of boredom between the intervention and control groups over the four weeks.

Table 5 Model fit indices for multilevel linear growth model in the level of boredom
Full size table

Temporal changes in attention

The parameter estimates for the linear growth model of increase in attention as a function of intervention are reported in Table 6. The interpretation of the estimates is as follows: (i) the intercept is the level of attention at day 1 of week 1 for the control group, (ii) the group estimate is the attention difference (intervention minus control) at day 1 of week 1, (iii) the time estimate is the change in attention in the control group over the four weeks of the study, and (iv) the time-by-group interaction is the difference in attention change between the intervention and control groups.

Table 6 Parameter estimates for linear growth model of increase in attention as a function of intervention
Full size table

Both groups showed an initial level of approximately 3.5 units on a 5-point scale (control = 3.545; intervention = 3.545 + (-0.045) = 3.500). No group difference in initial level of attention was found. Over the four weeks of experience sampling, the control group showed a -0.017 unit decrease in attention per day, whereas the intervention group showed a -0.017 + 0.033 = 0.016 unit increase in attention per day. The slope difference per day due to group was significant, B = 0.033 (SE = 0.004), p < .001.

The average slope (Fig. 1) shows how the level of attention increased throughout the four weeks of intervention in the intervention group. The level of attention decreased throughout the four weeks in the control group, which was not expected.

Fig. 1
figure 1

Linear growth model of increase in attention as a function of intervention

Full size image

Temporal changes in boredom

Table 7 presents the summary of the results of the linear growth model in the level of boredom. There was no group difference in the initial level of boredom on a 1 to 5 scale, control = 2.595; intervention = 2.595 + (-0.281) = 2.314. Over the four weeks of experience sampling, the control group showed a 0.008 unit increase in boredom per day, whereas the intervention group showed a 0.008 + (− 0.015) = -0.007 unit decrease in boredom per day. The slope difference per day due to group was significant, B = -0.015 (SE = 0.003), p < .001.

Table 7 Parameter estimates for linear growth model of decrease in boredom as a function of intervention
Full size table

As illustrated in Fig. 2, the level of boredom declined throughout the four weeks of intervention in the intervention group. Meanwhile, the level of boredom increased throughout the four weeks in the control group.

Fig. 2
figure 2

Linear growth model of decrease in boredom as a function of intervention

Full size image

The moderating effect of task differences

To investigate whether the intervention effects had specific influences on participants’ level of attention and boredom in different daily tasks, a 3-way interaction between activities and intervention throughout the study on the level of attention and boredom was analyzed. The model of fit of the 3-way interaction linear growth model in the level of attention and in the level of boredom are presented in Tables 8 and 9.

Table 8 Model fit indices for multilevel linear growth model in levels of attention and boredom influenced by activities and intervention over time
Full size table
Table 9 Parameter estimates for linear growth model of the level of attention influenced by different activities
Full size table

The chi-square difference test indices in Table 8 reveal significant differences between Model 1 (no interaction) and Model 2 (2-way interaction), χ² (11) = 148.102, p < .001. There was also a significant difference between Model 2 and Model 3 (3-way interaction), χ² (5) = 11.092, p < .05, indicating Model 3 as the best-fitted model, which suggests that there were significant interaction effects between activities and intervention on the level of attention over time.

When boredom was the outcome variable (Table 8), the chi-square difference test indices show a significant difference between Model 1 (no interaction) and Model 2 (2-way interaction), χ² (11) = 92.297, p < .001. No significant difference between Model 2 and Model 3 (3-way interaction), χ² (5) = 8.482, p = .132 was found. This indicates that activities did not influence the experience of boredom for the two groups differently over time.

The nature of activities participants engaged in explained 76% of the within-person variance in level of attention and 24% of the between-person variance in level of attention. As reported in Table 9, there was no change in the level of attention when participants were engaged with work in the control group over time, β = -0.0033, p = .320. However, there was significant growth in attention during work for the intervention group over time, β = (-0.0033) + 0.02379 = 0.02049, p < .001.

We also found a larger difference in change of level of attention between groups when they were doing nothing compared to during work, B = -0.03, p < .05. Over time, there was a significant decrease in attention for the control group when they were doing nothing compared to when they were working, β = (-0.0033) + (-0.0294) = -0.0327, p < .001.

Boredom proneness

An ANCOVA was conducted to test the efficacy of the attention training on boredom proneness with group as the predictor and pre-test boredom proneness as the covariate. The group variable was statistically significant, F(1,50) = 8.18, p = .006, eta-squared = 0.141. To obtain the effect sizes of between-group differences, we conducted a series of t-tests. At baseline, the independent t-test reveals no significant difference in scores on boredom proneness between the intervention group (M = 35.04, SD = 7.58) and the control group (M = 32.16, SD = 10.84), t(51) = 1.129, p = .264, d = 0.31. At post-test, compared to the control group (M = 33.36, SD = 10.5), the attention training group (M = 28.64, SD = 8.36) had significant lower levels of boredom proneness, t(47) = 2.99, p < .001, d = 0.50. No significant difference between the scores on boredom proneness at pre- vs. post-test was found in the control group.

Discussion

This study had two aims. The first was to test the efficacy of an attention training program that is generalizable to everyday tasks. The second aim was to examine whether improved attention also, in turn, attenuated boredom experience. The results support both hypotheses: those who received the four-week attention training showed a significant improvement in the level of attention in everyday life situations compared to participants who did not receive training. Those who received the training also showed a significant decline in the level of state boredom compared to the control group. Our exploratory analysis also suggests that the intervention was effective in reducing boredom proneness.

Attention training has been suggested as a possible way to reduce the experience of boredom based on the association found between attention and boredom (Carriere et al., 2008; Cheyne et al., 2006; Damrad-Frye & Laird, 1989; Danckert & Merrifield, 2018; Hunter & Eastwood, 2018; Westgate & Wilson, 2018). However, to the best of our knowledge, there has been no intervention study focusing on mitigating or coping with boredom experiences through attention training. Westgate and Wilson (2018; Study 2) attempt to experimentally manipulate attention by modifying the difficulty of a computerized air-traffic control simulation task and not by directly manipulating attention per se. Our study contributes to the investigation into the causal relationship between attention and boredom by showing that skills and techniques that are theoretically and empirically related to the improvement in the former can mitigate the latter. Interpreted from the perspective of the Boredom Feedback Model (Tam et al., 2021a, b), our attention training might have improved the intention to attend, the resources needed to attend, the ability to notice the shift in attention, and the efficiency in redirecting attention either to the task or an alternative task. The exact processes and potential individual differences, however, warrant further investigation.

Moreover, while no initial difference in boredom proneness between the two groups was found, at post-test, the intervention group saw a significantly larger reduction in boredom proneness compared to the control group. The effect size of the between-group difference at post-test was moderate (Cohen’s d = 0.50). To the best of our knowledge, the current study is the first to demonstrate the malleability of boredom proneness through intervention in general and attention training in particular. Boredom proneness is associated with a plethora of adverse outcomes, such as lower subjective well-being (e.g., Bai et al., 2021) and depressive and stress symptoms (e.g., Struk et al., 2017). Attention problems have also been identified as a transdiagnostic feature in psychopathology (e.g., Racer & Dishion, 2012). Our study points towards the possibility of considering attention training as a component of transdiagnostic interventions aimed at improving mental health.

Given that boredom proneness measures the frequency and intensity with which one experiences boredom, as well as the extent to which one perceives his or her life as boring (Tam et al., 2021a, b), this result further reflects that attention training may have an influence on individual’s perception of their boredom experience more generally. The impact on boredom proneness also suggests the construct, as measured by SBPS, is not necessarily stable or even trait-like. Indeed, many theorists have called into question both the definition of boredom proneness and the construct BPS and its short forms actually measure (e.g., van Tilburg et al., 2024). The extent to which the observed difference brought upon by the present attention training reflects a reduction in “the average intensity of boredom in response to a set of representative events over a defined period of time” (p.204, van Tilburg et al., 2024) and changes in the appraisal of one’s life (Tam et al., 2021a, b) warrants further research.

One alternative possibility is that the intervention was effective because it enhanced volitional, self-regulatory processes rather than attention per se.Footnote 1 In this study, the content of attentional training focuses on enhancing one’s internal capacity to gather resources to exert adequate attentional control in everyday life experiences. The outcome measure of the current study was subjective ratings of participants based on real-time reflection of everyday life experiences. To further understand the possible contributing factors towards the change in attentional experiences, we also included an open-ended question asking participants to report what they thought was most helpful in the training program (the full list is available on the project’s OSF page). It was observed that participants commonly found skills to selectively attend to tasks at hand for a limited time, working under emotional distress, organizing tasks, increased awareness of factors affecting attentional resources, and personal values were most helpful in enhancing their subjective attentional experiences. These participant feedbacks reflect that maintaining optimal attentional engagement in daily life entails the coordination of widespread cognitive processing to promote willed attention via a top-down process. As mentioned earlier, the skills training in the current study in part focuses on the top-down attentional process that consists of components of both attentional control and executive functioning. Future studies may further investigate their separate effects on participant’s experiences of boredom. This possibility of alleviating boredom experiences via its enhanced volitional, self-regulatory processes is in line with models of boredom that emphasize self-regulatory processes as a key psychological cause of boredom (e.g., Gorelik & Eastwood, 2024; Struk et al., 2016). These findings may also suggest that an enhanced attentional engagement involves dynamic interactions among different networks that involve higher-level cognitive functioning, including self-reflection, emotional regulation, and cognitive control. Such dynamic interaction is also found to be positively associated with subjective well-being (Shi et al., 2018).

The study further explored whether the effect of the training was particularly observable in some daily activities. Our results showed that the attention training significantly improved the ability to focus, particularly when one is working, compared to those in the control group. Moreover, somewhat surprisingly, in situations in which one was doing nothing, there was a significant decrease in attention among the control group. Such an effect was not found in the experimental group. The finding suggests that the attention training might have helped to prevent a drop in attention in situations when one is doing nothing. However, the nature of the activities did not influence the experience of boredom for either group. This implies that state boredom could be generally reduced by training individuals’ abilities to better engage with attentional resources, and the effect is the same across all activities. According to the Boredom Feedback Model, state boredom arises in the condition of IAE. That is, the discrepancy between the desired level of attentional engagement and the actual state of having nothing to do (i.e., having nothing to behaviorally engage with) is an antecedent to boredom experiences (Tam et al., 2021a, b). By helping participants manage their attentional expectations when they were not explicitly doing anything, their desired level of attentional engagement may be lowered to better match what the situation can afford. This might help prevent the emergence of or attenuate boredom.

The consistency in boredom reduction across diverse activities potentially challenges the notion that boredom’s alleviation is strictly tied to improvements in attention within the context of active engagement alone. Instead, it posits that boredom may be mitigated through mechanisms that are not exclusively dependent on the cognitive demands of a task or the level of attentional engagement it elicits. This invites a reevaluation of the interplay between attention enhancement and boredom reduction; the nature of boredom is multifaceted and, thus, also the factors that contribute to its experience and alleviation.

Although the training focuses on attention, our study cannot examine the temporal relationship between attention and boredom because they were measured concurrently at each EMA time point. Thus, our findings remain indirect in their explication of the causal relationship between attention and boredom. Furthermore, the improvement in attention appears to be stronger in work-related activities compared to when doing nothing, but the improvement in boredom did not show this between-activity variability. This inconsistency in the moderation effect underscores the need to further elucidate the relationship between (self-report) attention and boredom. Our study, in other words, does not resolve the theoretical question about the causality between attention and boredom.

Implications

Tam et al. (2021a, b) suggest that state boredom may arise when there is inadequate attentional engagement. Resolving inadequate attentional engagement may involve the effort and ability to redirect oneself into optimal attentional engagement in order to cope with the state of boredom and to successfully engage with meaningful tasks. The current study postulates that by improving one’s attention, one may have an improved chance to break through the state of boredom, which serves to pivot us from being stuck with boredom-inducing tasks to striving for challenges for optimal stimulations (Dahlen et al., 2004). Findings from the current study support such proposition and suggest that with improved ability in engaging with attentional resources, an individual might have a higher degree of flexibility in gathering and orienting their attention resources with tasks at hand, with a better chance of resolving the discrepancy between desired and actual levels of attentional engagement, hence, to avoid chronic boredom or cope with boredom more effectively. Given that prolonged experiences of boredom were found to be associated with negative behavioral outcomes (Crockett et al., 2015), this is particularly timely as studies have indicated an increase in boredom experience and boredom proneness, especially among young people (Gu et al., 2023; Weybright et al., 2020). Future studies can seek to unpack the intervention to identify components that are most conducive to the observed improvements while recognizing the potential between-participant heterogeneity in the type of support needed.

Limitations

The current study has several limitations. First, participants were not randomized into the two conditions. This was mostly due to the scheduling of the intervention; the participants who could attend the training sessions were assigned to the intervention group, whereas those who signed up late or could not attend the designated sessions were assigned to the control group. This quasi-experimental design has its inherent limitations, such as selection bias. Future studies should seek to conduct proper randomization of participants.

Second, the comparison group was a no-intervention control group, i.e., no attention training was provided to those in the control group after the study. This might have introduced demand characteristics, which can impede the validity of the results (Cuijpers et al., 2019). However, it has been noted in reviews of psychotherapy research that no-intervention control generally introduces less bias than waitlist control, i.e., the nocebo effect (e.g., Furukawa et al., 2014). That is, effect sizes based on the comparisons between treatment and waitlist control are generally larger than other passive control groups, including no-treatment. Nonetheless, future studies should consider including an active control group to better assess the efficacy of attention training and to reduce the influence of demand characteristics on the between-group comparisons.

Third, the EMA was conducted using a smartphone app that only offers up to five points on a Likert scale. The magnitude of change in attention and boredom experienced throughout the four weeks could be limited due to a ceiling or floor effect. Future studies can consider a more fine-grained scale rating to improve the sensitivity to fluctuation in attentional and boredom experiences.

Fourth, also related to the design of EMA, whilst answering too many questions at a time may risk burdening participants and affecting compliance, we had to minimize the number of questions asked on each occasion. Our key outcome variables were assessed with single-item questions, which may affect the reliability. Krueger and Schkade (2008) found reliabilities of 0.5–0.7 using a single momentary affect measure in their study, which was similar to the reliability of a general well-being measure, showing that single-item measures could be considered reliable in EMA research (Dockray et al., 2010; Kahneman et al., 2004; de Vries et al., 2021). Nevertheless, future studies may consider including more nuanced and sophisticated measures of attention level for a more precise understanding of the potential underlying mechanism of changes.

Fifth, responding to the EMA prompt itself could have been a distraction to the participants. This is particularly relevant given that the study was on sustained attention, and one of the intervention strategies was to reduce unnecessary smartphone use. This might also help explain why the control group saw a decrease in their attention in situations where they had nothing in particular to do, which, according to the BFM, is likely a situation when they had difficulty engaging with their attentional resources or low intention to attend. The EMA prompt might have become a distractor itself that further discounts the person’s attentional resource and exacerbates the sense of inadequate attentional engagement. However, as suggested by review studies (Degroote et al., 2020), EMA remains the most suitable method to collect data in real-life situations in real-time.

Lastly, owing to the non-controlled environment in which this study was conducted, we did not (and could not) account for environmental or individual factors that could influence the present findings. For example, the academic or work demands our participants faced could have fluctuated throughout the 28 days of the study. It would be of interest for future studies to consider factors such as examination periods of college students and annual peak seasons for employees on the changes in attention and boredom levels.

Conclusion

The present study developed and tested an attention training program for healthy adults. Momentary experiences of attention and boredom were assessed across 28 days among participants in the intervention and control groups. The training was found to be effective in improving the level of attention in everyday situations, particularly when participants were working, and prevented a drop in attention when participants found nothing to do. The increased level of attention over time was mirrored by a gradual decrease in boredom experience. Such an effect is similar across types of activities. Successfully coping with the experience of boredom, which signifies attention lapse, may afford us the opportunity to redirect our attention or readjust our environment to strive to overcome challenges and pursue meaning in life. This study lends support to the notion that one’s boredom experiences could be mitigated by an increased ability to engage one’s attention in the activity at hand.