For individuals working in safety-critical workplaces – such as aviation, medicine, the military and space exploration – achieving optimal human performance is paramount. Work systems such as these are complex and demanding and errors can have extremely serious consequences. There are many examples of how every day our safety depends on individuals working in safety-critical workplaces to maintain peak performance; from air traffic controllers that ensure safe separation between aircraft, to surgical teams performing emergency operations, or military teams conducting defence operations.
Psychological scientists and practitioners play an important role in supporting people working in safety-critical workplaces to sustain performance, and increasingly in designing work systems that optimise performance. Particularly relevant is the field of human factors, which is the application of psychological principles and practice to the design and engineering of industrial systems and workplaces. The overall goals of the field of human factors are to: optimise human performance, reduce human error risk, improve system efficiency and safety, and enhance user-experience and comfort. It is a multidisciplinary field, drawing upon cognitive science, social psychology, physiology, human-computer interaction and engineering to achieve these goals (Wickens et al., 2015).
Human factors methods
A common method for understanding threats to performance in safety-critical systems is to conduct field studies and expert interviews. While such approaches provide valuable insight to the conditions in which risks to performance might occur and the precursors thereof; these approaches lack the control required to understand the specific mental processes that underlie human performance. Additionally, human errors are fortunately extremely rare in the real world, and for obvious safety reasons, attempting to elicit them in the field is not possible.
To address this, human factors researchers often use ‘microworld’ simulations of complex work environments, in which task conditions can be carefully controlled and behavioural responses accurately measured (Gray, 2002). These environments allow us to evaluate and apply fundamental psychological theories to safety-critical workplace issues. By applying the principles of basic psychology to these simulations, we have learnt a great deal about the limits of human performance.
Critical performance factors
The field of human factors is broad, with psychologists working to apply our understanding of mind and behaviour across many domains and problem areas. Researchers may dedicate significant resources focusing on the nature of specific human performance risk factors (e.g., multitasking), yet the practitioners must integrate and apply this vast knowledge to support human performance. In this article, I will review four areas of human factors research in which psychological researchers and practitioners are crucial: (1) interruptions, (2) fatigue, (3) prospective memory, and (4) automation design. Below, I will introduce each of these topics, and describe the role of psychologists in translating from research to practice. However, it must be acknowledged these are only a subset of the many factors which influence human performance in safety-critical contexts, and do not exhaustively capture the full range of ways that psychologists support performance in safety-critical workplaces.
Interruptions
The nature of modern workplaces means interruptions are commonplace and often unavoidable. In part, this is due to our reliance on information and communication technologies, frequent interpersonal communication and multitasking to achieve task goals (McFarlane, 2002). Interruptions occur when one is forced to shift attention entirely away from a primary task (e.g., writing an email) to a secondary task (e.g., greeting a visitor). It should be noted that while the terms distraction and interruption are used almost interchangeably colloquially, they are psychologically quite distinct concepts. Unlike interruptions, distractions occur when one attends to a secondary task, without stopping the primary task (e.g., talking on phone while driving).
Interruptions can increase the time people take to complete the original task, and cause individuals to forget to complete intended tasks entirely (Trafton & Monk, 2007). Generally speaking, interruptions that are longer, more cognitively demanding, or require similar information processing systems to the primary task, tend to be more disruptive to performance (Wilson et al., 2018). For people working with complex displays, such as enroute air traffic controllers, there is evidence that interruptions can disrupt performance for several minutes after the interrupting task has been completed (Loft et al., 2015).
Unsurprisingly, interruptions are a serious risk that must be managed in safety-critical environments as human error can have disastrous consequences. A notable example of this was Northwest Airlines Flight 255. Prior to completing a pre-flight checklist, the flight crew were interrupted by air traffic control. Tragically, a failure to return to the checklist following the interruption resulted in an accident killing all on board except one passenger (National Transportation Safety Board, 1988).
The role for psychologists is to translate the wealth of research on interruptions into practical interventions and solutions for organisational systems. In many cases, changes to training policies and procedures can have substantive impact. For instance, it is now standard operating procedure in all modern airlines for pilots to restart a checklist if a significant interruption occurs, ensuring accidents such as Flight 255 are not repeated. Similar ‘interruption management’ policies have been trialled or implemented in other domains, such as healthcare. Psychologists have also contributed to the development of technological solutions, such as digital interruption management systems that recommend ‘optimal’ times to interrupt based on progress within a task sequence, or potentially from estimates of cognitive load and attention derived from biosensor technology (e.g., Katidioti et al., 2016).
Prospective memory
In everyday life, people often need to remember to complete an intended action or task at an appropriate future time (e.g., post a letter on the way home from work), which is referred to as a deferred task. The cognitive processes involved with successfully remembering to complete deferred tasks are referred to as prospective memory (Dismukes, 2012). Individuals in safety-critical environments also sometimes face situations where they must rely on prospective memory. For example, in medical contexts, a departing nurse may need to remember to pass on a message to an incoming doctor at the end of their shift, or they may need to remember to resume an interrupted medication preparation procedure at the correct re-entry point.
However, even in safety-critical contexts, individuals can sometimes forget to complete the intended deferred task, known as a ‘prospective memory error’. In a study of an intensive care unit, prospective memory errors were found to account for more than 50 per cent of all errors (Rothschild et al., 2005). In a study of commercial aviation pilots, it was found almost all incidents involving memory errors were due to failures of prospective memory, rather than of retrospective memory – that is, forgetting information of past events (see Dismukes, 2008). Furthermore, through interviews with air traffic controllers and analyses of incident reports, Shorrock (2005) found that prospective memory errors accounted for about 40 per cent of all memory-related errors across a range of controller duties.
The role of psychologists is not only to identify error occurrences and understand the mechanisms underlying prospective memory, but also to develop strategies and interventions to improve prospective memory performance (Loft et al., 2019). One strategy involves analysing work systems and identifying procedures where staff could benefit from implementation planning strategies (i.e., implementation intentions). Implementation planning involves forming a specific plan for how, where, and when one will execute an intended deferred task, and to mentally visualise executing that intention correctly (Gollwitzer & Sheeran, 2006).
In everyday tasks, such as medication adherence, forming these explicit intentions has been found to increase performance as much as two to four times. In safety-critical environments, where individuals must work under high cognitive load, effort has focused on developing computer-based display aids that can help workers keep track of deferred tasks. Generally, prospective memory is found to be most improved by aids that are salient, distinctive and related to the task goal at hand. However, there is still much work to be done by both psychological researchers and practitioners in ensuring a broad range of safety-critical workplaces are considering prospective memory and developing evidence-based interventions.
Mental fatigue
We have all experienced fatigue at some point in life. It is typically defined as a physiological state of reduced mental or physical performance capability, associated with tiredness and a lack of energy. Research shows that when fatigued, people make poorer decisions and experience a range of neurobehavioural deficits, including in memory, attention and emotional processing. In safety-critical workplaces, these deficits translate to increased risk of error, meaning fatigue is crucial to prevent. Fatigue is caused by several interacting factors. Most influentially, fatigue fluctuates in response to endogenous biological processes, such as our circadian rhythm (i.e., body clock) and homeostatic drive (i.e., need for sleep). However, work related factors, such as our motivation or the demands posed by our work can also influence an individual’s fatigue (Boksem et al., 2006; Hockey & Earle, 2006).
Organisational psychologists can help to mitigate fatigue in safety-critical industries by applying work-design principles. This involves considering how work systems and jobs themselves can be adapted to human psychophysiology. For example, one proactive approach psychologists take to protect against fatigue is roster optimisation, in which work rosters are assessed based on their risk of fatigue. This process is supported by using biomathematical modelling tools, which can forecast likely levels of fatigue based on a work or sleep schedule (Dawson et al., 2017). In Australia, many roster systems used in the aviation, rail and mining industries have been developed by psychologists who draw upon a combination of biomathematical approaches and work assessment. Further, by assessing the risk of proposed rostering schedules, psychologists can inform crewing requirements for work systems still under development, proactively addressing future fatigue risks.
Work design can also be applied to safety-critical workplaces to ensure that qualities of the jobs themselves are tolerable and mitigate fatigue risks. A growing body of research has shown that persistently high levels of workload and task demands (i.e., overload) can increase operator fatigue (Studnek et al., 2018; Wilson et al., 2021). This occurs because under high task demands, individuals need to compensate by mustering additional effort, which over time depletes energetic reserves. If left unaddressed, this can lead to more chronic states such as burnout.
Conversely, some passive monitoring jobs (e.g., automated plant monitoring; remote monitoring) can risk inducing underload. Underload is a state of low task engagement, poor concentration and fatigue that occurs from performing monotonous tasks that offer few opportunities for operators to exert active control (Young & Stanton, 2002). In underload situations, the work demands are too low to raise arousal to the level required for sustaining alertness and task engagement.
Organisational psychologists can reduce the likelihood of both underload and overload situations through effective work design. This can involve ensuring variety in day-to-day tasks, providing opportunities for skill development and progression, and reorganising the composition of work tasks (and rest breaks) to ensure demands are tolerable for each individual. Organisational psychologists also have an important responsibility to implement evidence-based fatigue management policy and training. For instance, employee education should cover the recent psychological research on energy management strategies such as microbreaks and effective recovery behaviours.
Automation design
The increase of automation and artificial intelligence in workplace systems has resulted in human factors researchers focusing on the psychology of human-automation interaction. To support performance in safety-critical environments, it is important that we understand the biases people may have when adopting and working with automated technologies, and that we develop automated technologies that actually support human performance. Ultimately, this helps ensure that the anticipated benefits of new technological innovations are realised (Parker & Grote, 2020).
There are reasons why automation may fail to deliver the expected benefits. A growing body of research has shown that humans are imperfect users of automation. Three biases commonly reported with regards to automation are automation ‘misuse’, ‘disuse’ and ‘abuse’ (Parasuraman & Riley, 1997). Misuse occurs when individuals overly rely on the automation. Indeed, several accidents have occurred when individuals have relied purely on automated advice and lost situational awareness. Disuse on the other hand occurs when individuals neglect or underutilise advice or capabilities provided by automated systems. Automation abuse is more systemic and occurs when system designers automate workplace functions without considering the implications it may have on human (and system) performance or on the operator’s situational awareness and ability to control the system. The recent incidents and issues surrounding the MCAS automated flight control system in Boeing’s 737-MAX serve as a timely reminder of the potential catastrophic consequences of failing to consider the human in system design.
In many cases, automation-related errors can be attributed to a mismatch between the cognitive requirements for a task, and the design of a system. For instance, automation is increasingly placing operators into ‘passive monitoring’ roles, in which they must monitor system states for long durations of time. However, studies of attentional processing show there are inherent challenges to sustained vigilance, including fatigue, poor motivation, and inattention (Warm et al., 2008). Although significant efforts have been placed in developing ‘human-centred automation’ that is predictable and transparent to operators, it is becoming clear that the rate of technological advancements means that active human control over technical systems cannot be guaranteed.
While there are no complete solutions to the challenges of optimising human performance in automated systems, there are opportunities here for psychologists to shift towards a more interdisciplinary practice. Parker and Grote (2020) argue that the development of increasingly autonomous work systems requires new ways of thinking about how to ensure safe, meaningful and productive work. Now more than ever is it crucial for psychologists to play an active role applying work design principles to new automated work systems.
Designing human-centred work systems
In this article, I have outlined four areas of human factors research and practice where industrial and organisational psychologists play an important role in optimising work systems for human performance. While each research area discussed appears distinct, optimising human performance in safety-critical workplaces requires considering many different parts of the work system ‘holistically’. For instance, interruptions may be the catalyst to a prospective memory failure, or a newly developed automated system may be responsible for increases in operator fatigue.
Achieving tangible benefits for workers in safety-critical industries requires careful translation of human factors research into practice, policy, and design. The challenge moving forward will be for psychologists to collaborate more directly with managers and system designers to proactively ensure good work and system design when new technologies are implemented. This collaborative and system-based approach is consistent with a socio-technical systems perspective, in which the human is seen as part of a broader system and emphasises the fit between human and technology. This approach is often applied as part of work systems design to ensure the technical, human, and organisational aspects of a system are considered together (Waterson et al., 2002).
Given the collaborative and systems-based approach that psychologists adopt, it is unfortunate that our work is commonly misunderstood as minimising instances of ‘human error’. This misconception is pervasive in media coverage of industrial disasters which all too often attribute the root cause of a particular accident to a single factor: human error, implying that a simple slip or mistake of a single individual can be placed at fault. However, human errors do not occur in a vacuum, but within complex socio-technological systems. In many cases, human error is simply the inevitable consequence of a poorly designed work system that failed to account for the limits of human performance or implement appropriate safeguards.
As such, the role of the psychologist in optimising human performance reaches well beyond understanding the biases and capacity of a single human in isolation, but in understanding human performance as a function of the entire work system and environment. Ultimately, it is through this multi-disciplinary and integrated perspective that psychological researchers and practitioners support performance in safety-critical workplaces.
Contact the author