Foraging for food seems to be a straightforward proposition: if you're a hunter, you hunt; if you're a gatherer, you gather; then you eat. What could be simpler? Well, plenty of things, because, as it turns out, foraging is a complex business. In fact, there's a whole foraging theory that was developed in the 1970s to explain animal foraging patterns and strategies. At its core is the idea of a cost-benefit analysis where an animal examines the available food (the benefit) and weighs the amount of energy required to obtain it (the cost). For example, a lion would derive enormous benefit from taking down an adult water buffalo, but the cost is just too high. It's better to pick off a young buffalo or two (lower benefit, but very low cost). Foraging theory also tells us that animals will move to a new foraging area as soon as the costs of foraging in the current area become too high relative to the remaining benefits.

Of course, none of these are conscious strategies on the animal's part. These techniques are hard-wired into animal brains, having been selected for by evolution over millennia. (Animals that can employ these foraging techniques well will have a natural advantage, so they'll be more likely to reproduce.)

We humans have these foraging mechanisms installed in our own brains, and that fact was the inspiration for the theory of information foraging. In the early 1990s, Peter Pirolli and Stuart Card of Xerox's Palo Alto Research Center observed that tracking down information was analogous to foraging for food, so they tried applying foraging theory to information hunting and gathering. Their results, which were first presented in a paper at the 1995 Conference on Human Factors in Computing Systems (see the earliest citation, below), showed that information seekers do use the same strategies as food foragers. That is, they use a (mostly unconscious) cost-benefit analysis, just like a food forager, where the benefit is the information they seek and the cost is the time it takes to find it. And once the costs of the current "information patch" outweigh its remaining benefits, they move on to a different Web site or database. Also, like food foragers, information foragers rely on "cues" that tell them whether a particular patch contains the data they seek (more on this in tomorrow's post).