Comprehensive Analysis of "Cognitive Load and Economic Decision-Making: A Multi-Technique Approach"
Key Publication Information
Authors: Authors are accessible from the full paper on European Economic Review
Publisher: Elsevier, European Economic Review
Published Date: 2021
Overview
This study examines the impact of cognitive load on decision-making across multiple domains, applying four distinct cognitive load techniques: a number memorization task, visual pattern recognition, auditory recall, and time pressure. By comparing these methods within the same participant pool, the researchers assessed whether cognitive load consistently impacts behavior in economic tasks, including math problems, logic puzzles, lottery tasks (risk-related), and allocation decisions. They observed that all load techniques led to poorer performance on math and logic tasks and increased risk aversion in the lottery tasks. However, cognitive load had little to no effect on allocation decisions.
Key Results and Findings
1. Consistent Behavioral Patterns Across Load Techniques:
Across all four cognitive load techniques, participants generally exhibited similar behaviors:
Reduced performance on math and logic tasks: Cognitive load impaired participants’ ability to solve problems accurately.
Increased risk aversion: Participants under cognitive load showed a preference for safer options in lottery tasks.
Minimal effect on allocation decisions: Cognitive load had no systematic effect on choices made in allocation tasks, suggesting that certain types of decisions are less influenced by cognitive load.
2. Comparative Effect of Load Techniques:
Number Memorization and Auditory Recall: These techniques showed the strongest impacts on math performance, logic problem-solving, and risk aversion in lottery tasks.
Time Pressure: This technique had the most significant effect overall, likely because it added immediate pressure, heightening error rates across tasks.
Visual Pattern Task: This approach showed the least deviation from baseline but still displayed effects in line with other load techniques, reinforcing that visual load impacts cognitive processing.
3. Within-Subject Consistency:
Participants whose performance was most affected by one cognitive load technique also tended to be most affected by other techniques, indicating a robust, cross-task influence of cognitive load.
4. Role of Cognitive Reflection in Load Susceptibility:
Cognitive Reflection Test (CRT): The study found that participants scoring above the median on the CRT—indicative of higher reflectiveness and a tendency to resist impulsive responses—were more affected by cognitive load. In contrast, those below the median showed negligible changes under load, suggesting that cognitive reflection may buffer against or predict vulnerability to load-induced performance degradation.
Experimental Design and Methodology
Participants completed a series of economic tasks under each cognitive load condition. These tasks included:
Math Problems: Tested analytical and problem-solving abilities.
Lottery Tasks: Assessed risk tolerance by asking participants to choose between probabilistic outcomes.
Logic Puzzles: Required deductive reasoning under cognitive constraints.
Allocation Decisions: Measured how participants distributed resources, serving as a test for economic fairness and efficiency preferences.
Each cognitive load technique was applied individually to isolate its effects on task performance and behavior. A no-load baseline condition served as a control to measure deviation under each load type.
Analysis and Discussion
Behavioral Impact of Cognitive Load:
Cognitive load reduced participants' ability to perform well on math and logic tasks, aligning with prior research on the cognitive demands of such activities. This decline was particularly evident under time pressure, suggesting that immediate processing constraints exacerbate performance degradation.
The increase in risk aversion, consistent across load types, suggests that cognitive load prompts more conservative, lower-risk choices, potentially as a protective mechanism against the higher cognitive demands of evaluating probabilistic options.
Load Techniques and Economic Decision-Making:
The similarity in outcomes across load techniques supports theories like dual-process models (e.g., Kahneman’s System 1 and System 2) and divisive normalization, indicating that cognitive load reliably shifts behavior toward intuitive or less effortful choices.
The minimal effect on allocation decisions suggests that such tasks may rely less on working memory or might engage different cognitive processes that are less susceptible to immediate load manipulations.
CRT as a Predictor of Load Susceptibility:
High CRT scorers, typically more reflective, demonstrated greater behavioral shifts under load, showing that cognitive reflection might predict load susceptibility. This outcome may reflect that those with higher cognitive control are more impacted when that control is compromised.
The lack of significant change in behavior among low CRT scorers implies that individuals less inclined to analytical processing may be less affected by cognitive load in decision-making contexts, as they rely less on cognitive resources for these tasks.
Conclusion
The study confirms that cognitive load consistently impairs performance in tasks demanding higher cognitive resources (math, logic) and increases risk aversion. Notably, the consistency of effects across load techniques suggests that cognitive load has a predictable impact on economic decision-making. The finding that individuals with higher CRT scores are more vulnerable to cognitive load also highlights the importance of cognitive reflection in managing complex tasks under load, suggesting that those prone to deeper analysis may be more impacted when cognitive resources are constrained.
Implications for Research and Application
Broader Applicability of Load Techniques: Given the robustness of load effects, various cognitive load methods can be used interchangeably in economic research to study decision-making under constraints.
Predictive Role of CRT: CRT may serve as a valuable predictor in research on cognitive load, highlighting which individuals are likely to be more affected by cognitive constraints in decision-making environments.
Decision Task-Specific Effects: The resilience of allocation decisions to cognitive load suggests that not all economic choices are equally impacted, inviting further research into the types of decisions most vulnerable to cognitive load effects.
Key Points
📉 Performance Under Load: Across all techniques, cognitive load reduced participants' math and logic task performance.
🎲 Risk Aversion Increase: Cognitive load led participants to prefer safer options in lottery tasks, indicating heightened risk aversion.
🧠 CRT as Susceptibility Marker: High CRT scorers were more affected by cognitive load, while low CRT scorers showed minimal behavioral changes.
⏲️ Time Pressure’s Unique Impact: Among techniques, time pressure caused the most significant deviation from baseline, emphasizing the impact of immediate processing limits.
🔄 Within-Subject Consistency: Participants affected by one load technique were generally impacted similarly by other techniques, suggesting universal load impact.
👀 Minimal Allocation Impact: Allocation decisions were largely unaffected by load, hinting that these choices may engage different or less load-sensitive cognitive processes.
🧩 Methodological Consistency: The use of four load techniques showed consistent behavioral patterns, validating load manipulation reliability.
💡 Insights for Cognitive Models: Findings support dual-process and divisive normalization models, demonstrating that load consistently shifts behavior toward intuitive responses.
📊 Experimental Rigor: Including a no-load baseline provided clear contrasts for evaluating the impact of each load technique.
🌐 Future Research Potential: Study encourages further exploration of cognitive load’s role in different decision types, such as ethical or moral decision-making under constraints.
Summary
Impact Consistency: All load techniques produced similar effects, reducing cognitive performance and increasing risk aversion.
Significance of Time Pressure: This technique had the strongest impact on performance, likely due to immediate decision constraints.
Cognitive Reflection Predictability: CRT results identified those most vulnerable to load impacts, showing the test's predictive value.
Robustness Across Tasks: Similar responses across tasks validate cognitive load as a reliable method for manipulating economic decision behavior.
Mathematical and Logical Constraints: Cognitive load reduced problem-solving ability across math and logic tasks.
Risk Aversion Patterns: Risk aversion in lottery tasks consistently increased under load, reflecting a preference for lower-risk choices when cognitive resources are stretched.
Within-Subject Effect Consistency: Participants' responses were consistent across techniques, strengthening generalizability.
Dual-Process Model Support: Findings align with cognitive theories that highlight intuitive vs. reflective responses under load.
Minimal Allocation Impact: Allocation tasks remained stable across load conditions, possibly due to lower working memory reliance.
Future Implications: Results support the use of diverse load techniques in economic decision research, emphasizing the predictability of cognitive load's effects.
Abstract
There are many ways to induce cognitive load. In this paper, we manipulate cognitive capacity using four common techniques: a number memorization task, a visual pattern task, an auditory recall task, and time pressure. Under each load manipulation (as well as under ‘no load’), every participant completes a series of math problems, lottery tasks, logic puzzles, and allocation decisions. We find similar behavioral responses across all techniques: poorer performance on the math problems and logic puzzles, more risk ~aversion~in the lottery tasks, and no systematic impact on allocation decisions. Using within-subject variation, we show that individuals whose math performance is most impacted for a given load manipulation (number memorization), are the same individuals whose performance is most impacted by other load manipulation, and in the other tasks. We also find that participants who scored above the median in a cognitive reflection test (CRT), and are thus able to resist the first response that comes to mind, are greatly impacted when placed under cognitive load; those scoring below the median in the CRT are not impacted much.
Introduction
There is a fast growing literature examining how cognitive load impacts economic decision making.1 Recent studies have focused on experimentation as a means to manipulate cognitive load, and among these, multiple elicitation techniques have been tried. For example, Deck and Jahedi (2015) use a number memorization task, and find that people become more risk averse, more impatient, more susceptible to anchoring, and correctly answer fewer math problems when under cognitive load. Similar conclusions are reached by Hinson et al. (2003) regarding analytical performance and by Benjamin et al. (2013) regarding risk and impatience.2 In contrast, Gerhardt et al. (2016) use a visual pattern task to distort cognitive function, while Schulz et al. (2014) use an auditory recall task.3 Rand et al. (2012) use time pressure as another technique still, and find it leads to more contributions to a public good. They interpret this result as evidence that people are intuitively cooperative.4
If cognitive load were robust in its pathway to impact decisionmaking, then in principle, the various techniques used to induce load should lead to approximately similar behavioral responses and generalizable insights, as opposed to responses that differ based on the interaction of the exact task and load manipulation. Further one would expect that decision makers who are the most impacted on one task by one technique, would be the most impacted by being placed under load via other techniques when facing other tasks. Such a finding would give more weight to models such as dual process (Fudenberg and Levine, 2006); (Kahneman, 2011), divisive normalization Heeger (1992), drift diffusion Ratcliff (1978), or attention Bordalo et al. (2015), that help to explain the effects of cognitive load. Importantly, it would provide researchers with a broader toolkit by which to study cognitive load and some confidence that the findings are commonly attributable to the load manipulation. This paper addresses this void and studies the robustness of decisionmaking under various load techniques. We experimentally test the impact of four commonly used techniques for manipulating cognitive capacity: a number memorization task, a visual pattern task, an auditory recall task, and time pressure. Under each manipulation, including a ‘no load’ baseline (control), every participant completes a series of math problems, lottery tasks, logic puzzles, and allocation decisions. Overall, we find the four techniques for inducing cognitive load yield similar directional effects on behavior; however, the size of the effects vary by technique. The results indicate that number memorization and auditory recall have comparable sized effects with both leading to poorer performance on math problems and logic puzzles as well as more risk aversion in the lottery tasks, but no discernible consistent impact on allocation choices. Time pressure tends to have the largest effect, in part due to calibration, which increased the frequency of errors even for the baseline tasks. The visual pattern technique yields only modest difference from the baseline, but the directional patterns match those of the other techniques. Finally, we show that certain individuals are more vulnerable to cognitive load than others. Using within-subject variation, we show that individuals who are most impacted in their math performance by one technique, are the same individuals whose performance is impacted by the other load manipulation techniques for the remaining tasks. This suggests that the techniques used to generate cognitive load are operating in a similar way. We further show that reflectiveness, as measured by performance in a post-experimental “Cognitive Reflection Test” (CRT), greatly predicts which individuals are impacted by cognitive load. If we split the sample in half, those scoring in the bottom half of the CRT do not change their behavior relative to the no-load baseline, whereas those scoring in the top half show very large changes to their behavior. This finding supplements the literature showing that an individual’s cognitive characteristics can be an important component of how they tackle complex problems. For instance, Oechssler et al. (2008) shows that those with low CRT are more likely to show affective play in negotiations. Fehr and Huck (2016) show that those with low CRT scores take actions that are less strategic in contests. Indeed, performance on CRT itself can be a manifestation of how cognitive resources are being used for decisions. The remainder of the paper is organized as follows. The next section discusses the background literature. Separate sections then detail the experimental design and behavioral findings. The final section contains a concluding discussion.
Last updated