Bored by bothering? A cost-value approach to pandemic boredom

At the time of data collection (December 2020), widespread non-pharmaceutical COVID-19 containment measures required high individual compliance to be effective. Prior work suggested boredom proneness undermines adherence, but it was unclear whether adherence itself elicits state boredom. The study’s primary research question was whether complying with behavioral recommendations (avoiding travel and gatherings) increases experienced or forecasted pandemic boredom, and how perceived value and effort of these behaviors contribute directly and indirectly (via compliance) to boredom. The authors hypothesized that lower perceived value and higher perceived effort would directly increase boredom, while simultaneously decreasing compliance and thereby indirectly reducing boredom, yielding opposing direct and indirect effects. They tested the model retrospectively (Thanksgiving) and prospectively (upcoming winter holidays), controlling for boredom proneness and examining robustness across US states with vs. without official restrictions.

Boredom is conceptualized as an aversive state marked by wanting but failing to engage meaningfully (Eastwood et al., 2012; Danckert et al., 2018). The MAC model (Westgate & Wilson, 2018) and Control-Value Theory (Pekrun, 2006) implicate perceived meaning/value and cognitive effort in boredom. Boredom arises when activities lack meaning or when engagement requires high mental effort (under- or overstimulation). Functional accounts position boredom as a signal to seek more rewarding alternatives; persisting in a low-value activity then requires self-control and feels effortful (Shenhav et al., 2017; Bieleke & Wolff, 2021). During the pandemic, monotony and constrained alternatives likely reduced perceived value and increased effort, potentially elevating boredom—especially during holidays when social restrictions conflict with traditions. Prior discourse noted “pandemic fatigue,” and research linked boredom proneness to lower adherence. The present study applies a cost-value framework: value and effort should shape both compliance and boredom.

Design and setting: Cross-sectional online survey administered December 18–19, 2020 via Qualtrics to US MTurk workers (≥50 HITs, ≥90% approval, age ≥21). Both retrospective (Thanksgiving) and prospective (upcoming winter holidays) assessments were included, enabling model testing at two time points. A multigroup analysis contrasted US states with vs. without official restrictions. Participants: N = 1,553 after excluding 13 who failed an instructional manipulation check. Mean age 40.36 (SD = 12.65); 47.3% female, 51.7% male, 1% other; 86.09% reported ≥13 years of education; 56.15% full-time employed, 13.46% self-employed. All US states represented. Ethics approval obtained; informed consent collected. Measures (5-point Likert unless noted):

• Compliance (Avoid travel/gathering): Two items each for Thanksgiving (retrospective) and winter holidays (prospective) adapted from Harris Poll, e.g., “I avoided traveling.”
• Perceived value: Multiple items for Thanksgiving and winter holidays assessing personal worth and societal worth (e.g., “avoiding travel was worthwhile for slowing the spread of COVID-19”).
• Perceived effort: Items assessing how effortful avoiding travel/gathering was (or would be) for Thanksgiving/winter holidays.
• Pandemic boredom: Three adjectives (bored, dull, monotonous) from the EES for Thanksgiving (experienced) and winter holidays (forecasted).
• Affect: Short PANAS (5 positive, 5 negative adjectives) for both periods.
• Subjective restrictions: Four items on perceived regional permissions to travel/gather (reverse coded so higher = more restrictions).
• Boredom proneness: Short Boredom Proneness Scale (8 items; Struk et al., 2017). Mean scores computed. Objective restrictions classification: States coded as with restrictions vs. without based on public sources (Gale 2020a/b) around the holidays. Procedure: After consent and manipulation check, participants completed retrospective Thanksgiving blocks (behavior, value, effort, emotions), then prospective winter holidays blocks, then personality scales and demographics. Three self-control single items were collected but not analyzed. Data preparation and reliability: For each period, items were aggregated across travel and gathering to form composite variables for avoid, value, effort, boredom, positive affect, negative affect, and subjective restrictions. Internal consistencies were good to excellent (e.g., Thanksgiving αs: Avoid .779, Effort .777, Value .935, Boredom .862; Winter holidays αs: Avoid .843, Effort .864, Value .950, Boredom .901). Statistical analysis: Structural equation modeling (SEM) with lavaan in R assessed direct effects of value and effort on boredom and indirect effects via avoid (mediator). Models controlled for boredom proneness. Separate SEMs were fit for Thanksgiving (retrospective) and winter holidays (prospective), with multigroup analyses by state restrictions. Model fit indices (RMSEA, SRMR, CFI, TLI), R2 for mediator and outcome, and bias-corrected bootstrap SEs (1,000 reps) were reported. Sensitivity analyses adjusted for age and gender.

Descriptive statistics:

• Compliance (avoid) high: Thanksgiving M = 4.19/5; Winter holidays M = 4.04/5.
• Value high: Thanksgiving M = 4.08; Winter holidays M = 4.00.
• Effort moderate: Thanksgiving M = 2.70; Winter holidays M = 2.90.
• Pandemic boredom low: Thanksgiving M = 1.95; Winter holidays M = 1.88.
• Boredom correlated strongly with negative affect (Thanksgiving r = 0.428; Winter r = 0.482) and negatively with positive affect (Thanksgiving r = −0.279; Winter r = −0.288).
• Objective vs. subjective restrictions correlations: Thanksgiving r = 0.178; Winter r = 0.207.
• Value strongly correlated with avoid (Thanksgiving r = 0.742; Winter r = 0.795). Boredom proneness correlated with pandemic boredom (Thanksgiving r = 0.388; Winter r = 0.395) and with negative affect (Thanksgiving r = 0.331; Winter r = 0.344).

Thanksgiving SEM (retrospective):

• Direct effects on boredom: Value b = −0.071, SE = 0.031, p = .021; Effort b = 0.096, SE = 0.018, p < .0001; Avoid b = 0.219, SE = 0.028, p < .001; Boredom proneness b = 0.403, SE = 0.026, p < .001.
• Paths to avoid: Value b = 0.780, SE = 0.024, p < .001; Effort b = −0.025, SE = 0.014, p = .088 (ns).
• Indirect effects on boredom via avoid: Value b = 0.171, SE = 0.023, p < .001; Effort b = −0.005, SE = 0.003, p = .104 (ns).
• Total effects on boredom: Value b = 0.100, SE = 0.020, p < .001; Effort b = 0.091, SE = 0.018, p < .001.
• Model fit: χ²(3) = 16.64, p = .001; RMSEA = 0.054 (90% CI [0.031, 0.081]); SRMR = 0.030; CFI = 0.992; TLI = 0.972. R2 Avoid = 55.2%; R2 Boredom = 20.5%.
• Adjusted for age/gender: Age had a small positive effect on avoid (b = 0.004, p = .004); no other changes.
• Multigroup (states with vs. without restrictions): Patterns similar; direct effect of value on boredom non-significant in states without restrictions (p = .386).

Winter holidays SEM (prospective):

• Direct effects on boredom: Value b = −0.093, SE = 0.034, p = .006; Effort b = 0.093, SE = 0.017, p < .0001; Avoid b = 0.256, SE = 0.032, p < .001; Boredom proneness b = 0.418, SE = 0.028, p < .001.
• Paths to avoid: Value b = 0.821, SE = 0.021, p < .001; Effort b = −0.051, SE = 0.013, p < .001.
• Indirect effects on boredom via avoid: Value b = 0.210, SE = 0.026, p < .001; Effort b = −0.013, SE = 0.004, p = .001.
• Total effects on boredom: Value b = 0.117, SE = 0.019, p < .001; Effort b = 0.080, SE = 0.018, p < .001.
• Model fit: χ²(3) = 20.124, p < .001; RMSEA = 0.061 (90% CI [0.037, 0.087]); SRMR = 0.028; CFI = 0.991; TLI = 0.971. R2 Avoid = 63.6%; R2 Boredom = 21.6%.
• Adjusted for age/gender: Both non-significant; results unchanged.
• Multigroup: Results similar; direct effect of value on boredom non-significant in states without restrictions (p = .195).

The study demonstrates that adherence to non-pharmaceutical COVID-19 measures, specifically avoiding travel and gatherings, is associated with higher pandemic boredom—supporting the notion that people can get “bored by bothering.” Perceived value and effort exert opposing direct and indirect influences: higher value directly reduces boredom but increases it indirectly by boosting compliance, whereas higher effort directly increases boredom but indirectly reduces it by lowering compliance. Net total effects were positive for both value and effort, indicating that overall, increased value contributes to boredom via greater adherence, and increased effort contributes to boredom primarily through its direct experiential cost. Boredom proneness robustly predicted pandemic boredom, exceeding the effect of behavior itself, highlighting vulnerable subgroups. Effects replicated across retrospective and prospective contexts and were largely invariant to objective state restrictions, suggesting broad generalizability. These findings underscore a trade-off for public health messaging: elevating perceived value of restrictions aids compliance but may heighten boredom unless coping strategies are provided; reducing perceived effort may both improve adherence and mitigate boredom.

This work integrates cost-value perspectives with pandemic behavior to elucidate mechanisms of pandemic boredom. It shows that value and effort predict both compliance and boredom, with compliance mediating part of these effects, and that boredom proneness is a key individual difference. Practically, interventions might combine value-focused messaging with strategies to manage boredom and reduce perceived effort (e.g., action planning) to support adherence without exacerbating negative affect. Future research should experimentally manipulate value and effort to establish causal mediation, employ longitudinal or experience-sampling designs to capture within-person dynamics and temporal precedence, and test generalizability across cultures and contexts.

• Cross-sectional mediation limits causal inference and temporal precedence; longitudinal/experimental designs are needed.
• Aggregated measures may not capture within-person variability; group-to-individual generalizability is limited.
• Retrospective and prospective self-reports are susceptible to memory and forecasting biases influenced by lay theories.
• US MTurk sample limits cultural generalizability; socio-cultural and resource disparities may moderate effects.