Human Mental Workload: Models and Applications

For human factors researchers and practitioners, mental workload remains both a crucial concept and a nebulous one. After decades of work in this field, there is still no real consensus on the construct of mental workload, although there is wide agreement about its multidimensional nature and the main ways to measure it. With increasing automation in many domains, the issue of underload has attracted a considerable proportion of research effort. This paper summarises work to propose a theory of underload based on the notion of malleable attentional resources, but also raises challenges that this theory – and, perhaps, underload in general – may be specific to automation. The paper goes on to discuss the elusive ‘redlines’ of overload and underload, and concludes by considering both theoretical and applied challenges for current research into mental workload.

While Mental Workload has been widely described in terms of the limited power of a digital computer, this analogy is becoming increasingly untenable. More recently the philosophical concept of Connectionism and the computational model of Parallel Distributed Processing (PDP) have provided an alternative paradigm for Mental Workload which explains some of the unexpected findings in recent research. It also suggests both that cognitive overload is a common, everyday problem and one which is heavily dependent on the whole environment in which it is measured.

Online education is gaining ground in our society due to the introduction of new educational technologies and the pandemic situation we are experiencing. The experience is showing that online teaching makes an extra demand on mental resources as compared to face-to-face teaching. For this reason, we are in need of methodologies to measure this demand for resources in order to propose how to mitigate it. In this paper we propose a methodology based on acoustic voice analysis to measure the mental resource demand of teachers. This methodology is similar to that being used successfully in other fields. The advantages of this methodology are that it does not require any costly and intensive instrumentation to record and analyse data. The only two instruments that the methodology requires are a tape recorder and a software for analysing the acoustic parameters of the voice that can be installed on the teachers’ own computer.

With current advancement in technology, it is expected and hoped that even a conditional or level 3 (L3) autonomous vehicle could alleviate older adults’ mobility issues. These conditional or level 3 autonomous vehicles allow the driver to engage in non-driving task (NDRT), but, it can request the driver to assume control of the vehicle via ‘Takeover request’ when it has reached its operational limits. Considering this could be a challenging for older drivers with their declined cognitive, perceptual, and motor capacities. A systematic review has been conducted to produce literature on their issues in a L3 autonomous vehicle. This review mainly focuses on older drivers’ challenges, perception of workload in AVs and takeover performance. This review is hoped to provide relevant literature on the subject and may help researchers improve and pursue research gaps identified in this paper.

A high mental workload level could significantly contribute to mental fatigue, decreased performance, or long-term health problems [14]. Recently, deep learning models have been trained on Electroencephalogram (EEG) signals to detect users’ mental workload. While such approaches show promising results, they either ignore the noise element inherent in the EEG signals or apply a random set of preprocessing techniques to reduce the noise. Such a lack of uniform preprocessing techniques in cleaning the EEG signals would not allow the comparison of the effectiveness of deep learning models across different studies even when they use the data collected from the same experiment. Therefore, in this study, we aim to investigate the effect of preprocessing techniques defined by neuroscientists in the effectiveness of deep learning models. To do so, we focused on the preprocessing techniques that can be automated and do not need any human intervention, namely a high-pass filter, the ADJUST algorithm, and a re-referencing. Using a publicly available mental workload dataset, STEW, we investigate the effect of these preprocessing techniques in three state-of-the-art deep learning models named Stacked LSTM, BLSTM, and BLSTM-LSTM. Our results show that ADJUST has the most significant effect on the performance of our models compare to other steps. Our findings also show that EEG signals that were prepossessed using the high-pass filter, ADJUST algorithm and re-referencing provided the highest classification performance across the investigated deep learning models. We believe this paper provides an important step towards defining a uniform methodological framework for using deep learning models on EEG signals.

This study evaluated the mental workload of military pilots during day and night conditions using night vision goggles (NVG) in a Flight Training Device, in order to rectify or ratify the fatigue correction factor used for flights using NVG. The experiment used basic military operational tasks measuring physiological data, specifically data on electrocardiography activity and galvanic skin response. Subjective data were gathered using NASA TLX and Psychomotor Vigilance Test methods. After collection, the data were subjected to treatment to correct possible errors during data acquisition and later analyzed in the domains of time and frequency. The analysis did not show a great change in mental workload between the day and night periods, which could be explained by the small sample, the small period between flights and the learning effect between day and night flight.

This study explored how students' main information problems during the information age, namely internet addiction, information overload, and social network addiction, influence holistic well-being and academic attainment. The participants were 226 university students, all UK based and regular internet users. They answered the Internet Addiction Test, Information Overload Scale, Bergen Social Media Addiction Scale, and the Wellbeing Process Questionnaire. Data were analysed with SPSS using correlation and linear regression analysis. The univariate analyses confirmed the negative impact of information overload, internet addiction and social media addiction on positive well-being but not academic attainment. However, multivariate analyses controlling for established predictors of well-being showed that the effects of information overload, internet addiction and social media addiction were largely non-significant, confirming other research using this analysis strategy. Future research should examine the type of internet use as well as the extent of it.

Covid-19 has caused a shift in the working environment, with people mandated to work from home where possible in the UK since March 2020. Cognitive workload is sensitive to environmental changes, so it’s possible that in moving from the office to working from home, people’s cognitive workload has been impacted. The research outlined presents findings from 11 interviews with office workers on whether their cognitive workload has been impacted due to changes in the working environment, consequence of Covid-19. Thematic analysis identified three themes that impact cognitive workload: The home environment, differing distractions and no longer having to commute. The paper finishes with a discussion of these themes in relation to cognitive workload and Covid-19 literature, as well as some recommendations on how employers should be flexible with employees to optimise workload.

The present research investigated the relationship between workload, fatigue, and sleep quality of physicians and nurses in psychiatric hospitals by conducting a cross-sectional survey and a diary study. Both studies were conducted in China in early 2021, investigating the effect of workload on fatigue and sleep quality among psychiatric staff in a real-life setting. Study 1 was a cross-sessional survey, investigating 334 responses from physicians and nurses in five psychiatric hospitals, and Study 2 was a diary study examining the association between workload, fatigue and sleep quality in the working week of 48 psychiatric staff. The findings from the first study showed that the staff reported a high workload, and fatigue and poor sleep quality were very prevalent. Workload was the strongest predictor of fatigue. In the diary study, workload and fatigue increased over the week, and sleep quality declined. This research has identified the importance of studying workload and its effects on psychiatric staff.