A Meta-Analytic Review of the Association Between Mental Effort and Negative Affect

David, L., Vassena, E., & Bijleveld, E. (2024). The unpleasantness of thinking: A meta-analytic review of the association between mental effort and negative affect. Psychological Bulletin, 150(9), 1070–1093. https://doi.org/10.1037/bul0000443

Rationale

The purpose of this meta-analytic study was to examine two key questions:

1. Is mental effort generally experienced as aversive?
2. What sample and task characteristics moderate the experienced aversiveness of mental effort?

These questions address a fundamental controversy in the literature on mental effort and motivation. On one hand, influential theories in psychology, neuroscience, and economics assume that exerting mental effort should feel aversive (Shenhav et al., 2017; Holmstrom & Milgrom, 1994).

This assumption aligns with classic work on the “law of less work” (Hull, 1943) and modern research showing that people often rely on heuristics to minimize cognitive effort (Shah & Oppenheimer, 2008).

On the other hand, some research suggests that mental effort can feel pleasant rather than aversive. For example, studies on the need for cognition indicate individual differences in the tendency to seek out and enjoy mentally effortful activities (Cacioppo et al., 1996).

Additionally, research on learned industriousness suggests that effort can become a secondary reinforcer through repeated reward (Eisenberger, 1992).

This controversy, termed the “effort paradox” by Inzlicht et al. (2018), highlights the need for a systematic examination of when and for whom mental effort feels aversive.

By meta-analyzing a large set of studies using a common measure of effort and affect (the NASA-TLX), this study aimed to provide a comprehensive test of the assumed aversiveness of mental effort across diverse populations and tasks.

Method

The researchers conducted a rapid review and meta-analysis of studies using the NASA Task Load Index (NASA-TLX). They focused on this instrument because it captures both mental effort and negative affect, allowing for a systematic examination of their relationship across many studies.

Search strategy and terms:

• Database: Scopus
• Search term: ALL (“NASA-TLX” OR “NASA Task Load Index”)
• Date range: 2015-2020

Inclusion and exclusion criteria:

Inclusion:

• Peer-reviewed articles in English reporting original data
• Sample size > 10
• Healthy adult participants not under pharmacological or severe psychological treatment
• NASA-TLX administered directly after a single, discrete task described in the article
• Tasks requiring at least some cognitive effort
• Reported means and standard deviations of both effort and frustration items from NASA-TLX

Exclusion:

• Dissertations
• Studies using NASA-TLX to probe experiences over extended periods (e.g., full working day)
• Tasks consisting only of physical exercise

The researchers applied a preregistered stopping rule, coding articles in reverse chronological order until either all articles were processed or it became April 1, 2021. They processed the most recent 1,484 articles from their search.

Statistical measures:

The meta-analysis was conducted using the metafor package in R. The researchers computed the best linear unbiased predictor for effort and adopted it as the main predictor in a multilevel mixed-effects metaregression model, using the raw mean of frustration as the outcome measure.

To examine moderators, they added each moderator individually to the model, testing main effects and interactions with effort.

For categorical moderators, they estimated the effect of effort on frustration separately for each category. For continuous moderators, they estimated the effect for several representative values.

The researchers used a multilevel mixed-effects structure to account for dependencies in the data, with random effects for article, study, and task levels.

They also reported cluster robust tests and confidence intervals to further account for unknown dependency structures.

Results

Main effect of effort on negative affect:

The meta-analysis revealed a strong positive association between mental effort and negative affect (β = 0.85, SE = 0.06, 95% CI [0.73, 0.96], p < .001).

This effect was large: with each point increase in effort, negative affect increased by 0.85 points on average.

Moderator analyses:

1. Learning history moderators:

• Education: No significant interaction (β = 0.14, 95% CI [-0.33, 0.61], p = .515)
• Work experience: No significant interaction (β = 0.03, 95% CI [-0.11, 0.06], p = .362)
• Skill-task fit: No significant interaction (β = 0.01, 95% CI [-0.28, 0.31], p = .918)
• Continent: Significant interaction, with effort less strongly associated with negative affect in studies from Asia compared to Europe (β = -0.29, 95% CI [-0.55, -0.02], p = .046) and North America (β = -0.49, 95% CI [-0.83, -0.16], p = .006)

2. Task design moderators:

No significant interactions were found for task variety, monitoring feedback, performance feedback, control, task significance, or task identity (all p > .05)

3. Exploratory moderators:

No significant interactions were found for age, gender, task duration, physical activity, or group setting (all p > .05)

Robustness analysis:

The association between effort and negative affect remained strong (β = 0.88, SE = 0.06, 95% CI [0.76, 1.00], p < .001) even after controlling for 10 moderators with >90% valid data points.

Insight

The key finding of this meta-analysis is that mental effort is strongly associated with negative affect across a wide range of populations and tasks.

This association was robust to various moderators related to learning history, task design, and individual characteristics. The ubiquity of this relationship suggests that mental effort may be inherently aversive.

This study extends previous research by providing a comprehensive, quantitative synthesis of the link between mental effort and negative affect.

While prior work has often focused on behavioral choices or specific populations, this meta-analysis demonstrates the consistency of the effort-affect relationship across diverse contexts.

The findings support models in psychology, economics, and neuroscience that conceptualize effort as a cost. However, they also raise questions about why people sometimes voluntarily engage in effortful mental activities.

To address this apparent paradox, the authors propose an “integrative value account.”

This account suggests that high-effort activities may be pleasant overall due to associated rewards (e.g., monetary, social, mastery-related), even if the effort itself feels aversive.

Future research could further investigate:

1. The causal mechanisms linking effort and negative affect
2. How performance-contingent rewards might modulate the aversiveness of effort
3. The role of individual differences (e.g., need for cognition) in effort experiences
4. The subjective experience of effort in clinical populations with altered effort-based decision-making

Strengths

The study had many methodological strengths including:

1. Large sample size: 358 tasks from 170 independent samples, with data from 4,670 unique individuals.
2. Diverse populations and tasks: Included studies from 27 different countries and a wide range of cognitive tasks.
3. Preregistered analysis plan: The researchers preregistered their hypotheses, procedure, coding scheme, and analysis plan, enhancing transparency and reducing researcher degrees of freedom.
4. Robust statistical approach: Used multilevel mixed-effects meta-analysis to account for dependencies in the data and reported cluster robust tests and confidence intervals.
5. Comprehensive moderator analyses: Examined a wide range of potential moderators related to learning history, task design, and individual characteristics.
6. Use of a common measurement tool (NASA-TLX): Allowed for consistent comparison across studies.
7. Robustness checks: Conducted analyses to assess the impact of influential cases and potential response biases.
8. Transparent reporting: Provided detailed information about deviations from preregistration and made data and analysis scripts publicly available.

Limitations

1. Observational nature: As a meta-analysis, the study cannot establish causal relationships between effort and negative affect.
2. Potential selection bias: The study included only recent articles (2019-2020) and those written in English, which may limit generalizability.
3. Reliance on self-report measures: The NASA-TLX is a subjective measure, which may be influenced by response biases or differences in interpretation across cultures.
4. Limited assessment of affect: The NASA-TLX combines several types of negative affect into one dimension, potentially obscuring nuances in emotional experiences.
5. Lack of physiological or behavioral measures: The study relied solely on self-reported effort, which may not fully capture objective effort expenditure.
6. Potential publication bias: Although the authors argue that publication bias is unlikely to affect their results, it remains a possibility in any meta-analysis.
7. Limited data on some moderators: Some potential moderators (e.g., performance-contingent incentives) could not be analyzed due to insufficient data.

These limitations suggest caution in generalizing the findings to all contexts and highlight the need for further research using diverse methodologies and measures of effort and affect.

Implications

The findings have significant implications for several areas:

1. Theoretical models: The results provide strong support for models that conceptualize mental effort as a cost, informing theories of motivation, decision-making, and cognitive control.
2. Work and task design: Understanding the inherent aversiveness of effort can inform strategies for designing more engaging and less mentally taxing work environments.
3. Educational practices: The findings may influence approaches to structuring learning tasks and motivating students, considering the balance between effort demands and associated rewards.
4. Clinical psychology: The study highlights the need to examine subjective experiences of effort in clinical populations, potentially informing treatments for conditions involving altered effort-based decision-making (e.g., depression, schizophrenia).
5. Behavioral economics: The results may inform models of how people make decisions involving trade-offs between effort and reward.
6. Neuroscience research: The findings support the importance of investigating neural systems involved in effort valuation and the integration of effort costs with potential rewards.
7. Individual differences research: While the effort-affect association was robust across populations, the slight differences found (e.g., between continents) suggest the need for further investigation of cultural and individual factors that may modulate effort experiences.
8. Measurement practices: The study demonstrates the value of using standardized measures like the NASA-TLX across diverse research contexts, enabling large-scale synthesis of findings.

References

Primary reference

David, L., Vassena, E., & Bijleveld, E. (2024). The unpleasantness of thinking: A meta-analytic review of the association between mental effort and negative affect. Psychological Bulletin, 150(9), 1070–1093. https://doi.org/10.1037/bul0000443

Other references

Cacioppo, J. T., Petty, R. E., Feinstein, J. A., & Jarvis, W. B. G. (1996). Dispositional differences in cognitive motivation: The life and times of individuals varying in need for cognition. Psychological Bulletin, 119(2), 197–253.

Eisenberger, R. (1992). Learned industriousness. Psychological Review, 99(2), 248–267.

Holmstrom, B., & Milgrom, P. (1994). The firm as an incentive system. The American Economic Review, 84(4), 972–991.

Hull, C. L. (1943). Principles of behavior: An introduction to behavior theory. Appleton-Century.

Inzlicht, M., Shenhav, A., & Olivola, C. Y. (2018). The effort paradox: Effort is both costly and valued. Trends in Cognitive Sciences, 22(4), 337–349.

Shah, A. K., & Oppenheimer, D. M. (2008). Heuristics made easy: An effort-reduction framework. Psychological Bulletin, 134(2), 207–222.

Shenhav, A., Musslick, S., Lieder, F., Kool, W., Griffiths, T. L., Cohen, J. D., & Botvinick, M. M. (2017). Toward a rational and mechanistic account of mental effort. Annual Review of Neuroscience, 40(1), 99–124.

Keep Learning

Socratic questions for a college class to discuss this paper:

1. How might the finding that mental effort is generally aversive influence our understanding of human motivation and decision-making?
2. What are some potential explanations for the slightly weaker association between effort and negative affect found in studies from Asia? How might cultural factors influence the experience of mental effort?
3. How does the “integrative value account” proposed by the authors help reconcile the apparent paradox between the aversiveness of effort and people’s willingness to engage in effortful activities? What are some potential limitations of this account?
4. Given the limitations of self-report measures like the NASA-TLX, what other methods might researchers use to study the relationship between mental effort and affect?
5. How might the findings of this study inform the design of educational environments or workplace tasks to optimize engagement and performance?
6. What are some potential implications of these findings for understanding and treating clinical conditions that involve alterations in effort-based decision-making, such as depression or schizophrenia?
7. How might individual differences, such as need for cognition or growth mindset, moderate the relationship between mental effort and negative affect? How could future research investigate these potential moderators?
8. Given the strong association found between effort and negative affect, what factors might explain why some individuals seem to enjoy and seek out mentally demanding activities?