Death by GPS. GPS mis-routing is the easiest and most relatable example of human-automaiton interaction. Unfortunately, to its detriment, this article does not discuss the automation literature, instead focusing on more basic processes that, I think, are less relevant.
This is the first post in an upcoming series about human factors graduate school.
If you have decided that you might want to further your education in human factors and ergonomics by going to graduate school, here is some useful information that Anne and I have collected over the years. While there are many sources of similar information, this one is tailored to potential HF students and answers questions that we’ve received.
First, graduate school will be very different from undergraduate. Yes, you take classes, but the most important experience is in conducting research–that is how you will be evaluated and ultimately what determines whether you are successful.
Most prospective students in HF are interested in the topic because they are interested in design or usability. It is important to realize that graduate school will not be like working in a design studio. Instead, it will be more like being in an experimental psychology program where you take courses in statistics, research methods, cognition, perception, etc.
You will also take specialized courses in usability or other evaluation methods but it will be one of many. The goal is to educate you on the fundamentals of human capabilities and limitations so that you can then use this knowledge in the design or evaluation of artifacts (for those going into applied fields).
In the rest of this series, we’ll discuss researching programs, contacting faculty, and various dos and don’ts.
The field of artificial intelligence is obviously a computational and engineering problem: designing a machine (i.e., robot) or software that can emulate thinking to a high degree. But eventually, any AI must interact with a human either by taking control of a situation from a human (e.g., flying a plane) or suggesting courses of action to a human.
I thought this recent news item about potentially dangerous AI might be a great segue to another discussion of human-automation interaction. Specifically, to a detail that does not frequently get discussed in splashy news articles or by non-human-factors people: degree of automation. This blog post is heavily informed by a proceedings paper by Wickens, Li, Santamaria, Sebok, and Sarter (2010).
First, to HF researchers, automation is a generic term that encompasses anything that carries out a task that was once done by a human. Such as robotic assembly, medical diagnostic aids, digital camera scene modes, and even hypothetical autonomous weapons with AI. These disparate examples simply differ in degree of automation.
Let’s back up for a bit: Automation can be characterized by two independent dimensions:
STAGE or TYPE: What is it doing and how is it doing it?
LEVEL: How much it is doing?
Stage/Type of automation describes the WHAT tasks are being automated and sometimes how. Is the task perceptual, like enhancing vision at night or amplifying certain sounds? Or is the automation carrying out a task that is more cognitive, like generating the three best ways to get to your destination in the least amount of time?
The second dimension, Level, refers to the balance of tasks shared between the automation and the human; is the automation doing a tiny bit of the task and then leaving the rest to the user? Or is the automation acting completely on its own with no input from the operator (or ability to override)?
If you imagine STAGE/TYPE (BLUE/GREEN) and LEVEL (RED) as the X and Y of a chart (below), it becomes clearer how various everyday examples of automation fit into the scheme. As LEVEL and/or TYPE increase, we get a higher degree of automation (dotted line).
Mainstream discussions of AI and its potential dangers seem to be focusing on a hypothetical ultra-high degree of automation. A hypothetical weapon that will, on its own, determine threats and act. There are actually very few examples of such a high level of automation in everyday life because cutting the human completely “out of the loop” can have severely negative human performance consequences.
The figure below shows some examples of automation and where they fit into the scheme:
Wickens et al., (2010) use the phrase, “the higher they are, the farther they fall.” This means that when humans interact with greater degrees of automation, they do fine if it works correctly, but will encounter catastrophic consequences when automation fails (and it always will at some point). Why? Users get complacent with high DOA automation, they forget how to do the task themselves, or they loose track of what was going on before the automation failed and thus cannot recover from the failure so easily.
You may have experienced a mild form of this if your car has a rear-backup camera. Have you ever rented a car without one? How do you feel? That feelings tends to get magnified with higher degrees of automation.
So, highly autonomous weapons (or any high degree of automation) is not only a philosophically bad/evil idea, it is bad for human performance!
First, a disclaimer: this isn’t a full-on review of the Watch. There are more qualified people to review the gadget. The newest one comes from one of the best and most thorough hardware review sites: Anandtech.
One part of the review was particularly insightful:
Overall, I found that the fitness component of this watch to be a real surprise. I often hear that Apple is good at making things we didn’t know we wanted, but this is probably the first time I’ve really believed that statement. Going into the review, I didn’t really realize that I wanted a solid fitness tracker on a smartwatch, but now I’m really convinced that there is value to such features.
This has been my experience as well. I’ve never cared to wear a fitness tracker but i’m surprised at how much I pore over the stats of my standing, activity, and workout levels. The watch also provides a surprisingly effective level of motivation (badges & activity circles).
My activity level (for someone who sits at a desk most of the time) has dramatically increased since the watch (see right; yellow line is when I got the watch).
We used to think that smartphones were the “ubiquitous” technology but there are times I leave it behind. The watch is always-on and there will be interesting use-cases and challenges in the future. I’d love to start my car with my watch!
Some other random thoughts:
The fitness features are great but I wish there was a better way to view my data:
View splits on outdoor runs
View all my workouts instead of looking for them in the calendar view.
Many reviews i’ve read assume the watch will replace the phone. But doing any extended activity really tires the shoulders! My interactions are really limited to much less than 5-10 seconds.
I notice that haptic feedback on the wrist is much less jarring and easier to dismiss (i.e., not as disruptive) as vibrating phones on the body.
The Apple Watch is made for travel:
Most airlines have applets for the watch that make it so easy to keep track of gates, departures, & arrivals.
Boarding a plane with your watch feels very futuristic but most pass readers are on the right side and I wear my watch on the left resulting in very awkward wrist positions. Even when the reader was on the left, it is facing upwards requiring me to turn my wrist downwards.
It is unobtrusive and looks like a watch, not like a gizmo on my wrist.
Apple Pay belongs on the watch. I’ve used Apple Pay on my phone but it is much more seamless on the watch.
Notifications are great if you pare down what can notify you. I only get notified of VIP mail (select senders) and text messages.
Controlling my thermostat, and other electrical devices from my wrist is pretty great.
The big news in tech last week was the unveiling of the Apple Watch. I think it is a nice moment to discuss a range of human factors topics. (This topic may elicit strong feelings for or against Apple or the idea of a smartwatch but let’s keep it about the science.)
The first is technology adoption/acceptance. Lots of people were probably scratching their heads asking, “who wears a watch, nowadays?” But you do see lots of people wearing fitness bands. Superficially, that contrast seems to demonstrate the Technology Acceptance Model (TAM) in action. TAM is a way to try to understand when people will adopt new technology. It boils down the essential factors to usability (does it seem easy to use?) and usefulness (does it seem like it will help my work or life?).
Fitness bands check both of the above boxes: since they are essentially single-function devices they are relatively easy to use and tracking fitness is perceived as useful for many people.
Back to the Watch, it may also check off both of the above boxes: it certainly appears easy to use (but we do not know yet), and because it has fitness tracking functions plus many others via apps it certainly may be perceived as useful to the same crowd that buys fitness bands.
The next topic that got me excited was the discussion of the so-called digital crown (shown below). Anne and I have previously studied the contrasts between touch screens and rotary knobs for a variety of computing tasks. Having both choices allows the user select the best input device for the task: touch for pushing big on-screen buttons and large-scale movement and knob for precise, linear movement without obscuring the screen. Using a knob is certainly easier than a touch screen if you have shaky hands or are riding a bumpy cab.
Two small items of note that were included in the Watch was the use of the two-finger gesture on the watch face to send a heart beat to another user–the same gesture many people intuitively think of when they want to feel their own heart beat.
Finally, the Watch has the ability to send animated emoij to other users. What was noteworthy is the ability to manipulate both eyes and mouth in emoji characters. I couldn’t find any literature but I recall somewhere that there is some cross-cultural differences in how people use and interpret emoji: Western users tend to focus on the mouth while Eastern users tend to focus on the eyes (if you know what reference I’m talking about or if I’m mis-remembering, feel free to comment).
There’s so much I haven’t brought up (haptic and multi-modal feedback, user interface design, automation, voice input and of course privacy)!
Anne sent me an example of, “why haven’t they thought of this before”: an air vent with the temperature display and control knob all in one.
In this article describing the new Audi TT with glass dashboard, they describe the novel control/display/air vent seen in the image above. I guess one problem here is if it is accessible to only the driver or if it’s centrally located.
The dashboard (shown in the linked article), however, is another story. While it looks futuristic, it looks like a distraction nightmare!
This is one creative solution to the overwhelming complexity of television remote controls. My only complaint is the very low contrast between the background and the text labels. I think i’ll try this with my Dad’s remote control.
This clip of Fox News’ new studio has been tearing up the internet. But what caught my eye was the touchscreen lag and general unresponsiveness/accidental touches of the users in the background (see image at top; video here). Starting at the 10 second mark, note the user on the right.