Research Methods Paper 3

Course:

Date:

HINDSIGHT BIAS

Introduction

Hindsight bias is the tendency to view an event as having been predictable after it has occurred even when there was little or no objective basis. It may affect different stages of processes, contexts, designs and situations. False theoretical outcomes may result from the recollection and reconstruction of content, causing memory distortion. Such a bias is common amongst physicians when recalling clinical trials, historians when describing the outcomes of wars and in judicial systems when investigating the predictability or responsibility of accidents. The purpose of this paper is to review three articles that describe hindsight bias and seek to investigate the hypothesis: memory distortion influences hindsight bias.

In Dunstin Calvillo’s research report on how rapid recollection of foresight judgements affect hindsight bias, suggests that memory distortion is a component of hindsight bias. A component that is measurable using a memory design. He indicates that a high working memory load corresponds with a smaller hindsight bias. He reports the findings of two experiments conducted to determine the memory distortion and its relation to hindsight bias. In the first experiment, it was found out that the magnitude of the bias was larger when less time was used to retrieve the original judgements. The second experiment showed that the bias was larger for participants who were required to respond quickly compared to the others who were required to respond more slowly. These findings are consistent with the claim that when the feedback is correct, there is an automatic update of knowledge especially when original judgements cannot be recalled. He suggests that the results are also in agreement with the claim that when a person fails to retrieve into working memory relevant information during the time of recollection, guessing then occurs depending on the information that is already in working memory (Calvillo, 2012).

In the study, he reports, rapid responding was a hindrance to information retrieval and, therefore, the large hindsight bias. Quick system processes, which are system one processing, are therefore responsible for the memory distortion component of hindsight bias and slower system processes, which are a system to processing, are necessary to reduce the bias. Measuring the ability to override system one outputs over in favour of the system two ones, may be used in the prediction of the components of hindsight bias. An example of such a measure would be the use of the Cognitive Reflect Test. The author indicates that the performance should have a positive correlation with the inevitability component and a negative one with the memory distortion component of the hindsight bias (Calvillo, 2012).

Hindsight bias has three components; these are an inevitability, foreseeability and memory distortion. Calvillo’s report indicates that system process has an effect of increasing foreseeability and inevitability but reduce memory distortion. The similarity between belief bias and hindsight bias becomes more apparent when the working memory load increases. Manipulation of response time in investigating the inevitability component shows a positive relationship between hindsight bias and response time (Calvillo, 2012).

In Nestler and Egloff’s report on the effect of surprise on hindsight bias, it is written that there are two hypotheses concerning the influence of surprising events on hindsight bias. One is the idea that surprising events tend to reverse hindsight bias while the other suggests that it increases it. The two perspectives can be reconciled with two of the three components of hindsight bias. When inevitability is considered, there is an increase in the bias while there is a reversal when foreseeability is considered. In their study, some participants were exposed to some information to a surprising event and others were not and then both made to go through the surprise. Among those who had received some information, there was found a strong hindsight bias regarding inevitability but a reversed one regarding foreseeability, as they perceived that the event was more inevitable and less foreseeable, than their counterparts who did not. Manipulation of the process that is presumed to underlie a hindsight bias determines the magnitude of the bias. Foreseeability is influenced by prior warning manipulation while inevitability is influenced mostly by explanation manipulation but remains unaffected by warning manipulation. They suggest that inevitability is dependent on the clarity of the explanation based on the causal event models. The results of their study, therefore, indicate that surprising events seem inevitable to the extent to which they are explainable. The type of assessed judgement matters in determining the effect of hindsight bias. Their study, therefore, suggests that the converse effects of surprise on this type of bias, are logically tied to the nature of the separate hindsight components. The conclusion of their study was, hindsight bias is not a unitary phenomenon but should be considered with the different and hindsight effects (Nestler and Egloff, 2009).

In another article, Roese and Vohs, while reviewing past research on hindsight bias, show many factors that make people so susceptible to this phenomenon. They have also identified a few ways of combating it. They propose three levels of this bias that stack on each other. They range from basic processes of memory, higher-level inference and even belief. Memory distortion is indicated as the first level of hindsight bias, involving forgetting or misremembering of an earlier option. The second level is centred on inevitability, believing that an event was bound to happen and so could not be evaded. The third level centres on foreseeability, involving the belief that an individual could have foreseen the outcome. There are certain factors that fuel our tendency to this bias. Ultimately, however, the bias negatively affect our learning experience in various situations. It could also create overconfidence in the approach to our judgements in different matters. It may also confer negative consequences in litigation such as in cases of medical malpractice, negligence or product liability. Roese and Vohs suggest that consideration of the opposite may help combat this cognitive fault (Roese and Vohs, 2012).

Reviewing these articles seems to confirm my hypothesis that memory distortion influences hindsight bias. A strong correlation between the magnitude of the bias and the time allowed for recollection is clearly presented.

REFERENCES

Calvillo, D, P. (2012). Rapid recollection of foresight judgements increases hindsight bias in a memory design. Journal of Experimental Psychology: American Psychological Association Learning, Memory, and Cognition 2013, Vol. 39, No. 3, 959–964.

Nestler, S. and Egloff, B. (2009). Increased or reversed? The effect of surprise on hindsight bias depends on the hindsight component. Journal of Experimental Psychology: American Psychological Association Learning, Memory, and Cognition 2009, Vol. 35, No. 6, 1539–1544...

Roese, N. and Vohs, K. (2012). “I Knew it All Along… Didn’t I?”- Understanding Hindsight Bias. Perspectives on Psychological Science.