Hand Signals and the Design of Everyday Things.

When it’s not raining here in the Emerald City, I like to commute by bike. Part of being a responsible cyclist is following the rules of the road as any other moving vehicle and using standardized hand signals.

A couple of weeks ago, I’m flying down the Counterbalance, slowing down at the corner of Queen Anne Ave and Mercer to take a left onto Mercer; this is a very busy intersection. So, I gave what I thought was the correct hand signal. But, as I was turning, a motorcyclist behind me yelled something in my direction. I couldn’t understand what he said, but I assume it had something to do with my hand signals. So, when I got home a few hours later, I looked them up online:

According to this diagram, I think what might have happened was while I was signaling a left turn, my arm may have been bent instead of completely straight, indicating a right turn. Note to self: keep your left arm straight when signaling left. But the more I looked at this diagram, the more I started thinking about how you could critique this system according to Donald Norman’s principles in his book, The Design of Everyday Things, something we had just been talking about in the interaction design class I’m teaching at the UW.

In Norman’s book, he talks about the Gulf of Evaluation, conceptual models, and mapping, three principles that the right turn signal illustrated above fails to address. As a cyclist, I know what these hand signals are because I need to use them every time I bike. But I can guarantee you that the majority of motor vehicle drivers don’t know what these mean, with the exception of the left turn signal. Why? Because this signal is mapped very well and conforms to a universal conceptual model. Your left hand is sticking straight out in the direction you want to go. The stop signal is less obvious, although it does make sense upon evaluation. Moving the arm down fits our conceptual model of something moving downward communicating a decrease.

Now, the right hand signal? This is the one I have big problems with. If I were a driver behind this cyclist and had no knowledge of hand signals, there is no way I would conclude that this signal indicates a right turn. First, no logical mapping exists. Using your left arm to signal a right turn? That makes no sense. Second, the arm is bent and pointed upwards. In terms of conceptual models, what does an upward movement generally indicate? An increase. These two problems with this communication design lead to what Norman calls the Gulf of Evaluation.

"The gulf of evaluation is the difficulty of assessing the state of the system and how well the artifact supports the discovery and interpretation of that state."

In this instance, I, the cyclist, am the system. The driver behind me is looking to my hand signals, e.g., the artifact, to determine the state of the system (what I’m about to do next). In the case of the right hand turn signal, the artifact miserably fails to support the discovery and interpretation of my intended actions.

In light of this overly academic analysis, from here on out I’m going to just use my right arm, straight and fully extended, to signal a right turn. I think Norman would approve.