Explainable Artificial Intelligence (XAI) 2.0: A manifesto of open challenges and interdisciplinary research directions

Information Fusion, Journal Year: 2024, Volume and Issue: 106, P. 102301 - 102301

Published: Feb. 15, 2024

Understanding black box models has become paramount as systems based on opaque Artificial Intelligence (AI) continue to flourish in diverse real-world applications. In response, Explainable AI (XAI) emerged a field of research with practical and ethical benefits across various domains. This paper highlights the advancements XAI its application scenarios addresses ongoing challenges within XAI, emphasizing need for broader perspectives collaborative efforts. We bring together experts from fields identify open problems, striving synchronize agendas accelerate By fostering discussion interdisciplinary cooperation, we aim propel forward, contributing continued success. develop comprehensive proposal advancing XAI. To achieve this goal, present manifesto 28 problems categorized into nine categories. These encapsulate complexities nuances offer road map future research. For each problem, provide promising directions hope harnessing collective intelligence interested stakeholders.

Language: Английский

---

How and why humans trust: A meta-analysis and elaborated model

Frontiers in Psychology, Journal Year: 2023, Volume and Issue: 14

Published: March 27, 2023

Trust exerts an impact on essentially all forms of social relationships. It affects individuals in deciding whether and how they will or not interact with other people. Equally, trust also influences the stance entire nations their mutual dealings. In consequence, understanding factors that influence decision to trust, is crucial full spectrum Here, we report most comprehensive extant meta-analysis experimental findings relating such human-to-human trust. Our analysis provides a quantitative evaluation interpersonal initial propensity as well assessment general trusting others. Over 2,000 relevant studies were initially identified for potential inclusion meta-analysis. Of these, (n = 338) passed screening criteria provided therefrom total 2,185) effect sizes analysis. The dependent variables trustworthiness, supervisors subordinates express each other. Correlational results demonstrated large range trustor, trustee, shared, contextual within working emphasis present work being one several dimensions herein originated. Experimental established reputation trustee shared closeness trustor predictive trustworthiness outcome. From these collective findings, propose elaborated, overarching descriptive theory which special note taken theory's application growing human need non-human entities. latter include diverse automation, robots, artificially intelligent entities, specific implementations driverless vehicles name but few. Future directions momentary dynamics development, its sustenance dissipation are evaluated.

Language: Английский

---

Mental Workload and Human-Robot Interaction in Collaborative Tasks: A Scoping Review

International Journal of Human-Computer Interaction, Journal Year: 2023, Volume and Issue: 40(20), P. 6458 - 6477

Published: Sept. 20, 2023

In human-robot collaboration (HRC), operators work side by with collaborative robots (cobots), a new type of machines able to function safely on tasks shared humans. Despite the increasing presence cobots in factories, quality experience associated workers HRC is an underexplored topic. This review focused mental workload (MWL) reported interacting cobots, its major sources, and potential solutions optimize it during HRC. Out 165 papers identified, 23 were selected as specifically devoted exploration workers' MWL activity. Cobot motion, predictability, task organization communication patterns emerged factors contributing operators' Endowing capacity meet both demands human needs through modulation motion rhythm, flexible physical interaction, more efficient may contribute mitigating MWL.

Language: Английский

---

A multidimensional taxonomy for human-robot interaction in construction

Automation in Construction, Journal Year: 2023, Volume and Issue: 150, P. 104845 - 104845

Published: March 28, 2023

Language: Английский

---

Mental workload assessment by monitoring brain, heart, and eye with six biomedical modalities during six cognitive tasks

Frontiers in Neuroergonomics, Journal Year: 2024, Volume and Issue: 5

Published: March 12, 2024

Introduction The efficiency and safety of complex high precision human-machine systems such as in aerospace robotic surgery are closely related to the cognitive readiness, ability manage workload, situational awareness their operators. Accurate assessment mental workload could help preventing operator error allow for pertinent intervention by predicting performance declines that can arise from either work overload or under stimulation. Neuroergonomic approaches based on measures human body brain activity collectively provide sensitive reliable training environments. Methods In this study, we developed a new six-cognitive-domain task protocol, coupling it with six biomedical monitoring modalities concurrently capture correlates across longitudinal multi-day investigation. Utilizing two distinct each aspect cardiac (ECG PPG), ocular (EOG eye-tracking), (EEG fNIRS), 23 participants engaged four sessions over 4 weeks, performing tasks associated working memory, vigilance, risk assessment, shifting attention, situation awareness, inhibitory control. Results results revealed varying levels sensitivity within modality. While certain exhibited consistency tasks, neuroimaging modalities, particular, unveiled meaningful differences between conditions domains. Discussion This is first comprehensive comparison these brain-body multiple days findings underscore potential wearable sensing methods evaluating workload. Such neuroergonomic inform development next generation neuroadaptive interfaces more efficient interaction skill acquisition.

Language: Английский

---

Assessing Physiological Signal Utility and Sensor Burden in Estimating Trust, Situation Awareness, and Mental Workload

Journal of Cognitive Engineering and Decision Making, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 4, 2025

Effective human-autonomy teaming is increasingly important to ensuring mission success in operational environments. Modeling operators’ cognitive states, including trust, situation awareness (SA), and mental workload (WL), may improve team performance by informing autonomous systems about human operators. Subjective questionnaires are often used measure these states but obtrusive impractical for real-world operations. Integrating observable physiological measures could enable unobtrusive measurements of states. We created models estimate SA, WL using observable, physiological, operator background information (OBI) measures. collected data from 15 subjects during a spacecraft docking simulation. developed LASSO-based algorithm select features, generated multivariate regression models, assessed predictive capabilities. Observable OBI features combined led the best performing model, indicating that signals do not add significant power. Simultaneous feature selection SA yielded comparable fit single state, did reduce number required sensors. The algorithm, use multiple modalities, simultaneously fitting capability can be leveraged better estimating teaming.

Language: Английский

---

Human–Robot Interactions: A Pilot Study of Psychoaffective and Cognitive Factors to Boost the Acceptance and Usability of Assistive Wearable Devices

Multimodal Technologies and Interaction, Journal Year: 2025, Volume and Issue: 9(1), P. 5 - 5

Published: Jan. 9, 2025

Robotic technology to assist rehabilitation provides practical advantages compared with traditional treatments, but its efficacy is still disputed. This controversial effectiveness due different factors, including a lack of guidelines adapt devices users’ individual needs. These needs include the specific clinical conditions people disabilities, as well their psychological and cognitive profiles. pilot study aims investigate relationships between psychological, cognitive, robot-related factors playing role in human–robot interaction promote human-centric approach robotic rehabilitation. Ten able-bodied volunteers were assessed for anxiety, experienced workload, reserve, perceived exoskeleton usability before after task lower-limb (i.e., 10 m path walking trials). Pre-trial anxiety levels higher than post-trial ones (p < 0.01). While trait predictive effort (Adjusted-r2 = 0.43, p 0.02), state score was overall workload 0.45, 0.02). High–average reserve scores perception A negative correlation emerged personal identification (r −0.67, p-value 0.03). preliminary evidence impact psychoaffective on device appreciation training. It also suggests pragmatic measures such familiarization time reduce end-user selection based assessments may provide guidance personalization

Language: Английский

---

Mental workload in worker-drone communication in future construction: Considering coexistence, cooperation, and collaboration interaction levels

Advanced Engineering Informatics, Journal Year: 2025, Volume and Issue: 65, P. 103110 - 103110

Published: Jan. 13, 2025

Language: Английский

---

The Mathematical Meaninglessness of the NASA Task Load Index: A Level of Measurement Analysis

IEEE Transactions on Human-Machine Systems, Journal Year: 2023, Volume and Issue: 53(3), P. 590 - 599

Published: May 23, 2023

Human mental workload can profoundly impact human performance and is thus an important consideration in the design operation of many systems. The standard method for assessing NASA Task Load Index (NASA-TLX). This involves a operator subjectively rating task based on six dimensions. These dimensions are combined into single score using one two methods: scaling summing (where scales derived from paired comparisons procedure) or averaging together. Despite its widespread use, level measurement NASA-TLX's computed has not been investigated. Additionally, nobody researched whether approaches computing overall mathematically meaningful with respect to constituent dimensions' levels measurement. serious deficiency. Knowing what NASA-TLX scores will determine mathematics be meaningfully applied them. Furthermore, if syntheses meaningless, then measure lacks construct validity. research presented this article used previously developed evaluate Results show that can, most situations, treated as interval population analyses ordinal individuals. Our results also suggest methods combining meaningless. We recommend analysts without

Language: Английский

---

MATB for assessing different mental workload levels

Frontiers in Physiology, Journal Year: 2024, Volume and Issue: 15

Published: July 23, 2024

Multi-Attribute Task Battery (MATB) is a computerized flight simulator for aviation-related tasks, suitable non-pilots and available in many versions, including open source. MATB requires the individual or simultaneous execution of 4 sub-tasks: system monitoring (SYSMON), tracking (TRACK), communications (COMM), resource management (RESMAN). Fully customizable, design test duration, number sub-tasks used, event rates, response times overlap, create different levels mental load. can be combined with an additional auditory attention (Oddball) task, physiological constraints (i.e., sleep loss, exercise, hypoxia). We aimed to assess main characteristics assessing workload levels. identified reviewed 19 articles which effects low high were analyzed. Although has shown promise detecting performance degradation due increase workload, studies have yielded conflicting unclear results regarding configurations. Increased (multitasking), overlap are associated increased perceived score (ex. NASA-TLX), decreased (especially tracking), neurophysiological responses, while no effect time-on-task observed. The median duration used 20 min (range 12–60) level 10 4–15). To stimuli respectively 3 events/min 0.6–17.2) low, 23.5 9–65) level. In this review, we give some recommendations standardization design, configuration, description training, order improve reproducibility comparison between studies, challenge future researches, as human-machine interaction digital influx pilots. also discussion on possible use context aeronautical/operational changes . Thus, appropriate difficulty, simulation tool study aircraft pilots, during operational constraints.

Language: Английский

---