Critical Review of Perceptuo-Motor, Cognitive, and Description-Based Decision-Making

Summary

Jarvstad, Hahn, Rushton, & Warren (2013) found that contrary to classical studies (Kahneman & Tversky, 1979; Hertwig, Barron, Weber, & Erev, 2004; Weber, Shafir, & Blais, 2004) people are equally good at making low-level and high-level cognitive decisions. This means making decision based on experience and described information respectively. The study by Jarvstad et al. (2013) showed that description-experience gap (when people make choices based on experience, they take fewer probable events too lightly) and perception-cognition gap (when people make high-level cognitive decision they weight the unlikely events more than they should) are illusory. Additionally, Jarvstad’s et al. (2013) paper suggests that individual differences between people who make choices are more vital than distinctions between each domain when predicting their decisions. The study used within-subject design, repeated measures on three tasks such as, perceptuo-motor, cognitive and description. There were 18 participants who were paid £6 per hour and an extra financial reward if they performed well. They had to attend a trial on the first day and come the day after for an initial decision session. The critical review will evaluate the methodology, how the conclusion about both, description-experience and perception-cognition gaps were made as well as include possible biases which could affect the outcome of the study and its validity. It will also analyse for the findings of this paper.

Review

The description of the course of the experiment is important to understand the strengths and weaknesses of this paper. The participants of the study first had to go through a learning session to prepare themselves for a good performance in the actual experimental tasks. In a perceptuo-motor task, they had to decide which bar is more likely to give them points and hit it with their finger on the computer screen. After the task they could see if they missed the bar which resulted in a loss of 700 points whereas, a correct hit would give them 100 points. The point system in this session was to prepare the subjects to score well in the decision-making session. Empirical error distribution, based on perceptuo-motor and mental arithmetic trials, was measured to check subjects’ abilities across these domains. It also designed the actual decision session tasks based on both domains from a trial session to match a classical task. Participants were told that they should learn as much as possible from the experience in the learning session as it will improve their performance and affect their final earnings (Jarvstad et al., 2013). This researchers’ suggestion may be related to a framing effect. It is people’s avoidance of taking risks when they are put into a situation where they can benefit from something (Tversky & Kahneman, 1981). During the decision period the participants had to pick between two probabilities that had assigned rewards and choose the most beneficial option. They had to make these choices across three different domains. After that, participants’ real decisions were compared with the best hypothetical option which had the highest expected value. It analysed if there are any differences between their between-tasks performance. They showed that most of the subjects individually performed equally well across all domains. In addition, the participants’ results were compared separately for each task to see if they performed equally well. A better predictor of people’s choices are differences between the decision-makers rather than between each domain. The results showed that there were big individual differences between participants, but there were not many differences with the between-tasks performance. In the past studies, participants who took part in similar experiments had to choose between two financial decisions only few times (Hau, Pleskac, Kiefer, & Hertwig 2008; Hertwig et al., 2004) Therefore, they had fewer opportunities to learn from the experience which may have led to the presence of a wide description experience gap (Hau et al., 2008) Would that be the reason for this gap reduction in this study?

Get Help With Your Essay

If you need assistance with writing your essay, our professional essay writing service is here to help!
Find out more about our Essay Writing Service

The methodology used in Jarvstad et al. (2013) experiment will be analysed to establish its strengths and weaknesses. A within-subject design made the study thorough. It is because, the results show that there are big individual differences between participants. If it was a between-subjects design the results would have been inaccurate, as the comparison of two people’s results do not illustrate one’s ability to make optimal decisions in both; experience based and analytical decisions. For example, one person may perform well on a decision from experience task and the other poorly on a decision from description task. That does not describe how someone performs overall across tasks. It could explain opposite findings to this study (Hertwig et al., 2004). Moreover, participants were not told about amount of their financial reward until the end of the experiment. It is because, when people find out that they failed, they are more likely to learn from that incident and perform better later (Santos & Rosati, 2015). This could bias the results of these subjects in all tasks. Furthermore, the study was conducted on only 18 participants. It may be possible that if the experiment was replicated with more participants, as it has been done previously with the similar experiments (Hertwig et al., 2004; Barkan, Zohar, & Erev, 1998), the results of the study would show greater variability with people’s individual performance across all domains.  One reason for this could be that it is not clear whether some participants may have had a bad experience of gambling in the past. These kinds of events tend to alter future behaviours in a way that one tries to avoid making these mistakes again (Kahneman & Tversky, 1979; Hertwig et al. 2004; Wu, Delgado, & Maloney, 2009). This may have caused some participants’ results to be above average. Moreover, Huffman (1974) proposed that education improves one’s ability to decode, analyse and classify financial and technical data. It is not known what educational background participants had. However, it is certain that these participants were university students who were not studying any economic nor mathematical major. Some of them may have poor analytical skills to perform well on these tasks. This may have affected participants’ understanding of probabilities and ability to calculate them to make the most optimal decisions. Therefore, it may explain why the results between participants varied greatly.

The explanation for different findings is partially based on lack of perception-cognition and description-experience gaps. It is of big importance to evaluate the validity of these findings. The absence of perception-cognition gap in this study arises from the different ways in which performance was measured. When it was assessed by standard low-level metrics, the performance was more accurate whereas when high-level metrics were used (calculation based on cumulative prospect theory) the performance was lower for both high and low-level decisions. Description-experience gap in this study disappeared. However, the method used to lead to this conclusion may not be valid. The replication of a key qualitative discrepancies showed that it fitted the objective probabilities, but it did not fit the subjective probabilities. It indicated an assumption that both probabilities should overlap themselves is incorrect. However, there are different methods of measuring the subjective probability. It may be possible that when the probability will be measured differently, the results would be different (Jarvstad et al. 2013). A replication of the experiment is advised to make sure that the results are not biased by the previous participants’ experiences which may have affected their judgement and the methodology used to calculate subjective probability.

This experiment contradicts previous papers including one of the most cited research papers in economics (Kahneman & Tversky, 1979). Jarvstad et al. (2013) study has been cited only 38 times whereas Kahneman & Tversky (1979) research over 53000 times. It does not make this paper less influential. It is because some of the studies used Jarvstad et al. (2013) methodology or findings to prove points in their own research papers (Meyniel, Sigman, & Mainen, 2015; Oaksford & Hall, 2016; Dekker & Nardini, 2015) Also none of the researchers who cited this paper criticised it, which may suggest that this study was thorough and should be replicated to examine further details of such contradicting findings. However, nowadays research studies are not replicated due to concerning falsified results, outcomes of the studies by chance where small samples are used, different behaviour over time leading to various or opposite results (Diener & Diener, 2017) Could that be a barrier for the funders to pay for such research replication?

Conclusion

This paper was well thought as it analysed previous research studies to reach to the decision of the experiment replication with a different methodology. Also, it used many strategies to avoid confounds and biases. For instance, it considered framing effect, amount of repeated measures to make sure that the study is not under sampled and used within-subject design to avoid confounds. On the other hand, it did not check participants’ past experiences, which could affect their results across domains and their mathematical skills. As opposed to previous papers which studied decision-making, this experiment found that people are equally good at making choices based on experience and described information. The outcome of this paper was not influential, as it was cited only 38 times and the other paper over 53000 times, the field of psychology nor economy as opposed to Kahneman, & Tversky (1979) study. It is hypothesised that the reason for such opposite findings may be an improved different methodology used.  The study needs to be replicated to make sure that methodology does not falsify the results or make them unclear. It is hoped that it happens although there is a trend that nowadays many studies are not being replicated.

References

Barkan, R., Zohar, D., & Erev, I. (1998). Accidents and decision making under uncertainty: A comparison of four models. Organizational Behaviour and Human Decision Processes, 74, 118–144.

Dekker, T. M., Nardini, M. (2016). Risky visuomotor choices during rapid reaching in childhood. Developmental Science, 19(3), 427-439.

Diener, R., & Diener, E. (2017). The replication crisis in psychology. Champaign, IL: DEF Publishers.

Hau, R., Pleskac, T. J., Kiefer, J & Hertwig, R. (2008). The Description–Experience Gap in Risky Choice: The Role of Sample Size and Experienced. Journal of Behavioral Decision- Making. 21, 493–518.

Hertwig, R., Barron, G., Weber, E. U., & Erev, I. (2004). Decisions from experience and the effect of rare events in risky choice. Psychological Science, 15(8), 534-539.

Huffman, W. E. (1974). Decision Making: The Role of Education.  American Journal of Agricultural Economics. 56(1), 85-97.

Jarvstad, A., Hahn, U., Rushton, S.K., & Warren, P.A. (2013). Perceptuo-motor, cognitive, and description-based decision-making seem equally good. Proc. Natl. Acad. Sci. 16271-16276.

Kahneman, D., Tversky, A. (1979). Handbook of the Fundamentals of Financial Decision Making. World Scientific.

Meyniel, F., Sigman, M., Mainen, Z. F. (2015). Confidence as Bayesian Probability: From Neural Origins to Behavior. Neuron, 88(1), 78-92.

Oaksford, M., Hall, S. (2016). On the Source of Human Irrationality. Trends Cogn Sci, 20(5), 336-344.

Santos, L. R., & Rosati, A. G. (2015). The Evolutionary Roots of Human Decision Making. Annu Rev Psychol. 66, 321–347.

Tversky, A., Kahneman, D. (1981) The framing of decisions and the psychology of choice. Science. 211, 453–58.

Weber, E. U., Shafir, S., & Blais, A. R. (2004). Predicting risk sensitivity in humans and lower animals: Risk as variance or coefficient of variation. Psychological Review, 111, 430–445.

Wu, S. W., Delgado, M. R., & Maloney, L. T. (2009) Economic decision-making compared with an equivalent motor task. Proc Natl Acad Sci, 106(15), 6088-6093.