AIXPERIMENTATIONLAB — How does the use of augmented intelligence in the work context change the working conditions of employees and consequently their mental strain?
Towards an institutionalized format for the design, development, use and diffusion of human-centred artificial intelligence applications
AI-based technologies are seen as having significant potential to increase efficiency and productivity in organisations (Benbya et al., 2020). Therefore, they are increasingly finding their way into companies and thus into our working lives (von Krogh, 2018), which is predicted to change significantly with the implementation of AI-based systems (Webster & Ivanov, 2020). The great power of change attributed to AI-based technologies in the work context has inspired a plethora of books, reports and scholarly articles (see e.g. Thiebes et al., 2021; Susskind, 2020; Makridakis, 2017), which are fuelling an emotional discourse in society. This discourse highlights the current uncertainty in society about what impact the increased use of AI-based technologies, in the work context, will have in the future on a societal, organisational and individual level. This uncertainty is also difficult to address, as the question cannot be answered in a blanket manner. It can be assumed that the effect of AI-based technologies is influenced by a number of parameters and is particularly dependent on how the use of AI-based technologies is designed at the three levels.
Over the past years, we can note an increased use of the design approach of Augmented intelligence in the working context (see for example Jain et al., 2021, Hellebrandt et al., 2021). Compared to fully automated AI applications that replicate or even surpass human performance, this approach pursues a synergistic complement of human and machine intelligence. In doing so, AI-based applications analyse large amounts of data that are almost impossible for humans to grasp. It can, for example, find patterns that can be used as a basis for decision-making. The sovereignty over the final decision-making and the resulting actions remains entirely with the human being. The human in turn “feeds” the AI-based system with their expertise and experience, in order to improve the algorithms for the next case (Hellebrandt et al., 2021).
The human-machine collaboration should create a win-win situation for companies and employees. Organisations should have a more efficient and less error-prone decision-making process (Hellebrandt et al., 2021), and employees should be cognitively relieved by AI-based support (Maes, 1995). This approach and its associated objectives are being tested as part of the ongoing research project AIXPERIMENTATIONLAB. From an occupational psychology perspective, it is of particular interest whether the use of augmented intelligence actually changes the mental workload of employees and consequently reduces negative mental strain.
Differentiation between mental workload and mental strain
When speaking about psychological impacts on humans in working life, it is very important to distinguish between the terms mental workload and mental strain. These might often be confused as being the same but they are not. Also, despite these words being associated with something negative in day to day life, it is important to state that each of them can be concerned with something positive or negative. Essentially, mental workload is seen as the combination of external forces that influence a human being in the work context psychologically. This for example, refers to working conditions like the content and organisation of work, as well as, social contacts at work. Whereas mental strain is the immediate effect of mental workload on individuals, in dependence of their own capabilities. This can either be a positive effect, for example being motivated, or negative, like fatigue in the short term and absenteeism in the long term. Here it is important to note that the same workload can lead to different mental strains (see figure 1; DIN German Institute for Standardization, 2018).
The presumed direct and indirect effect of augmented intelligence systems on mental strain
AI-based assistance systems are part of the working conditions and can therefore be described as a mental workload factor. It can be assumed that this workload factor influences mental strain in two different ways — directly and indirectly (see figure 2).
Expectations are that the immediate interaction between augmented intelligence systems and humans has a direct influence on their mental strain. Whether the influence is positive or negative, is likely to depend on the three system factors — usability, performance and usefulness. It is expected that a low usability of a user interface, which is for example, characterised by a low clarity triggers directly a negative experience of mental strain. Similarly, it can be assumed that low system performance and low system usefulness will psychologically stress the user in direct interaction.
In addition, it is probable that AI-based assistance systems have an indirect influence on the mental stress of employees. This is because existing working conditions often correlate with each other (Rothe et al., 2017). Thus, it can be expected that augmented intelligence systems influence other existing workload factors, which in turn are directly related to mental strain (see figure 2). Such interdependence is also specifically targeted in the augmented intelligence approach. This is because AI-based decision support systems are intended to reduce information deficits and the associated uncertain decision-making situations and, as a result, minimise incorrect strain. These mechanisms of action will be explored in detail in the further course of the AIXPERIMENTATIONLAB. We will keep you informed about the findings here.
The research project AIXPERIMENTATIONLAB runs until the end of 2023 and is funded by the Federal Ministry of Labour and Social Affairs (BMAS) as part of the funding programme “Future-oriented companies and administrations in the digital transformation (room for learning and experimental AI)” — EXP.01.00016.20.
The authors gratefully acknowledge the support of the German Federal Ministry of Labour and Social Affairs (BMAS).
For more information please visit the project’s official website (not optimized for mobile, in German only).
References
Benbya, H., Davenport, T. H., & Pachidi, S. (2020). Artificial intelligence in organizations: Current state and future opportunities. MIS Quarterly Executive, 19(4), 9–21. http://dx.doi.org/10.2139/ssrn.3741983
DIN German Institute for Standardization (2018). DIN EN ISO 10075–1 Ergonomische Grundlagen bezüglich psychischer Arbeitsbelastung — Teil 1: Allgemeine Aspekte und Konzepte und Begriffe (ISO 10075–1:2017); Deutsche Fassung EN ISO 10075–1:2017. Beuth.
Hellebrandt, T., Huebser, L., Adam, T., Heine, I., & Schmitt, R. H. (2021). Augmented Intelligence — Mensch trifft Künstliche Intelligenz. Zeitschrift für wirtschaftlichen Fabrikbetrieb, 116(6), 433–437. https://doi.org/10.1515/zwf-2021-0104
Jain, H., Padmanabhan, B., Pavlou, P. A., & Raghu, T. S. (2021). Editorial for the Special Section on Humans, Algorithms, and Augmented Intelligence: The Future of Work, Organizations, and Society. Information Systems Research, 32(3), 675–687. https://doi.org/10.1287/isre.2021.1046
Maes, P. (1995). Agents that reduce work and information overload. In R. M. Baecker, J. Grudin, W. A. S. Buxton, & S. Greenberg (Hrsg.), Human-Computer Interaction: Toward the Year 2000 (S. 811–821). Morgan Kaufmann. doi.org/10.1016/B978–0–08–051574–8.50084–4
Makridakis, S. (2017). The forthcoming Artificial Intelligence (AI) revolution: Its impact on society and firms. Futures, 90, 46–60. https://doi.org/10.1016/j.futures.2017.03.006
Neuner, R. (2016). Psychische Gesundheit bei der Arbeit (2. Aufl.). Springer Gabler.
Rothe, I., Adolph, L., Beermann, B., Schütte, M., Windel, A., Grewer, A., Lenhardt, U., Michel, J., Thomson, B., & Formazin, M. (2017). Psychische Gesundheit in der Arbeitswelt: Wissenschaftliche Standortbestimmung. Bundesanstalt für Arbeitsschutz und Arbeitsmedizin. https://www.baua.de/DE/Angebote/Publikationen/Berichte/Psychische-Gesundheit.pdf?__blob=publicationFile
Susskind, D. (2020). A world without work: Technology, automation and how we should respond. Penguin UK.
Thiebes, S., Lins, S., & Sunyaev, A. (2021). Trustworthy artificial intelligence. Electronic Markets, 31(2), 447–464.
Von Krogh, G. (2018). Artificial intelligence in organizations: New opportunities for phenomenon-based theorizing. Academy of Management Discoveries, 4(4), 404–409. https://doi.org/10.5465/amd.2018.0084
Webster, C., & Ivanov, S. (2020). Robotics, artificial intelligence, and the evolving nature of work. In B. George & J. Paul (Hrsg.), Digital Transformation in Business and Society (S. 127–143). Palgrave Macmillan.
