TY - GEN

T1 - Dysfunctional User States in Interface Use and their Dependency on Work Environment and Task Complexity

AU - Якунин, Александр Васильевич

AU - Бодрунова, Светлана Сергеевна

N1 - Yakunin, A. V., Bodrunova, S. S. Dysfunctional User States in Interface Use and their Dependency on Work Environment and Task Complexity // Social Computing and Social Media 15th International Conference, SCSM 2023, Held as Part of the 25th HCI International Conference, HCII 2023, Copenhagen, Denmark, July 23–28, 2023, Proceedings, Part I Сборник - https://link.springer.com/book/10.1007/978-3-031-35915-6 Статья: DOI https://doi.org/10.1007/978-3-031-35915-6_34 https://link.springer.com/chapter/10.1007/978-3-031-35915-6_34#citeas

PY - 2023

Y1 - 2023

N2 - Usability testing today comprises quantitative and qualitative approaches. Within the former, factors that shape test results include four major ones put together within the ‘contextual fidelity’ model. These factors include product features, task complexity, user traits (including cultural belonging and gender), and experiment settings. However, most usability tests only consider one or two, not four of them. Our earlier research [1] has shown that, when four factors are assessed in parallel, cumulative impact of all of them makes test results highly diverge. Our current study complements this research by showing how usability tests diverge when task complexity varies highly. Another gap in usability research is that they miss the point in their final target, as they measure relative efficiency of interfaces for various groups of users but do not study formation of dysfunctional psychological states that critically prevent efficient Internet use and task performance. This happens despite the growing evidence of mental harm brought by interfaces to, e.g., youngsters’ health or office workers. By testing 60 assessors in either groups or individually on tasks that induce dysfunctional states, namely monotony and anxiety, and introducing tasks of varying complexity, we show that excessive cognitive load leads to rapid rise of user dysfunctionality. Our results suggest that anxiety reduction in real-world tasks may not be reached by reducing task complexity. We recommend group performance on monotonous tasks and individual performance on anxiety-inducing tasks. By that, we illustrate our doubt of the possibility to reach any ‘objective’ results of usability tests; the latter need to be treated as fundamentally conditioned by the contextual fidelity factors. This has two consequences: First, usability testing needs to be conducted in accordance with prospective goals of interface use; second, multi-functional interfaces of general use need to pass through multiple usability tests that would combine contextual fidelity factors in various ways.

AB - Usability testing today comprises quantitative and qualitative approaches. Within the former, factors that shape test results include four major ones put together within the ‘contextual fidelity’ model. These factors include product features, task complexity, user traits (including cultural belonging and gender), and experiment settings. However, most usability tests only consider one or two, not four of them. Our earlier research [1] has shown that, when four factors are assessed in parallel, cumulative impact of all of them makes test results highly diverge. Our current study complements this research by showing how usability tests diverge when task complexity varies highly. Another gap in usability research is that they miss the point in their final target, as they measure relative efficiency of interfaces for various groups of users but do not study formation of dysfunctional psychological states that critically prevent efficient Internet use and task performance. This happens despite the growing evidence of mental harm brought by interfaces to, e.g., youngsters’ health or office workers. By testing 60 assessors in either groups or individually on tasks that induce dysfunctional states, namely monotony and anxiety, and introducing tasks of varying complexity, we show that excessive cognitive load leads to rapid rise of user dysfunctionality. Our results suggest that anxiety reduction in real-world tasks may not be reached by reducing task complexity. We recommend group performance on monotonous tasks and individual performance on anxiety-inducing tasks. By that, we illustrate our doubt of the possibility to reach any ‘objective’ results of usability tests; the latter need to be treated as fundamentally conditioned by the contextual fidelity factors. This has two consequences: First, usability testing needs to be conducted in accordance with prospective goals of interface use; second, multi-functional interfaces of general use need to pass through multiple usability tests that would combine contextual fidelity factors in various ways.

KW - anxiety

KW - contextual fidelity model

KW - dysfunctional state

KW - human-computer interaction

KW - interface

KW - monotony

KW - usability

KW - usability testing

KW - user states

UR - https://www.mendeley.com/catalogue/8d959797-5048-352c-bfa6-89a4556c5972/

U2 - 10.1007/978-3-031-35915-6_34

DO - 10.1007/978-3-031-35915-6_34

M3 - Conference contribution

SN - 9783031359149

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 470

EP - 485

BT - Social Computing and Social Media 15th International Conference, SCSM 2023, Held as Part of the 25th HCI International Conference, HCII 2023, Copenhagen, Denmark, July 23–28, 2023, Proceedings, Part I

Y2 - 23 July 2023 through 28 July 2023

ER -