scholarly journals Pre-screening workers to overcome bias amplification in online labour markets

PLoS ONE ◽  
2021 ◽  
Vol 16 (3) ◽  
pp. e0249051
Author(s):  
Ans Vercammen ◽  
Alexandru Marcoci ◽  
Mark Burgman

Groups have access to more diverse information and typically outperform individuals on problem solving tasks. Crowdsolving utilises this principle to generate novel and/or superior solutions to intellective tasks by pooling the inputs from a distributed online crowd. However, it is unclear whether this particular instance of “wisdom of the crowd” can overcome the influence of potent cognitive biases that habitually lead individuals to commit reasoning errors. We empirically test the prevalence of cognitive bias on a popular crowdsourcing platform, examining susceptibility to bias of online panels at the individual and aggregate levels. We then investigate the use of the Cognitive Reflection Test, notable for its predictive validity for both susceptibility to cognitive biases in test settings and real-life reasoning, as a screening tool to improve collective performance. We find that systematic biases in crowdsourced answers are not as prevalent as anticipated, but when they occur, biases are amplified with increasing group size, as predicted by the Condorcet Jury Theorem. The results further suggest that pre-screening individuals with the Cognitive Reflection Test can substantially enhance collective judgement and improve crowdsolving performance.

2019 ◽  
Author(s):  
Roman Burič ◽  
Jakub Šrol

Studies on individual differences in susceptibility to cognitive biases have identified several cognitive dispositions which were thought to predict reasoning by contributing to the efficiency of analytic thought. Recently formulated hybrid models, however, suggest that substantial differences between reasoners may arise early already in the intuitive stages of the reasoning process. To address this possibility, we examined standard individual difference measures, mindware instantiation, and conflict detection efficiency as predictors of the accuracy on conflict reasoning problems presented under a two-response paradigm. This was intended to tease apart the predictors of intuitive responding from those factors which only contribute to reasoning when participants have enough time for analytic engagement. We found that participants correctly solved almost half of conflict reasoning problems already at the initial response stage and that the individual differences in initial reasoning performance were predicted by their cognitive reflection, mindware instantiation, and detection efficiency. The findings advance the specification of hybrid dual-process models and provide corroborating evidence that a part of the link between bias susceptibility and cognitive dispositions is due to differences in intuitive processing.


1999 ◽  
Vol 4 (4) ◽  
pp. 205-218 ◽  
Author(s):  
David Magnusson

A description of two cases from my time as a school psychologist in the middle of the 1950s forms the background to the following question: Has anything important happened since then in psychological research to help us to a better understanding of how and why individuals think, feel, act, and react as they do in real life and how they develop over time? The studies serve as a background for some general propositions about the nature of the phenomena that concerns us in developmental research, for a summary description of the developments in psychological research over the last 40 years as I see them, and for some suggestions about future directions.


2018 ◽  
Vol 39 (2) ◽  
pp. 99-106 ◽  
Author(s):  
Michał Białek ◽  
Przemysław Sawicki

Abstract. In this work, we investigated individual differences in cognitive reflection effects on delay discounting – a preference for smaller sooner over larger later payoff. People are claimed to prefer more these alternatives they considered first – so-called reference point – over the alternatives they considered later. Cognitive reflection affects the way individuals process information, with less reflective individuals relying predominantly on the first information they consider, thus, being more susceptible to reference points as compared to more reflective individuals. In Experiment 1, we confirmed that individuals who scored high on the Cognitive Reflection Test discount less strongly than less reflective individuals, but we also show that such individuals are less susceptible to imposed reference points. Experiment 2 replicated these findings additionally providing evidence that cognitive reflection predicts discounting strength and (in)dependency to reference points over and above individual difference in numeracy.


2014 ◽  
Author(s):  
Kilian James Garvey ◽  
Laura Folse ◽  
Crystal Curry

2019 ◽  
Vol 130 (629) ◽  
pp. 1384-1415 ◽  
Author(s):  
Ralph Hertwig ◽  
Michael D Ryall

ABSTRACT Thaler and Sunstein (2008) advance the concept of ‘nudge’ policies—non-regulatory and non-fiscal mechanisms designed to enlist people's cognitive biases or motivational deficits so as to guide their behaviour in a desired direction. A core assumption of this approach is that policymakers make artful use of people's cognitive biases and motivational deficits in ways that serve the ultimate interests of the nudged individual. We analyse a model of dynamic policymaking in which the policymaker's preferences are not always aligned with those of the individual. One novelty of our set-up is that the policymaker has the option to implement a ‘boost’ policy, equipping the individual with the competence to overcome the nudge-enabling bias once and for all. Our main result identifies conditions under which the policymaker chooses not to boost in order to preserve the option of using the nudge (and its associated bias) in the future—even though boosting is in the immediate best interests of both the policymaker and the individual. We extend our analysis to situations in which the policymaker can be removed (e.g., through an election) and in which the policymaker is similarly prone to bias. We conclude with a discussion of some policy implications of these findings.


2021 ◽  
pp. 174702182110693
Author(s):  
Cyril Thomas ◽  
Marion Botella ◽  
André Didierjean

To facilitate our interactions with the surroundings, the human brain sometimes reshapes the situations that it faces in order to simplify them. This phenomenon has been widely studied in the context of reasoning, especially through the attribute substitution error. It has however been given much less attention in the field of perception. Recent research on the bat-and-ball problem suggests that reasoners are able to intuitively detect attribute substitution errors. Using a perceptual illusion drawn from the field of magic, we investigate the extent to which a perceptual form of attribute substitution depends on executive resources and can be detected. We also investigate the relationship between susceptibility to attribute substitution error in the flushtration count illusion and in a French adaptation of the bat-and ball problem. Finally, we investigate the link between the intuitive cognitive style (assessed by the Cognitive Reflection Test) and the susceptibility to the flushtration count illusion. Our results suggest that participants do not detect perceptual attribute substitution error, that this phenomenon could be independent of the executive resources allocated to the task, and could rest on mechanisms distinct from those that produce errors in reasoning. We discuss differences between these two phenomena, and factors that may explain them.


2016 ◽  
Vol 13 (122) ◽  
pp. 20160414 ◽  
Author(s):  
Mehdi Moussaïd ◽  
Mubbasir Kapadia ◽  
Tyler Thrash ◽  
Robert W. Sumner ◽  
Markus Gross ◽  
...  

Understanding the collective dynamics of crowd movements during stressful emergency situations is central to reducing the risk of deadly crowd disasters. Yet, their systematic experimental study remains a challenging open problem due to ethical and methodological constraints. In this paper, we demonstrate the viability of shared three-dimensional virtual environments as an experimental platform for conducting crowd experiments with real people. In particular, we show that crowds of real human subjects moving and interacting in an immersive three-dimensional virtual environment exhibit typical patterns of real crowds as observed in real-life crowded situations. These include the manifestation of social conventions and the emergence of self-organized patterns during egress scenarios. High-stress evacuation experiments conducted in this virtual environment reveal movements characterized by mass herding and dangerous overcrowding as they occur in crowd disasters. We describe the behavioural mechanisms at play under such extreme conditions and identify critical zones where overcrowding may occur. Furthermore, we show that herding spontaneously emerges from a density effect without the need to assume an increase of the individual tendency to imitate peers. Our experiments reveal the promise of immersive virtual environments as an ethical, cost-efficient, yet accurate platform for exploring crowd behaviour in high-risk situations with real human subjects.


2008 ◽  
Vol 4 (3) ◽  
pp. 191 ◽  
Author(s):  
Muhannad Quwaider ◽  
Subir Biswas

This paper presents the architecture of a wearable sensor network and a Hidden Markov Model (HMM) processingframework for stochastic identification of body postures andphysical contexts. The key idea is to collect multi-modal sensor data from strategically placed wireless sensors over a human subject’s body segments, and to process that using HMM in order to identify the subject’s instantaneous physical context. The key contribution of the proposed multi-modal approach is a significant extension of traditional uni-modal accelerometry in which only the individual body segment movements, without their relative proximities and orientation modalities, is used for physical context identification. Through real-life experiments with body mounted sensors it is demonstrated that while the unimodal accelerometry can be used for differentiating activityintensive postures such as walking and running, they are not effective for identification and differentiation between lowactivity postures such as sitting, standing, lying down, etc. In the proposed system, three sensor modalities namely acceleration, relative proximity and orientation are used for context identification through Hidden Markov Model (HMM) based stochastic processing. Controlled experiments using human subjects are carried out for evaluating the accuracy of the HMMidentified postures compared to a naïve threshold based mechanism over different human subjects.


Sign in / Sign up

Export Citation Format

Share Document