No description available.

Cognitive theories propose that interpretive biases play an important role in the onset and maintenance of emotional disorders. To investigate the proposed causal role of interpretive biases, this study examined if it is possible to train interpretations of ambiguous situations, and if this training affects emotional vulnerability and memory. The results indicated that the interpretive training was effective in inducing the intended group differences in interpretive bias, but that the positive training was more effective than negative training. These findings also highlight the potential benefits of inducing positive interpretive biases on mood and emotional vulnerability for some individuals (i.e., individuals with high depression or rumination scores). Additionally, results from the current study demonstrate that manipulating interpretive biases can result in corresponding changes in memory. In sum, findings from the current study hold valuable implications for the effects of positively manipulating interpretive biases on mood, emotional vulnerability, and memory. Future research can further elucidate the relationship between interpretive biases and mood, as well as extend current findings to a clinical population.

In the general population, psychosis-like experiences have been extensively studied under the psychometric rubric of schizotypy (psychosis-proneness). As such, Phase 1 of this thesis aimed to assess the distribution of schizotypal traits and associated personality correlates displayed within a quota sample of the general population stratified by Gender and Ageband, the emphasis being upon anomalous experiences (positive schizotypy). Respondents (N = 130) completed a battery of established self-report measures assessing thirteen areas of personal experience. Correlational analysis revealed that eleven of the measures of ostensibly anomalous experiences possessed significant intercorrelations. Subsequent principal components analysis identified three factors accounting for 64.91% of the total variance; the factor accounting for the greatest proportion of variance (42.97%) was interpreted as representing a psychological disposition towards reporting ‘Anomalous Cognitions’. The other two factors are named ‘Lifeview System’ (12.60% of total variance) and ‘Social Adaptation Skills’ (9.34% of total variance). From the principal factor inclusion criteria for Phase 2 of the research are explicated. No significant mean Gender differences were revealed for the six self-report measures that provided unique contributions toward anomalous cognitions, the two self-report measures that uniquely contributed toward a lifeview system, or for the single self-report measure that uniquely contributed toward social adaptation skills. Additionally, four of the nine self-report measures which provided unique factorial contributions generated significant mean differences between Agebands, with a further three providing trends toward significant mean differences. Implications for the role of anomalous cognitions, a framework for a lifeview system, and social adaptation skills with regard to psychosis-proneness are discussed. From the primary factor (anomalous cognitions), three experimental groups were identified for cognitive testing: respondents scoring 1) ≤ 20th percentile; 2) 10% ± the mean; and 3) ≥ 80th percentile. This procedure provided a total of 78 participants (three x 26) for Phase 2 testing. Phase 2 of this thesis sought to identify some of the cognitive mechanisms underpinning subclinical anomalous cognitions with a view to deciphering which measures best predicted experimental group membership. A comprehensive literature review highlighted six domains of cognition, five accompanying self-report measures, and two measures of intelligence functioning (verbal and fluid/visuoconstructive), which, following previous research, were utilised as covariate measures. Based on prescribed delineation points, participants were allocated, according to scores on the primary factor from Phase 1, to one of three experimental groups (low-, mid, and high-anomalous cognitions). Of the six cognitive domains—1) sustained visual attention; 2) false (illusory) memory; 3) probability reasoning (decision making); 4) object recognition; 5) reality monitoring; and 6) self-monitoring—four succeeded in eliciting significant mean differences between experimental groups with the noted exceptions of sustained visual attention and self-monitoring. Subsequent canonical discriminant analyses identified that the best predictors of XPG membership were the number of critical lures recognised on the false memory test, the number of correct responses and confidence when uncertain on the object recognition test, plus two self-report measures pertaining to comorbid psychopathology and the vividness of visual imagery. In light of previous research, the inclusion of fasle (illusory) memory biases, the comorbidity of mental pathology (especially, depressive and anxiety-related symptoms), and the vividness of visual imagery are unsurprising; however, the two object recognition variables (the ‘number of correct responses’ and ‘confidence when uncertain’) offer exciting avenues for future research into the continuum of psychosis. Moreover, the ‘confidence when uncertain’ data from the object recognition test (perceptual) and the probability reasoning (decision making) data from the Beads test suggest that cognitive underconfidence may well be an enduring personality disposition in those reporting elevated levels of anomalous cognitions, including positive and disorganised schizotypal personality traits. The results of Phase 2 add confirmatory evidence to previous research suggestive of memory and perceptual biases plus comorbid psychopathology and the vividness of visual imagery as being integral to the psychogenesis of psychosis-like (anomalous) symptomatology.

<p>Stock recommendations,frequently produced under time pressure, are susceptible to being the result of automatic and intuitive thinking. This is associated with using heuristics in decision-making which is studied by an entire school of research – the heuristics and bias approach. Heuristics of representativeness, availability and anchoring including associated biases as defined by Tversky and Kahneman provide the theoretical framework for the study. This study is aimed at extending the understanding of biases in general and cognitive biases with regard to stock recommendations. A total of thirty equity recommendations were analyzed. A t-test showed that more biases were present in the incorrect recommendations. Overconfidence, illusion of validity and anchoring were among the most frequently observed. The vast majority of recommendations were characterized by insensitivity to predictability indicating that forecasters are seemingly unaware of the difficulty of accurately predicting where the stock price is going to be within the next three to six months.</p>

Stock recommendations,frequently produced under time pressure, are susceptible to being the result of automatic and intuitive thinking. This is associated with using heuristics in decision-making which is studied by an entire school of research – the heuristics and bias approach. Heuristics of representativeness, availability and anchoring including associated biases as defined by Tversky and Kahneman provide the theoretical framework for the study. This study is aimed at extending the understanding of biases in general and cognitive biases with regard to stock recommendations. A total of thirty equity recommendations were analyzed. A t-test showed that more biases were present in the incorrect recommendations. Overconfidence, illusion of validity and anchoring were among the most frequently observed. The vast majority of recommendations were characterized by insensitivity to predictability indicating that forecasters are seemingly unaware of the difficulty of accurately predicting where the stock price is going to be within the next three to six months.

Increasing amounts of data are being produced and consumed on a daily basis. Every mouse movement and click on a website can be analyzed to discover usage patterns and cognitive load (Jenkins et al., 2014), companies mine purchase histories to discover customer shopping patterns (Brin et al., 1997) and historical business transaction information can be used to improve business processes (Ghattas et al., 2014). Using sophisticated algorithms, data can be turned into information that helps guide marketers, policy makers, business managers, and other decision-makers. However, history has shown that increases in the amount and quality of information do not necessarily lead to better decision outcomes (Dawes et al., 1989). Human decision-makers may fail to understand the information, ignore it, or simply not believe it. Methods for effectively conveying information to humans must be studied so that the full value of information systems can be realized. This dissertation uses three studies to explain the factors that make technology persuasive. In the first study, attitudes toward technology measure how beliefs about technology influence the way people process information. Ordering effects are also examined to determine how people view information from decision support systems, and to find the optimal time to present information to decision-makers. In the second study, the persuasive power of text and audio modalities are compared. Additionally, the loss aversion bias is investigated to determine the utility of leveraging this cognitive bias in a technology context. In the third study, Protection Motivation Theory (Rogers, 1983) is used to extend the loss aversion model from study two. The study also investigates how message vividness and user participation through software personalization influence attitudes and behavior. Together, these experiments extend existing theoretical frameworks while giving actionable guidance to information systems practitioners. The studies demonstrate the importance of understanding cognitive biases, attitudes toward technology, and message delivery in a decision support scenario. These investigations are the first step in creating a more comprehensive model of factors that influence the persuasive power of technology.

Data shows that many inventors continue to expend resources on their inventions even after they have received expert advice suggesting that they cease effort. Using a sample of inventors seeking outside advice from a Canadian evaluative agency, this paper examines how overconfidence, optimism, and illusion of control explain this fact. While overconfidence did not have a significant effect on inventor's decisions, illusion of control and optimism did have an effect. An additional interesting finding is that the more time people have spent working on inventions, the more likely they are to discount this expert advice.

Behavioural economics builds on psychology rather than on sociology, and on cognitive science rather than the science of culture. The same is true for new behavioural scholarship in the legal discipline, whether this is referred to as 'behavioural law and economics' or 'law and the behavioural sciences'. The result of a one-sided definition of a more realist research agenda in legal scholarship is an impoverished understanding of the 'social'. In Thaler and Sunstein's famous concept of nudging, social conformity appears as a property of the individual, which can be instrumentalized by social nudges. More generally, the cognitive strand of behavioural economics lends itself to strategies of regulatory 'debiasing', which suggests that it is possible to get down to pure preferences that are free from any distortions. While this approach neglects the endogeneity, or social contingency, of individual preferences, the social strand of behavioural economics is explicitly concerned with the dynamics of social interaction, or the effects of social interdependence. However, both strands of behavioural economics are still higher on methodological individualism, naturalism or positivism and lower on institutionalism, culturalism or constructivism than a genuinely sociological approach. More specifically, their understanding of the 'social' does not sufficiently account for the social embeddedness of both rational and irrational economic action. What is more, behavioural economics also lacks the means to reflect on the link between science and politics, which includes the question of why different models of economic man are attractive at different points in time. The conceptual move from rational to behavioural economic man bears distinctive policy implications, which are in line with the transformation of welfare capitalism towards 'less state' and 'more market'. While the overall direction of this project gets blurred in Thaler and Sunstein's branding of 'libertarian paternalism', it is evident in the adaptation of consumer policies, which proceeds under the imperative of market-conformity. Accordingly, a strategy of nudging does not put into question the wider institutional context but offers a technical solution to what is defined as a problem of individual behavioural rigidities and cognitive biases in the market environment.

This thesis investigates to what extent do cognitive biases a ect human understanding of interpretable machine learning models, in particular of rules discovered from data. Twenty cognitive biases (illusions, e ects) are analysed in detail, including identi cation of possibly e ective debiasing techniques that can be adopted by designers of machine learning algorithms and software. This qualitative research is complemented by multiple experiments aimed to verify, whether, and to what extent, do selected cognitive biases in uence human understanding of actual rule learning results. Two experiments were performed, one focused on eliciting plausibility judgments for pairs of inductively learned rules, second experiment involved replication of the Linda experiment with crowdsourcing and two of its modi cations. Altogether nearly 3.000 human judgments were collected. We obtained empirical evidence for the insensitivity to sample size e ect. There is also limited evidence for the disjunction fallacy, misunderstanding of and , weak evidence e ect and availability heuristic. While there seems no universal approach for eliminating all the identi ed cognitive biases, it follows from our analysis that the e ect of many biases can be ameliorated by making rule-based models more concise. To this end, in the second part of thesis we propose a novel machine learning framework which postprocesses rules on the output of the seminal association rule classi cation algorithm CBA [Liu et al, 1998]. The framework uses original undiscretized numerical attributes to optimize the discovered association rules, re ning the boundaries of literals in the antecedent of the rules produced by CBA. Some rules as well as literals from the rules can consequently be removed, which makes the resulting classi er smaller. Benchmark of our approach on 22 UCI datasets shows average 53% decrease in the total size of the model as measured by the total number of conditions in all rules. Model accuracy remains on the same level as for CBA.

Cognitive theories of depression have long posited automatic interpretation biases (AIB) as a central contributor to depressed mood. The current study was first to examine AIB in a clinically defined depressed sample. While assessing AIB using a semantic association paradigm, pupillary reactivity was simultaneously recorded to build insight into the AIB process. A total of 53 individuals (25 depressed and 28 healthy control) completed the Word Sentence Association Paradigm for Depression (WSAP-D) while pupillary reactivity was recorded. Results revealed the depressed group was significantly more likely to endorse negative AIB and less likely to endorse benign AIB compared to healthy controls. The depressed group demonstrated a modest effect size difference indicating they were faster to endorse negative AIB compared to the healthy controls, but did not differ in endorsing benign AIB or in rejecting either valence. Pupillary reactivity was found to differentiate behaviorally defined AIB type from a natural processing condition when counter to theorized, group relevant AIB. The depressed group demonstrated greater initial pupillary constriction during initial presentation of ambiguous information and comparatively less pupillary dilation during and after endorsing a benign AIB. Taken together, the results suggest that theorized negative AIB and lack of benign AIB are characteristic of depression, that greater cognitive effort is required to reject interpretations consistent with theorized biases consistent with reinterpretation processes, and that depressed individuals are less engaged with benign AIB compared to healthy controls, possibly associated with hedonic deficits. Theoretical implications and future directions are discussed.