Research reveals people overestimate slopes, heights

These three illustrations show the importance of proportions in large-scale form, according to Dennis Proffitt's research, which reveals that people overestimate slants and heights. The Rotunda (left) and an original drawing by Thomas Jefferson (center) illustrate how the elevation of the building is based on a circle embedded in a square (right). The illustration appeared with an article, titled "Seeing Big Things: Overestimation of Heights is Greater for Real Objects than for Objects in Pictures," in the July issue of Perception.

By Ida Lee Wootten

Steep hill? It's probably not as steep as you think.

Hills appear steeper than they actually are. For example, everyone judges hills with only a seven-degree slant as extremely steep. The tendency to overestimate steepness is even greater for elderly people, people carrying a heavy load and people who are tired, of low fitness or in poor health.

These are part of the first systematic studies on people's perception of common geographical inclines. Led by psychology professor Dennis R. Proffitt, the research results were published recently in the national periodical, Journal of Experimental Psychology: Human Perception & Performance.

Far from being a problem, the exaggeration of slant perception actually helps people ascend hills, said Proffitt, who noted that earlier work showed that overestimates of pitch are even more pronounced when people view steep hills from the top rather than the bottom.

"The overestimates are functional. They help people pace themselves when ascending hills and may even prevent people from undertaking climbs that would be too difficult. In a similar vein, as anyone in San Francisco can tell you, descending steep hills is difficult. So overestimates encourage people to be cautious," he said.

In a series of experiments on perceptual bias, participants consistently judged a hill with a 10-degree slant to be about 30 degrees. Their guesses for a five-degree hill averaged 20 degrees. When told that their estimates were off by more than 15 degrees -- and that by law the steepest public road in Virginia can have no more than nine degrees in inclination -- most participants were incredulous.

"To look at a 10-degree hill, typically judged to be about 30 degrees by verbal estimates and visual matching, and to be told that it is actually 10 degrees is an astonishing experience for anyone unfamiliar with the facts of geographical slant overestimation," said Proffitt, who holds U.Va.'s Cavalier Distinguished Teaching Chair.

In the experiments, 130 U.Va. students reported their judgments of inclines in three different ways. A verbal report was a simple estimate in degrees of the hill's incline. Participants made reports while standing at the base of hills, looking directly ahead at them. To gain a visual estimate, participants used a disk with an angle that could be adjusted to represent the hill's incline. Judgments based on touch (called "haptic judgments") were made by using a tilt board with a flat palm rest, the tilt of which could be adjusted to match the inclination.

Four experiments were conducted. They showed that hills appear steeper when people carry a heavy load, are fatigued by running, are not physically fit, are elderly or in poor health. The verbal and visual judgments grossly overestimated incline, while haptic judgments were significantly more accurate.

The study participants' verbal and visual overestimates reflect their awareness of how challenging hills can be to climb, Proffitt believes. "The highly exaggerated estimates of slant show that human perceptions are not simple reflections of reality, although most people tend to assume they are," he said. "However, the visually guided actions, reflected in the haptic response, are accurate and unaffected by fatigue, load, fitness, age and health."

In other experiments investigating how people perceive heights, U.Va. researchers found that the larger real objects are outdoors, the greater is the overestimate of their height. The findings were published in the July issue of the national journal, Perception.

The studies, involving 230 U.Va. students, investigated why people experience very small vertical-horizontal distortions when viewing photographs and line drawings, but tend to vastly overestimate vertical extent when viewing real objects outdoors. For example, when viewed in person, the Rotunda appears elongated vertically when compared to Thomas Jefferson's illustration showing the building's horizontal and vertical dimensions are physically equal.

In six experiments with the 230 students, participants viewed objects presented in line drawings or in photographs and found that there was only a small overestimation of their vertical extent. These same objects were also viewed in an outdoor setting and in a 3D virtual reality scene shown in a head-mounted display. In these situations the vertical overestimation was about twice as large as that found when viewing the two-dimension images.

"As with the studies on geographical slant, these findings demonstrate that our everyday perceptions often exhibit large distortions of which we are generally unaware," Proffitt said.