JUDGEMENT AND DECISION MAKING
Palak Agrawal
Article 1 – Mindless math
There is evidence that people perform mathematical operations in response to problems even if they don’t need to, which we call “mindless math”. Three pre-registered studies (total N = 3,193) investigated how mindless math related to perceived problem difficulty, problem representation, and accuracy. According to Study 1, more numeric demands lead to more mindless math (and fewer correct answers). According to study 2, this effect isn’t caused by people being wary of easy problems. Throughout Study 3, we demonstrate that this effect is robust over a wide range of numeric demands, and we discuss two possible mechanisms that could explain this effect. However, there is a caveat that at higher levels of numeric demands, the effect may reverse, resulting in higher accuracy compared to moderately hard math at much harder levels. (Lawson, M. Asher et al., 2022)
Article 2 – Search for value-driven information in partner selection.
Generally, people use non-compensatory choice heuristics to find a relationship partner based on incomplete information. In recent experimental studies, however, partners are typically chosen using compensatory strategies. In order to bridge the gap between theory and experimental evidence, we separate the information search process from the evaluation process in order to characterize the mate choice problem. According to an eye-tracking experiment and a MouseLab experiment, people show strong value-directed search heuristics in response to all types of cues, and the magnitude of value-directed searches increases with cue primacy. Valued-directed search is also explained by cue primacy as a result of the interaction between cue type and participant sex. Further, value-directed search does not necessarily result in non-compensatory choice rules, but may serve compensatory decision-making rather than non-compensatory choices. Our results demonstrate that people may adopt remarkably smart search heuristics to find an ideal partner efficiently. (Wang, Hongyi et al, 2022)
Article 3 – Judgements with description and experience.
When modest probabilities are sampled or presented, decision-makers give varying weights to each. We distinguish between the information that decision makers acquire (population probabilities or sample frequencies), our innovative explanation, and how they receive it (via description or experience), the literature’s dominant focus. The latter establishes statistical confidence, or how confidently one can say that an option is better than expected. Sample judgements to react to statistical confidence are revealed by two laboratory investigations and a review of earlier work. In fact, more strongly than decisions based on population probability, which results in larger expected payoffs. As a result, our research not only provides a more reliable way for locating description-experience gaps. Additionally, it illustrates how probability weighting. (Klingebiel, Ronald et al, 2022)
Article 4 – Stress and Risk – Preferences versus noise
In a pre-registered laboratory experiment with 194 participants, we examine the effect of acute stress on risky decision. By isolating decision-making noise from a real shift in preferences, we examine the causal effect of stress on the stability of risk preferences. Using descriptive analyses, structural estimations for risk aversion, and various noise structures, we find no appreciable variations in risk attitudes across situations on the whole. We also find statistically significant evidence that poorer cognitive capacities are associated with more noise in decision-making generally, which is consistent with the prior literature. There is no discernible relationship between cognitive function and stress that affects noise levels. (Parslow, Elle et al, 2022)
Article 5 – Explain human sampling rates in various decision domains.
The optimal stopping literature frequently contains under sampling biases, particularly for economic full choice problems. The sampling rate of participants in these types of number-based studies appears to be influenced by the moments of the distribution of values that creates the alternatives, or the generating distribution. On a different form of optimal stopping task, where participants selected potential love partners from pictures of faces, a recent study, however, found an oversampling bias (Furl et al., 2019). The authors postulated that this bias may be unique to mate selection. We tested whether sampling rates across several image-based decision-making domains a) reflect various over- or under sampling biases, or b) depend on the moments of the generating distributions after preregistering this hypothesis (as shown for economic number-based tasks). We discovered evidence that the preregistered hypothesis was false in two experiments (N = 208 and N = 96). Participants oversampled to the same extent across domains (in comparison to a Bayesian ideal observer model), and their sampling rates were similar to number-based paradigms in that they were dependent on the mean and skewness of the generating distribution. Also, just like participants, the optimality model sampling was somewhat dependent on the skewness of the generating distribution. We come to the conclusion that sampling rates in image-based paradigms, like sampling rates in number-based paradigms, depend on the generating distribution and are not driven by the mate choice domain. (van de Wouw, Didrika S. et al, 2022)
Article 6 – Faith in karma is coupled with perceived (but not real) trustworthiness.
Karma devotees believe in the ethical causality theory, which holds that both good and bad consequences can be linked to previous moral and immoral deeds. The concept of karma may have significant interpersonal repercussions. We looked into whether American Christians who believe in karma have higher standards for their interaction partners and are more willing to trust them. We carried out a trust game incentive study with interaction partners who had various god and karma views. Participants had higher expectations of and were more likely to believe in karma beliefs. Expectations did not match action, and karmic belief did not predict real reliability. These results imply that people could mistakenly infer someone’s reliability based on their karmic belief. (Ong, How Hwee et al, 2022)
Article 7 – Pay or Not Pay: Measurement of risk preferences in the lab and field.
Using financial incentives to gauge risk preferences is expensive. It may also be risky and unjust in the pitch. The widely used Holt and Laury (2002) measure is time- and money-consuming because it depends on a dozen lottery options and fees. The differential rewards produced by good and bad luck raises moral questions as well. Tensions between the researcher and subjects may also arise if some subjects are paid but not all. We apply a condensed version of Holt and Laury in a pre-registered study in Honduras, Nigeria, and Spain where we address all three issues. We discover that there is no difference between paying with or without probabilistic rules in practise. Our simplified and hypothetical version makes (Brañas-Garza, Pablo. Et al, 2021)
Article 8 – Psychological Task Management: The trap of smaller tasks.
How do people choose which task to complete first when given a choice between several? Normally, the work with the best cost-benefit ratio (or efficiency) should be given priority. We propose that people continuously prioritise smaller (less time-consuming) tasks and stick with them even when this approach proves to be ineffective. We call this phenomenon the “smaller tasks trap.” We further expect that individual differences in the propensity for rational thought are adversely correlated with the preference for the smaller assignments. We created a unique paradigm using an incentive-compatible task management game where participants are given a variety of tasks and must choose how to complete them in order to test these predictions. The findings support the smaller task trap and show that variations in logical thinking among individuals predict vulnerability to this trap. That is, regardless of how effective the smaller tasks were, participants with low levels of logical thought preferred to begin with a smaller (vs. bigger) task. As a result, they performed much worse overall in the work management game. We go over the theoretical and practical ramifications of these findings and make some suggestions for potential treatments that could aid in task management improvement. (Rusou, Zohar. Et al, 2020)
Article 9 – Risk of experiment framing: A methodological note.
In traditional studies on judgement and decision-making under risk, participants are given a description of the potential outcomes and probabilities to help them understand the risk. A new paradigm, known as decisions-by-experience, has been developed recently in research, where participants learn about risk by taking samples from the results rather than reading summary summaries. According to the latter study, there is a description-experience gap, which shows that some of the traditional risk attitude patterns change when people really face the risk. The decisions-by-experience paradigm’s risky choice framing has been the subject of recent research. I go over the issues this research has with correctly changing the framing of experience-based decisions. I contend that framing effects exist in experience tasks as well, drawing on studies on framing with animals. The classic Asian Disease task, however, awaits proper translation into an experience paradigm. (Kühberger, Anton, 2021)
Article 10 – To vote under pressure.
In a well-designed lab experiment, we test the hypothesis that time pressure affects voting choices, specifically the amount of strategic (false) voting. When adopting the popular Plurality Voting method, we discover that participants are more honest while under time pressure. In other words, a time crunch could make voting less strategic and lead to an inaccurate depiction of preferences. Yet, Approval Voting has no effects, which is consistent with claims that this procedure doesn’t offer any incentives for strategic voting. (Alós-Ferrer. et al, 2022)
Article 11 – Conclusion
According to Study 1, more numeric demands lead to more mindless math (and fewer correct answers). According to an eye-tracking experiment and a MouseLab experiment, people show strong value-directed search heuristics in response to all types of cues, and the magnitude of value-directed searches increases with cue primacy. Our results demonstrate that people may adopt remarkably smart search heuristics to find an ideal partner efficiently. The latter establishes statistical confidence, or how confidently one can say that an option is better than expected. We also find statistically significant evidence that poorer cognitive capacities are associated with more noise in decision-making generally, which is consistent with the prior literature. The authors postulated that this bias may be unique to mate selection. We discovered evidence that the preregistered hypothesis was false in two experiments (N = 208 and N = 96). The concept of karma may have significant interpersonal repercussions. We carried out a trust game incentive study with interaction partners who had various god and karma views. These results imply that people could mistakenly infer someone’s reliability based on their karmic belief. We apply a condensed version of Holt and Laury in a pre-registered study in Honduras, Nigeria, and Spain where we address all three issues. We discover that there is no difference between paying with or without probabilistic rules in practise. How do people choose which task to complete first when given a choice between several? Normally, the work with the best cost-benefit ratio (or efficiency) should be given priority. We propose that people continuously prioritise smaller (less time-consuming) tasks and stick with them even when this approach proves to be ineffective. The decisions-by-experience paradigm’s risky choice framing has been the subject of recent research. I go over the issues this research has with correctly changing the framing of experience-based decisions. In a well-designed lab experiment, we test the hypothesis that time pressure affects voting choices, specifically the amount of strategic (false) voting. When adopting the popular Plurality Voting method, we discover that participants are more honest while under time pressure.
REFERENCES
Alós-Ferrer, Carlos & Garagnani, Michele, 2022. “Voting under time pressure,” Judgment and Decision Making, vol. 17(5), pages 1072-1093, September 2022, Cambridge University Press.
Brañas-Garza, Pablo & Estepa-Mohedano, Lorenzo & Jorrat, Diego & Orozco, Victor & Rascón-Ramírez, Ericka, 2021. “To pay or not to pay: Measuring risk preferences in lab and field,” Judgment and Decision Making, vol. 16(5), pages 1290-1313, September 2021, Cambridge University Press.
Klingebiel, Ronald & Zhu, Feibai, 2022. “Sample decisions with description and experience,” Judgment and Decision Making, vol. 17(5), pages 1146-1175, September 2022, Cambridge University Press.
Kühberger, Anton, 2021. “Risky choice framing by experience: A methodological note,” Judgment and Decision Making, vol. 16(5), pages 1314-1323, September 2021, Cambridge University Press.
Lawson, M. Asher, Larrick, Richard P. & Soll, Jack B., 2022. “When and why people perform mindless math,” Judgment and Decision Making, vol. 17(6), pages 1208-1228, November 2022, Cambridge University Press.
Ong, How Hwee & Evans, Anthony M. & Nelissen, Rob M. A. & van Beest, Ilja, 2022. “Belief in karma is associated with perceived (but not actual) trustworthiness,” Judgment and Decision Making, vol. 17(2), pages 362-377, March 2022, Cambridge University Press.
Parslow, Elle & Rose, Julia, 2022. “Stress and risk — Preferences versus noise,” Judgment and Decision Making, vol. 17(4), pages 883-936, July 2022, Cambridge University Press.
Rusou, Zohar & Amar, Moty & Ayal, Shahar, 2020. “The psychology of task management: The smaller tasks trap,” Judgment and Decision Making,vol. 15(4), pages 586-599, July 2020, Cambridge University Press.
van de Wouw, Didrika S. & McKay, Ryan T. & Averbeck, Bruno B. & Furl, Nicholas, 2022. “Explaining human sampling rates across different decision domains,” Judgment and Decision Making, vol. 17(3), pages 487-512, May 2022, Cambridge University Press.
Wang, Hongyi & Ma, Jiaxin & He, Lisheng, 2022. “Value-directed information search in partner choice,” Judgment and Decision Making, vol. 17(6), pages 1287-1312, November 2022, Cambridge University Press.