VIDEO + QUIZ: ADVENTURE SURVIVAL GUIDE Got what it takes to outsmart a grizzly, escape a great white, or stare down a charging tusker? Test your survival instincts with our new video series.  See the videos >> AUDIO INTERVIEW: SURVIVAL EXPERT Author  Laurence Gonzales shares his survival insights in an audio interview. Download the audio interview >> YOUR STORY: ARE YOU A SURVIVOR? Life-and-death situations don't only occur in the wild. Read about our readers' experiences or send in your own. Read the survival stories >>

When I was a child, my grandmother owned an ashtray that I loved. It was a ceramic rattlesnake, and its coiled body formed the tray for the ashes. It was dark and dusty and very realistic. Decades later, while hiking in the Santa Monica Mountains, I came across the ruin of a stone house. Everything was shattered and broken. Only the chimney remained standing. I thought, I'll see if I can find a souvenir to take home, a bit of a broken dish with a design on it or some old tool. As I poked through the wreckage, at once I saw it: my grandmother's ashtray, complete and unbroken. Delighted, I reached out to take it. And then its tongue came out. I froze. Its tongue came out again. The serpent was smelling me. I carefully backed away.

Mental models have been the subject of intense research by psychologists for at least two decades. A classic one can be seen on the parking sign for the handicapped. It displays an image that we instantly recognize as a wheelchair, even though it looks very little like one. That's because we code information in an abbreviated form for quick reference. We can also create much more elaborate models. Most people, for example, have a complex model for driving that allows them to do so while talking on the phone and drinking coffee. Once models are established, they require no thought. They're efficient, which is probably why they were selected by evolution.

My mistake in trying to pick up a rattlesnake is an example of how our models can betray us. I had two models working against me. The first was my model of a rattlesnake. It was my grandmother's ashtray. It was ceramic and not real. My nostalgic emotional attachment made this a powerful and persistent model. My lack of experience with real rattlesnakes made for a much weaker competing model. During that split second (about one-twentieth of a second, actually) in which seeing became recognition, my brain selected the strongest mental model: the ashtray.

The second trap was set by my model of what I was doing. I had told myself that I was looking for a souvenir. I had not told myself what should have been obvious and what I knew intellectually to be true: I was in the dry mountain wilderness of California where rattlesnakes are common; I was in a rock ruin where rattlesnakes like to look for mice. The models we make automatically push this sort of intellectual knowledge aside.

At that moment, when I reached out to the rattlesnake, I was exhibiting a flaw that attends many accidents in the wilderness and in daily life. I'd been lulled into complacency by a natural tendency of mammals: to create models. The models include the thing itself and the way it behaves, but also the value of our behavior in relation to the thing—good, bad, or indifferent. This system uses our previous experience to prescribe our behavior in new situations. For example: Skiing deep powder results in a big rush of adrenaline. The next time we find ourselves poised above a great expanse of deep powder, we may launch into it without stopping to assess the danger of an avalanche, even as a tiny voice says: "Wait just a moment, please."

Most things eventually fall into the category labeled "ignore." That's why you can put a new picture on your wall and find that soon you don't even see it. We don't really perceive the world most of the time. We take in perceptions through our senses and then pull up what seems like the most relevant mental model. We see, in other words (and hear and smell and feel), by analogy. This system lets us move smoothly through the world without having to stop all the time and reexamine something we've already examined. It's the OHIO rule: Only Handle It Once. But that efficiency comes at the cost of careful analysis. That's why I reached out my hand to a rattlesnake that was coiled to strike. That also explains how many accidents happen in what appear to be benign places in the wild.

The Merced River in Yosemite National Park flows over Vernal and Nevada Falls, providing ideal places for mental models to set traps for people. The hike to the top of the falls is long and strenuous, climbing 2,050 feet (625 meters) in three and a half miles (six kilometers). You arrive at the top parched and hot. The river is cool and inviting. The model that most of us have will urge us to action: "Drink some of that cool water," it says. "Better still, jump in and cool off." Moreover, the competing model for danger is weak. As John Dill, a Yosemite search-and-rescue ranger, put it, "The dangers in a river are not as obvious as the dangers of standing on the edge of a cliff. It's not an innate fear."

So without further thought you plunge in and then either drown or are swept over the falls (Nevada Fall is almost 600 feet high [183 meters]). That makes the Merced River the number one killer in the park, with Vernal in fourth place and Nevada in fifth. (Numbers two and three are the popular El Capitan for wall climbing and the notoriously hazardous Ledge Trail.)

These models form the basis not only of how we act but what we perceive and believe. We tend not to notice things that are inconsistent with the models, and we tend not to try what they tell us is bad or impossible. Henry Plotkin, a psychobiologist at University College in London, refers to this phenomenon as the "primary heuristic," a tendency to "generalize into the future what worked in the past." Until Reinhold Messner climbed Mount Everest without oxygen in 1978, everyone was held back by the belief that it was impossible. Today an average of seven climbers do it every year. Human evolution had not changed in those few years. What changed was the mental model. Seymour Cray, famous for inventing the fastest supercomputers in his day, liked to hire kids straight out of college, because unlike senior engineers they hadn't yet learned what was impossible.

These models can become remarkably stable even in the face of clear information that would seem to contradict them. That's why we can continue on into deteriorating conditions or ignore obvious hazards. That's why Lynn Hill fell. And that's why we ignore the warning signs posted by the Park Service, climb a guard rail, and throw ourselves into the Merced River a mere 50 yards from the falls.

Continue reading on the next page >>

Survival Intro >>

Page 2 - The Darwin of Dumb >>

Page 3 -  When Mental Models Go Wrong >>

Page 4 - The Trouble With Success >>

Page 5 - Learning How to S.T.O.P. >>

Page 6 - Living Mindfully >>

Page 7 - The Survivor: Rulon Gardner >>

Subscribe now and save!