Human Decision Research Articles

Development of an Expert System for Diagnosing Musculoskeletal Disease

Musculoskeletal diseases (MSDs), encompasses various conditions affecting muscles, bones, tendons, ligaments, and joints, resulting to pain, inflammation, and limited mobility, significantly impacting individuals' quality of life. Diagnosing these diseases poses a challenge for healthcare professionals due to symptom similarities with other conditions. To address this, the development of expert systems tailored for musculoskeletal diagnosis has emerged as a promising approach to enhance clinical decision-making and improve patient outcomes. This study aims at developing and evaluating an expert system for musculoskeletal disease diagnosis, by leveraging a knowledge base containing information on common musculoskeletal diseases and symptoms. The system utilized a combination of rule-based and machine learning techniques to provide diagnostic recommendations to physicians. Comparative analysis with experienced physicians, using a dataset of patients with known musculoskeletal diseases, revealed the expert system’s diagnostic accuracy of 92%, recall of 98%, Precision of 91%, F1-Score of 94% and a quicker diagnosis compared to physicians. Additionally, the system demonstrated ease of use and user-friendliness. This project focuses on predictive algorithms, leveraging expert systems dating back to the 1970s, emulating human expert decision-making, particularly in disease diagnosis. The development of an expert system for musculoskeletal disease diagnosis symbolizes the convergence of medical expertise, computer science, and artificial intelligence. By integrating machine learning, natural language processing, and decision support systems, these expert systems have the potential to revolutionize musculoskeletal healthcare delivery. In conclusion, our results show that expert systems hold promise in transforming clinical practice and improving patient outcomes in musculoskeletal healthcare through interdisciplinary collaboration and continuous innovation.

Research on shared control of robots based on hybrid brain-computer interface

BackgroundWith the arrival of the new generation of artificial intelligence wave, new human-robot interaction technologies continue to emerge. Brain–computer interface (BCI) offers a pathway for state monitoring and interaction control between human and robot. However, the unstable mental state reduce the accuracy of human brain intent decoding, and consequently affects the precision of BCI control. New methodsThis paper proposes a hybrid BCI-based shared control (HB-SC) method for brain-controlled robot navigation. Hybrid BCI fuses electroencephalogram (EEG) and electromyography (EMG) for mental state monitoring and interactive control to output human perception and decision. The shared control based on multi-sensory fusion integrates the special obstacle information perceived by humans with the regular environmental information perceived by the robot. In this process, valid BCI commands are screened by mental state assessment and output to a layered costmap for fusion. ResultsEight subjects participated in the navigation experiment with dynamically changing mental state levels to validate the effects of a hybrid brain-computer interface through two shared control modes. The results show that the proposed HB-SC reduces collisions by 37.50 %, improves the success rate of traversing obstacles by 25.00 %, and the navigation trajectory is more consistent with expectations. ConclusionsThe HB-SC method can dynamically and intelligently adjust command output according to different brain states, helping to reduce errors made by subjects in a unstable mental state, thereby greatly enhancing the system's safety.

C-frame thinking: Embedding behavioral economics into ecological economics

This paper aims to explore opportunities for integrating Behavioral Economics (BE) into Ecological Economics (EE). By examining the frames of analysis for both disciplines, this study categorizes BE as operating at the individual level (i-frame), while EE addresses systemic aspects of society (s-frame) and extends its considerations to the biosphere (n-frame), advocating for collective action through bottom-up intermediate-level interventions (c-frame).The study posits that EE can benefit from BE’s rich insights into human behavior and decision-making, especially for c-frame action strategies. However, integrating these disciplines requires finding common ontological and epistemological ground to avoid eclecticism and methodological flaws. The integration is approached in two steps: first, adapting BE epistemology to the systems thinking approach of EE, and second, addressing the ontological gap in BE regarding the world surrounding the individual. This paper argues that embedding BE within EE’s ontology points to the necessity of c-frame thinking for human decision-making.A case study of the ex-GKN factory in Italy demonstrates the practical benefits of c-frame thinking in a complex decision process. An alliance of workers, researchers, and civil society movements collaboratively developed a future plan that considered the needs of all stakeholders, showcasing the effectiveness of collective action.

Individuals with anxiety and depression use atypical decision strategies in an uncertain world.

Previous studies on reinforcement learning have identified three prominent phenomena: (1) individuals with anxiety or depression exhibit a reduced learning rate compared to healthy subjects; (2) learning rates may increase or decrease in environments with rapidly changing (i.e. volatile) or stable feedback conditions, a phenomenon termed learning rate adaptation; and (3) reduced learning rate adaptation is associated with several psychiatric disorders. In other words, multiple learning rate parameters are needed to account for behavioral differences across participant populations and volatility contexts in this flexible learning rate (FLR) model. Here, we propose an alternative explanation, suggesting that behavioral variation across participant populations and volatile contexts arises from the use of mixed decision strategies. To test this hypothesis, we constructed a mixture-of-strategies (MOS) model and used it to analyze the behaviors of 54 healthy controls and 32 patients with anxiety and depression in volatile reversal learning tasks. Compared to the FLR model, the MOS model can reproduce the three classic phenomena by using a single set of strategy preference parameters without introducing any learning rate differences. In addition, the MOS model can successfully account for several novel behavioral patterns that cannot be explained by the FLR model. Preferences for different strategies also predict individual variations in symptom severity. These findings underscore the importance of considering mixed strategy use in human learning and decision-making and suggest atypical strategy preference as a potential mechanism for learning deficits in psychiatric disorders.

Individuals with anxiety and depression use atypical decision strategies in an uncertain world

Previous studies on reinforcement learning have identified three prominent phenomena: (1) individuals with anxiety or depression exhibit a reduced learning rate compared to healthy subjects; (2) learning rates may increase or decrease in environments with rapidly changing (i.e. volatile) or stable feedback conditions, a phenomenon termed learning rate adaptation; and (3) reduced learning rate adaptation is associated with several psychiatric disorders. In other words, multiple learning rate parameters are needed to account for behavioral differences across participant populations and volatility contexts in this flexible learning rate (FLR) model. Here, we propose an alternative explanation, suggesting that behavioral variation across participant populations and volatile contexts arises from the use of mixed decision strategies. To test this hypothesis, we constructed a mixture-of-strategies (MOS) model and used it to analyze the behaviors of 54 healthy controls and 32 patients with anxiety and depression in volatile reversal learning tasks. Compared to the FLR model, the MOS model can reproduce the three classic phenomena by using a single set of strategy preference parameters without introducing any learning rate differences. In addition, the MOS model can successfully account for several novel behavioral patterns that cannot be explained by the FLR model. Preferences for different strategies also predict individual variations in symptom severity. These findings underscore the importance of considering mixed strategy use in human learning and decision-making and suggest atypical strategy preference as a potential mechanism for learning deficits in psychiatric disorders.

Michael is better than Mehmet: exploring the perils of algorithmic biases and selective adherence to advice from automated decision support systems in hiring.

Artificial intelligence algorithms are increasingly adopted as decisional aides in many contexts such as human resources, often with the promise of being fast, efficient, and even capable of overcoming biases of human decision-makers. Simultaneously, this promise of objectivity and the increasing supervisory role of humans may make it more likely for existing biases in algorithms to be overlooked, as humans are prone to over-rely on such automated systems. This study therefore aims to investigate such reliance on biased algorithmic advice in a hiring context. Simulating the algorithmic pre-selection of applicants we confronted participants with biased or non-biased recommendations in a 1 × 2 between-subjects online experiment (n = 260). The findings suggest that the algorithmic bias went unnoticed for about 60% of the participants in the bias condition when explicitly asking for this. However, overall individuals relied less on biased algorithms making more changes to the algorithmic scores. Reduced reliance on the algorithms led to the increased noticing of the bias. The biased recommendations did not lower general attitudes toward algorithms but only evaluations for this specific hiring algorithm, while explicitly noticing the bias affected both. Individuals with a more negative attitude toward decision subjects were more likely to not notice the bias. This study extends the literature by examining the interplay of (biased) human operators and biased algorithmic decision support systems to highlight the potential negative impacts of such automation for vulnerable and disadvantaged individuals.

Transfer Learning-Based Steering Angle Prediction and Control with Fuzzy Signatures-Enhanced Fuzzy Systems for Autonomous Vehicles

This research introduces an innovative approach for End-to-End steering angle prediction and its control in electric power steering (EPS) systems. The methodology integrates transfer learning-based computer vision techniques for prediction and control with fuzzy signatures-enhanced fuzzy systems. Fuzzy signatures are unique multidimensional data structures that represent data symbolically. This enhancement enables the fuzzy systems to effectively manage the inherent imprecision and uncertainty in various driving scenarios. The ultimate goal of this work is to assess the efficiency and performance of this combined approach by highlighting the pivotal role of steering angle prediction and control in the field of autonomous driving systems. Specifically, within EPS systems, the control of the motor directly influences the vehicle’s path and maneuverability. A significant breakthrough of this study is the successful application of transfer learning-based computer vision techniques to extract respective visual data without the need for large datasets. This represents an advancement in reducing the extensive data collection and computational load typically required. The findings of this research reveal the potential of this approach within EPS systems, with an MSE score of 0.0386 against 0.0476, by outperforming the existing NVIDIA model. This result provides a 22.63% better Mean Squared Error (MSE) score than NVIDIA’s model. The proposed model also showed better performance compared with all other three references found in the literature. Furthermore, we identify potential areas for refinement, such as decreasing model loss and simplifying the complex decision model of fuzzy systems, which can represent the symmetry and asymmetry of human decision-making systems. This study, therefore, contributes significantly to the ongoing evolution of autonomous driving systems.

THE ROLE OF AI AND WEB 2.0 TOOLS IN DEVELOPING PRAGMATIC COMPETENCE OF L2 ENGLISH LEARNERS

Artificial Intelligence (AI) and web 2.0 tools has transformed various sectors, like healthcare and finance, by imitating human intelligence and decision-making. However, AI still struggles with understanding and effectively participating in human communication, which is a crucial challenge. Pragmalinguistics, a branch of linguistics focused on the relationship between language and context, addresses this issue. It examines pragmatic aspects such as speaker intention, context, and shared knowledge to grasp the intended meaning behind communication. By integrating pragmalinguistics into AI systems, researchers and developers aim to narrow the gap between machines and human communication. This article explores how AI and web 2.0tools are helpful in pragmalinguistics collaborate to improve AI capabilities.

Interpersonal strategy for controlling unpredictable opponents in soft tennis

Competition in sports, unlike cooperation in everyday life, does not involve a single solution because individuals aim to behave unpredictably, thereby preventing others from predicting their actions. This study determined how individuals in court-based sports attempted to control others’ unpredictable behaviors, addressing the gap in previous studies regarding the lack of clarity around strategies employed by individuals in competitive situations. We achieved this by applying a switching hybrid dynamics model, considering external inputs to analyze individual behaviors. Consequently, the study indicates that skilled individuals, in contrast to intermediate players, exhibit greater consistency in their behaviors. These skilled individuals lead others to anticipate their consistency and subsequently employ strategies to disrupt these expectations. This strategy exploits the principles of active human inference, implying that competition involves cooperation. This was revealed by an analysis of both human decision-making and behavior in actual matches as discrete and continuous dynamical systems. This interpersonal strategy could assist policymakers in the field of everyday life to enhance competitiveness. This strategy enables policymakers to adopt new policies that promote cooperation with competitors, ultimately increasing competitiveness in various aspects of our daily lives.

Hybrid reward-punishment in feedback-evolving game for common resource governance.

How to maintain the sustainability of common resources is a persistent challenge, as overexploiters often undermine collective efforts by prioritizing personal gain. To mitigate the overexploitation of resources by violators, previous theoretical studies have revealed that the introduction of additional incentives, whether to reward rule-abiding cooperators or to punish those who overexploit, can be beneficial for the sustainability of common resources when the resource growth rate is not particularly low. However, these studies have typically considered rewarding and punishing in isolation, thus overlooking the role of their combination in common resource governance. Here, we introduce a hybrid incentive strategy based on reward and punishment within a feedback-evolving game, in which there is a complex interaction between human decision making and resource quantity. Our coevolutionary dynamics reveal that resources will be depleted entirely, even with cooperative strategies for prudent exploitation, when resource growth is slow. When the rate of resource growth is not particularly low, we find that the coupled system can generate a state where resource sustainability and cooperation can be maintained. Furthermore, when the rate of resource growth is moderate, we find that achieving this state cannot simply allocate all incentive budgets to reward. In addition, the increase in per capita incentives significantly promotes the stability of this state.

Good decisions in an imperfect world: a human-focused approach to automated decision-making

ABSTRACT Legal rules are based on an imagined regulatory scene that contains presumptions about the reality a regulation addresses. Regarding automated decision-making (ADM), these include a belief in the ‘good human decision’ that is mirrored in the cautious approach in the GDPR. Yet the ‘good human decision’ defies psychological insight into human weaknesses in decision-making. Instead, it reflects a general unease about algorithmic decisions. Against this background I explore how algorithms become part of human relationships and whether the use of decision systems causes a conflict with human needs, values and the prevailing socio-legal framework. Inspired by the concept of Human-Centered AI, I then discuss how the law may address the apprehension towards decision systems. I outline a human-focused approach to regulating ADM that focuses on improving the practice of decision-making. The interaction between humans and machines is an essential part of the regulation. It must address socio-legal changes caused by decision systems both to integrate them into the existing value system and adapt the latter to changes brought forth by ADM. A human-focused approach thus connects the benefits of technology with human needs and societal values.

From Human Perception of Good Practices to Horse (Equus Caballus) Welfare: Example of Equine-Assisted Activities.

Equine-assisted intervention (EAI) studies deal with clients, whereas very few studies focused on the effects on animals. EAI equids are also submitted to management, which influences their welfare. Management and working conditions depend on human decisions and perception. We gathered information through a survey about facilities managers' strategies (n = 51) and obtained direct information on management and working practices and their consequences on equids' welfare through an observational study (n = eight facilities, 174 equids). Differences in managers' perceptions of good management practices were related to the facility's involvement in EAI, e.g., increased awareness of equids' needs (housing and feeding), especially when EAI was the main activity. A detailed observational study on eight additional facilities confirmed that. Facility management profiles were paralleled by equids' welfare profiles. Clear correlates were found between management decisions and welfare consequences. One major factor influencing welfare and human-equid interactions appeared to be working modalities, with more EAI facilities practicing groundwork and bitless work. Facilities where equids were the most involved in mixed activities had the most equids with compromised welfare. Given EAI clients' particularities, conventional working modalities are less adapted and at risk of increased discomfort for equids. Overall, survey and observational approaches converged but some discrepancies (choice of equid type) appeared between the reported and observed prevalence.

An integrated model of prosecutor decision-making

AbstractSociolegal scholarship has long noted the many ways in which the law is interpreted and selectively applied by human decision-makers. Yet, the processes underlying one of the most significant discretionary waypoints in the criminal legal process – prosecutorial charging decisions – remain opaque. Using data from interviews and focus groups with prosecutors in three midsized jurisdictions, we propose a model of charging that integrates legal considerations, social identity, and organizational constraints. We find that felony prosecutors weave together legal and extralegal factors, often relying heavily on criminal history, to evaluate defendants’ moral character. Based on their evaluation of a defendant’s character, prosecutors charge strategically to secure a final disposition and sentence they view as appropriate for the defendant. Prosecutors’ identities and experiences act as lenses through which they interpret case facts in their evaluation of defendants’ character. However, the level of discretion provided by their chief prosecutor and the culture of the court community in which they work condition the process by which prosecutors achieve their desired outcomes for cases.

Toward Constructing Frameworks for Task- and Design-Optimized Visualizations.

Visualization is crucial to augment and enhance human understanding and decision-making in today's data-driven world. However, the way data are visualized can influence and drastically change the conclusions people draw using data. The findings around visualization effectiveness are nuanced, and guidelines for effective visualization design depend on the visual channels used, chart types, and analysis tasks. This points to a significant need to understand the intersection of these factors to create optimized visualizations. We need a framework to define this intersection that fills the gap by providing a task-optimized visualization design for better quality and higher decision-making confidence that gives designers objective guidance. A task-optimized visualization design framework strategically integrates visual channels, visualization types, and specific low-level tasks to enhance data interpretation and optimize user task performance. We discuss constructing a visualization framework that considers both human perception for encoding techniques and the task being performed, enabling optimizing visualization design to maximize efficiency. Furthermore, we highlight a task-optimized framework's impact and potential applications.

Timely, Granular, and Actionable: Designing a Social Listening Platform for Public Health 3.0

Every day, patients access and generate online health content through a variety of channels, creating an ever-expanding sea of digital data. At the same time, proponents of public health have recently called for timely, granular, and actionable data to address a range of public health issues, stressing the need for social listening platforms that can identify and compile this valuable data. Yet previous attempts at social listening in healthcare have yielded mixed results, largely because they have failed to incorporate sufficient context to understand the communications they seek to analyze. Guided by activity theory to design HealthSense, we propose a platform for efficiently sensing and gathering data across the web for real-time analysis to support public health outcomes. HealthSense couples theory-guided content analysis and graph propagation with graph neural networks (GNNs) to assess the relevance and credibility of information, as well as intelligently navigate the complex online channel landscape, leading to significant improvements over existing social listening tools. We demonstrate the value of our artifact in gathering information to support two exemplar public health tasks: (1) performing postmarket drug surveillance for adverse reactions and (2) addressing the opioid crisis by monitoring for potent synthetic opioids released into communities. Our results across data, user, and event experiments show that effective design artifacts can enable better outcomes across both automated and human decision-making contexts, making social listening for public health possible, practical, and valuable. Through our design process, we extend activity theory to address the complexities of modern online communication platforms, where information resides not only in the collection of individual communication activities but also in the complex network of interactions among them.

The bounded intelligence of AI: Superficiality and deceivability

Artificial Intelligence (AI) is typically designed to replace or augment human beings in making and executing decisions. In the domain of organizational management, AI has already found wide applications in many areas such as HR functions, marketing campaigns, competitive analysis, and strategy formulations, exerting increasing impacts on a host of executives and management professionals. Accurate and timely replicating or capturing the know-how of human decisions and actions becomes imperative, as AI has to be, at least as intelligent as, or even more intelligent than, those human actors whose data are being fed to AI during its training and learning iterations. Just as human beings are subject to bounded rationality due to their limited capacities in information processing, AI, too, is subject to bounded intelligence, in the sense that it could appear not as smart as expected, due to its limited capacities in replicating and capturing human expertise, to say nothing of augmenting or extending it. Such bounded intelligence of AI hinges on two underlying factors: superficiality and deceivability. Superficiality refers to the inability of AI to fully replicate human expertise during its learning process due to natural or technical barriers such as imperfect capturability, relational embeddedness, and time sensitivity. Deceivability refers to the inability of AI to accurately capture human expertise during its learning process due to human-actor-created barriers such as passive evasion, deliberate manipulation, and malicious sabotage. Understanding such boundedness of AI and the specific mechanisms of superficiality and deceivability will help us better appreciate the usefulness and limitations of AI and take measures to both remedy and utilize the bounded intelligence of AI.

Advanced Fuzzy-Based Decision-Making: The Linear Diophantine Fuzzy CODAS Method for Logistic Specialist Selection

This study addresses the inherent uncertainty in human decision-making by leveraging fuzzy sets, which were introduced to better capture the imprecision associated with human thoughts and judgments. As an extension to traditional fuzzy sets, the Linear Diophantine Fuzzy Set (LDFS) was developed, offering a more flexible approach by relaxing existing limitations on grade values. The LDFS has found applications across various fields, demonstrating its versatility and effectiveness. In this study, we explore the application of the Linear Diophantine Fuzzy Set within the framework of the Combinative Distance-based Assessment (CODAS) method. The CODAS method stands out because it incorporates Euclidean and Taxicab distances. Decision-making should consider not only the direct distances between ideal solutions and alternatives but also the indirect distances. The foremost objective of this research is to propose a novel approach by integrating the CODAS method with the LDFS to address complex decision-making problems characterized by uncertainty and imprecision. To illustrate the practical utility of the proposed method, we apply it to a numerical example involving the selection of a logistics specialist, a critical decision in emergency logistics optimization. The research also includes a case study on a logistics specialist, highlighting the practical application of the methods discussed. Furthermore, this study provides a comprehensive sensitivity analysis of the parameters' weights to evaluate the robustness and reliability of the proposed method. The results highlight the effectiveness of the LDF CODAS method in making informed and reliable decisions under conditions of uncertainty, paving the way for future applications in other decision-making scenarios.

Effects of machine learning errors on human decision-making: manipulations of model accuracy, error types, and error importance

This study addressed the cognitive impacts of providing correct and incorrect machine learning (ML) outputs in support of an object detection task. The study consisted of five experiments that manipulated the accuracy and importance of mock ML outputs. In each of the experiments, participants were given the T and L task with T-shaped targets and L-shaped distractors. They were tasked with categorizing each image as target present or target absent. In Experiment 1, they performed this task without the aid of ML outputs. In Experiments 2–5, they were shown images with bounding boxes, representing the output of an ML model. The outputs could be correct (hits and correct rejections), or they could be erroneous (false alarms and misses). Experiment 2 manipulated the overall accuracy of these mock ML outputs. Experiment 3 manipulated the proportion of different types of errors. Experiments 4 and 5 manipulated the importance of specific types of stimuli or model errors, as well as the framing of the task in terms of human or model performance. These experiments showed that model misses were consistently harder for participants to detect than model false alarms. In general, as the model’s performance increased, human performance increased as well, but in many cases the participants were more likely to overlook model errors when the model had high accuracy overall. Warning participants to be on the lookout for specific types of model errors had very little impact on their performance. Overall, our results emphasize the importance of considering human cognition when determining what level of model performance and types of model errors are acceptable for a given task.

Human bias in AI models? Anchoring effects and mitigation strategies in large language models

This study builds on the seminal work of Tversky and Kahneman (1974), exploring the presence and extent of anchoring bias in forecasts generated by four Large Language Models (LLMs): GPT-4, Claude 2, Gemini Pro and GPT-3.5. In contrast to recent findings of advanced reasoning capabilities in LLMs, our randomised controlled trials reveal the presence of anchoring bias across all models: forecasts are significantly influenced by prior mention of high or low values. We examine two mitigation prompting strategies, ‘Chain of Thought’ and ‘ignore previous’, finding limited and varying degrees of effectiveness. Our results extend the anchoring bias research in finance beyond human decision-making to encompass LLMs, highlighting the importance of deliberate and informed prompting in AI forecasting in both ad hoc LLM use and in crafting few-shot examples.

Mapping cognitive biases in multidisciplinary team (MDT) decision-making for cancer care in Scotland: a cognitive ethnography study protocol

IntroductionThe efficiency of multidisciplinary teams (MDTs) in cancer care hinges on facilitating clinicians’ cognitive processes as they navigate complex and uncertain judgements during treatment planning. When systems and workflows are...