Arathi Sethumadhavan

Self-Driving Cars: Enabling Safer Human–Automation Interaction

Several car manufacturers and technology companies currently have autonomous cars in their road maps. However, highly reliable automated systems introduce a huge conundrum (Endsley, 2016) – it makes it difficult for human operators to monitor critical pieces of information and take over manual control when needed. For example, drivers with high automation trust are less likely to monitor the automated driving system (e.g., Hergeth, Lorenz, Vilimek, & Krems, 2016a) and more likely to engage in nondriving tasks (e.g., Carsten, Lai, Barnard, Jamson, & Merat, 2012). Whether we like it or not, automation is here to stay and to grow. Given that, how can we make human–automation interaction safer (Endsley, 2016; Hergeth, Lorenz, Vilimek, & Krems, 2016b)?

In-Vehicle Technologies and Driver Distraction

E R G O N O M I C S I N D E S I G N 2 9 Shutko and Tijerina article in this issue.) Although speech-based systems reduce the load on a driver’s visual system, such systems are not without flaws. For example, Lee, Caven, Haake, and Brown (2001) found that speechbased e-mail systems slowed drivers’ response to the braking of a lead vehicle. Drivers also perceived their workload as being higher when interacting with the system tested in this study. Designers should recognize that speech-based interfaces are not infallible and can compete with some of the cognitive resources required for driving – and can even give rise to distraction. It is important that speechbased systems be tested for effects on driving, as should visual-manual interfaces, prior to release in a product. (Also see Geisler and Tengler’s article in this issue.) In addition to design interventions, training can mitigate the distracting effects of concurrent tasks. Horrey, Lesch, Kramer, and Melton (2009) examined the effects of computerbased training on young drivers’ willingness to engage in in-vehicle distracting activities. The training technique aimed at developing drivers’ metacognitive skills so that they were able to monitor and adjust their performance. The training module included general information and demonstrations of distraction, as well as techniques for dealing with it. Drivers who received the training perceived the risk associated with concurrent in-vehicle distracting activities as higher and reported being less willing to perform such activities while driving, compared with the group that received no training. The individuals who received the training were also more likely to wait until the vehicle stopped to engage in the distracting activities. (In addition to training, other types of outreach are covered by Strayer and by McGehee in this issue.) RESEARCH DIGEST

Principles of Persuasive Design

Persuasion skills are essential if you are an entrepreneur aiming to convince your investors or if you are a health care provider wanting your patients to adopt healthy behaviors associated with disease and medication adherence. Today, the task of persuasion can be accomplished by technology perhaps even better than by a human. In fact, persuasive technology refers to technology that is designed to change users’ behaviors without pressuring them. It can take various forms, including wearables, smartphone apps, games, virtual reality environments, and robots (e.g., Song & Fiore, 2017). In the domain of well-being and health, persuasive technology is being used increasingly to encourage individuals to make healthy choices and to reduce overall health care costs. Persuasive systems have the advantage of being interactive, persistent, anonymous, and multimodal, and they can make sense of large amounts of data (e.g., Reza & Burns, 2014). In this Research Digest column, I outline some of the factors to keep in mind when designing persuasive technology:

Use of Sound in Display Design

• • Expose operators to auditory alarms ahead of time to prepare them to detect those alarms subsequently in complex, high-workload situations (Dehais et al., 2014). • • Make auditory content easy to comprehend and stimulating to reduce the negative effects of reduced auditory comprehension during night shifts, when operators are sleep deprived (Pilcher, Jennings, Phillips, & McCubbin, 2016). • • Provide a cycling earcon display to help operators monitor multiple variables. Earcons have the potential to be less intrusive and more informative than auditory alarms. Earcons are short keynotes that are played irregularly to convey information. The sound of an earcon changes when the information it represents changes (e.g., a tremolo high can be used to represent high heartbeat, and a tremolo low can be used to represent low heartbeat, in a multiple-patient display; Hickling, Brecknell, Loeb, & Sanderson, 2017). • • Exercise caution when employing auditory feedback in environments where operators perform tasks that require high working-memory capacity. Although operators are able to ignore learned alarms and prioritize their current high-priority task over the alarms, spoken alarms interfere with operators’ workingmemory resources even when operators try to ignore the alarms (Lacherez, Donaldson, & Burt, 2016). • • Use speech messages rather than auditory sounds or visual texts in highly automated systems (e.g., driverless cars) when operators who are presumably engaged in other activities (e.g., talking, texting, entertainment) need to be alerted about automation status to help maintain their situation awareness (Nees, Helbein, & Porter, 2016). 716444 ERGXXX10.1177/1064804617716444ergonomics in designergonomics in design research-article2017

Health Care Design: What Can We Do Better?

Use color and labeling wisely: Medication overdose is a major patient safety hazard, especially with the increasing reliance on over-the-counter medications. For example, acetaminophen overdose is considered the leading cause of liver failure in the United States; medication package labeling has been identified as playing an important role in such overdoses. Endestad, Wortinger, Madsen, and Hortemo (2016) found that when two medication packages that contain the same ingredient have different package coloring, users conclude that the packages contain two different ingredients, with elderly users making more erroneous conclusions than young users. Highlighting and placing the name of the active ingredient in the upper right-hand corner of the medication package can help patients realize that two medications can have the same ingredient and has the potential to reduce the risk of unintentional overdose.

Building High-Performance Teams

A team is a set of two or more individuals who are working interdependently toward a common goal. Modern work settings or activities, such as spaceship expeditions, product launches, intensive care units, professional soccer, and military reconnaissance, all involve teams. A Google search for “teams” and “HBR” gives a plethora of Harvard Business Review articles on various topics, including the discipline of teams, leadership, emotional intelligence, diversity, and team collaboration and communication. This is a testament to how work settings today are passionate about creating high-performing, cohesive, innovative teams. In this Research Digest I present some of the key factors that affect team performance, based on an analysis of the human factors literature.

Five User Interface Design Tenets

• A permanent, noninterrupting, and dynamic visualization is superior to text-based alerts to aid complex decision making during uncertain situations (Rayo et al., 2015). Dynamic visualizations are advantageous in that they (a) do not interrupt the operator, (b) visually present the risks and benefits of alternatives, and (c) promote comprehension and algorithm transparency of the decision support system. • With smaller keyboards being increasingly used today with portable devices, such as tablets and laptops, it is important that key spacing have a positive impact on typing performance and usability. A key gap of at least 3 mm and center-to-center key spacing of 17 mm is recommended to ensure higher typing accuracy (Madison et al., 2015). • Present dark characters on light background versus light characters on dark background when designing displays. This is especially advantageous with smaller font sizes to facilitate better perception of details (Piepenbrock, Mayr, & Buchner, 2014). The higher luminance provided by dark characters on a light background leads to higher pupillary contraction, increases the depth of field, and leads to higher acuity and superior perception of fine print. • With smartphones being used for myriad activities, such as social networking, messaging, and e-mailing by both younger and older adults, it is important to have reliable text entry methods. Smith and Chaparro (2015) showed that voice input is the most effective text entry method for younger and older adults alike, followed by physical Qwerty, tracing, onscreen Qwerty, and handwriting. Therefore, smartphone manufacturers should consider releasing smartphone models with physical Qwerty keyboards to gain a competitive edge. Handwriting is the least usable text entry method for both older and younger age groups. Tracing input is superior to onscreen Qwerty and handwriting among the onscreen input methods. • When designing large touch-screen user interfaces for young and old users, provide (a) click-to-zoom or drag-to-zoom options as well as a two-finger zooming gesture, (b) a click-to-rotate gesture design as well as a two-finger continuous rotating gesture, (c) touch-screen inclination angles of 60° to 75° to facilitate zooming and rotating, (d) button sizes that are at least 15.9 × 9 mm, and (e) dragging tasks that have right and downward movements (Gao & Sun, 2015).

Designing the Safety-Critical Ecosystem

Several safety-critical domains require operators to maintain their attention for long periods; military, anesthesiology, and baggage screening are all environments where lack of, or late detection of, a critical signal can have severe consequences. The decrement in operator performance can occur as early as within the first 5 min of assuming responsibility. In addition to vigilance decrement issues, safety-critical domains, such as aviation, process control, military, and health care, can also be affected by the sleep disruption of its workforce (e.g., Wickens, Hutchins, Laux, & Sebok, 2015). This Research Digest notes some principles to keep in mind when designing for safety-critical environments involving sustained-attention tasks, shiftwork, or long-duration assignments with sleep deprivation.

Designing the Intensive Care Unit

Intensive care units (ICUs) are complex environments within a hospital that take care of critically ill patients. The ICU provides several opportunities for human factors practitioners to design systems, environments, and clinical work flows that have a positive influence on patient care and health outcomes. In this issue of the Research Digest, I present a few guidelines that should be taken into account when designing a patient-centered ICU environment.

Designing Trustworthy Automation

From self-learning thermostats to agricultural vehicles to baggagescreening systems, automation is becoming an integral part of modern life. In safety-critical domains, automation offers incredible benefits by improving the accuracy of complex tasks. Because automation failures are inevitable, appropriate calibration of operator trust in automation is integral for the optimal performance of human– automation teams. Trust in automation is affected by several factors, such as automation reliability, operator workload and selfconfidence, task difficulty and novelty, consequences of automation misses, and automation display format and content. Trust is therefore a multidimensional concept. So what can be done to appropriately calibrate operator trust level on automation?