The main aim of the thesis was to investigate the effect of the use of a foreign language - in this case, English - on the strength of the framing effect in framing types that have not been studied in this respect before. According to the results, framing through the manipulation of emotionally charged keywords is indeed mitigated by the use of a foreign language. The main factors at play here are whether the respondent started learning the foreign language at an early age before the emotion-regulation systems were developed, and whether the respondent is accustomed to using the language routinely. Visual framing yielded the opposite results - in the absence of the keywords, only the effect of greater cognitive load remained for respondents completing the questionnaire in English, which magnified the framing bias. This effect was stronger for respondents whose English skills were weaker and who used English less. In the case of framing using the status quo, the results were inconclusive as there was a bias most likely caused by an inaccuracy in translation.

In 1984, Miller published the paper: Bias among forensic document examiners: A need for procedural changes, with the intent to elicit some concern about the amount of cognitive bias among forensic document examiners. There is a need for the development of procedures regarding how a document examiner can minimize the amount of cognitive bias that may lead to erroneous conclusions by the examiner. Such procedures would serve to demonstrate that a conscientious effort was made by the examiner and the submitting agency to control extraneous variables that could bias the results of the examination. Some 28 years after Miller1 the forensic sciences are confronted with serious criticism with respect to cognitive bias (e.g. Risinger et al.2, and the NAS report3). It appears that not much of Millers suggestions have been applied in practice. No good general procedures have been implemented for minimizing the risk of cognitive bias in most institutes. In this paper we address the main issues raised in the 1984 paper, and describe the current state of affairs with respect to minimizing cognitive bias in the forensic sciences. There is still a need for procedural changes in the forensic sciences.

No description available.

No description available.

No description available.

No description available.

No description available.

Heavy/at-risk alcohol use among young adults is an important public health problem given its association with health risk behavior and negative consequences. Dual process models of self-control characterize heavy/at-risk alcohol use as a function of (1) automatic appetitive processes and (2) conscious control processes. While existing interventions typically focus on individuals’ conscious control processes, recent research has focused on targeting automatic appetitive processes directly through strategies such as Cognitive Bias Modification (CBM). Although there is some evidence that CBM may improve outcomes for patients in treatment for alcohol use disorder, there is less evidence for its efficacy among young adults. This has been attributed to utilizing CBM as a stand-alone intervention and recruiting samples with low motivation to change drinking in prior studies. To address these limitations, this study examined the efficacy of CBM as an adjunct to brief web-based motivational intervention among young adults interested in changing their drinking. Heavy drinkers, ages 18-34, were recruited through social media platforms and completed an online screener. Eligible participants (N = 81) completed baseline assessment and a web-based motivational intervention (NIAAA- Rethinking Drinking). Participants were randomized to one of two adjunct intervention conditions—CBM or SHAM (control)—that was completed following the motivational intervention and over 4 sessions in the ensuing week. Weekly number of drinks, weekend drinking, alcohol-related consequences, readiness to change, and implicit biases toward alcohol were measured at baseline, 1-week and 1-month follow-up. Intervention condition did not significantly predict the primary outcomes of weekly drinking or consequences at 1-month, controlling for sex. In addition, intervention condition was not associated with implicit bias toward alcohol at 1-week or 1-month follow-up, controlling for sex. Post-hoc analyses revealed the intervention may be differentially efficacious for individuals high, versus low, in readiness to change. Overall, this web based CBM intervention did not reduce drinking behavior, consequences, or implicit biases compared to SHAM. Future work should focus on alternative strategies to modify automatic processes to enhance motivational web-based interventions, particularly among individuals highly motivated to change their behavior.

When entrepreneurs are searching for more information to develop their business ideas, they generally do not have enough time to do extensive research and thus are inclined to take shortcuts. This could undermine the rationality of their decisions, depending on what information is not found. However, one popular shortcut offers an interesting opportunity. By consulting experts, entrepreneurs can save time and, if they apply the search principles identified in this study, they can further avoid search obstacles, such as cognitive biases and poor decision framing. For example, by consulting skeptics of a technology, entrepreneurs can counteract their own optimism. While the right mitigation techniques for bias may seem obvious once the bias has been identified, acting pro-actively is not obvious and seems to require experience or prior learning. The results of this study were obtained by applying grounded theory on data obtained from semi-structured interviews. Nine interviews were done in Stockholm, Sweden. The interviewees included founders of companies, a business developer and a business coach for entrepreneurs.

This paper investigates the conversational capabilities of voice-controlled virtual assistants with respect to biased questions and answers. Three commercial virtual assistants (Google Assistant, Alexa and Siri) are tested for the presence of three cognitive biases (wording, framing and confirmation) in the answers given. The results show that all assistants are susceptible to wording and framing biases to varying degrees, and have limited ability to recognise questions designed to induce cognitive biases. The paper describes the different response strategies available to voice user interfaces, the differences between them, and discusses the role of strategy in relation to biased content.