We spend our lives engaged in problem solving: When should I leave the house to get to work on time? What can I make for dinner given the stuff in my refrigerator? How much work do I need to get done today in order to leave a bit earlier tomorrow? What’s the best driving route given the traffic report coming over the car radio? Can I make the this green light? Can I talk my way out of a traffic ticket? What’s the maximum amount I can pack into my trunk after a COSTCO run? How can I get that stain off the carpet? Is this blog good-enough to post?
Looking over this sample list of problems, it’s easy to see that some have to do with temporal and spatial processing (e.g. packing the trunk, picking the best route, judging speed, making schedules), some with background knowledge manipulation (e.g. coming up with a recipe given a list of ingredients, looking up cleaning strategies), some with social processing (e.g. ability to analyze social situations and make correct predictions of possible outcomes—”I will get that ticket, if I run that red light.”), and some with metacognitive tasks (e.g. judging quality, comparing standards for work output). Most of the problem solving we do during ordinary, every day tasks doesn’t feel like problem solving and doesn’t take very long (certainly not like my son’s science homework assignments). If we had to resort to physics to calculate stopping distance given our speed, weight of our car, and road conditions each time we approached an intersection, we would have had our license taken away a long time ago. Good drivers can rely on past experience to make quick decisions—most of the time, it’s painless problem solving.
In fact, most of the time when we have to solve a problem, we try to shortcut the procedure and pull up something from memory that worked for us in a similar situation in the past. Professor Andrea diSessa from University of California at Berkeley calls such shortcuts p-prim (for phenomenological primitives), others call them “folsky wisdoms.” Regardless of what we call them, these shortcuts save us a lot of time…but only if they give us to the right solution. The problem is that these shortcuts often lead us astray.
When there’s no time pressure, we can approach our problem solving in a leisurely way—carefully considering all of the data and using a systematic approach to solving a problem. But most of the time, we engage in difficult problem solving under extreme conditions. Consider the Three Mile Island problem. Nuclear plant operators, faced with a possible core meltdown, didn’t have the luxury of time, each second counted. What’s more, these operators only had a limited view of the problem (what their instruments were communicating to them) and only limited access to real physical data (they couldn’t actually see that the valve controlling the water flow to the nuclear core was stuck). Under extreme pressure, people resorted to cognitive shortcuts to solve the problem. Fortunately, after a few false starts, the Three Mile Island operators were able to avert disaster (but not before radioactive toxins were already released into the environment).
As product designers, we have to be careful in trying to understand the collection of cognitive shortcuts our users are likely to bring to product interaction. These shortcuts vary by culture and by situation, and our shortcuts are unlikely to be shared by all our users.
“Grabbity” is a great shortcut to understanding simple gravity problems, but in a more complex situation, it could leave a problem solver hanging.