In his 1937 APA presidential address, the noted neobehaviorist Edward Chace Tolman, PhD, made a startling claim: "Everything important in psychology … can be investigated in essence through the continued experimental and theoretical analysis of the determinants of rat behavior at a choice-point in a maze."
Even in its day, this was quite an assertion: psychology boils down to what makes a rat turn left or right in a maze. Tolman was known to overstate the case for effect, but the quote does say something about the importance of maze learning to psychological scientists in the 1930s.
What are the origins of this iconic apparatus, and how did the maze come to be held in such esteem?
Most historians agree that the animal maze was first developed at Clark University in the late 1890s, in the laboratory of Edmund Sanford, PhD, in a study by his graduate student, Willard Small. At about the same time, Edward Thorndike, of cats-in-puzzle-boxes fame, had been experimenting with baby chicks in maze-like devices (he called them "pens") constructed by placing books on end in various configurations, but the Clark experiments were the first real maze studies. They launched a rats-in-mazes tradition that continues to this day.
The idea for the first maze study was sparked by a conversation between Sanford and another Clark graduate student, Linus Kline. Small and Kline were both interested in the then-new Darwin-inspired field of comparative psychology. They had been studying rats and were especially interested in what they called the rat's "home-finding" ability. Kline told Sanford he had observed "runways … made by large feral rats to their nests under the porch of an old cabin on [his] father's farm in Virginia." When these runways were exposed during an excavation, their maze-like appearance immediately suggested to Sanford using the Hampton Court Maze design to study "home-finding."
At that time, the Hampton Court Maze in England was a popular tourist stop, arguably the world's most famous hedge maze. It was part of the sprawling attraction of Hampton Court, just outside London, built as a home away from throne for the British royal family. Built in 1690, the maze consists of twists and turns and six-foot-tall hedges that continue to perplex visitors today. At the time of his conversation with Kline, Sanford had just returned from London; it is conceivable that he had visited the maze on that trip.
Whatever the origins of Sanford's suggestion, the Clark lab soon had its own mini-version of the Hampton Court Maze, redesigned slightly to make it rectangular instead of trapezoidal. The 6' x 8' maze had a wooden floor and wire mesh walls. Small became the lead researcher on the project when Kline had to step away for other research. In 1899, Small began his research, publishing his results two years later. This was a time when psychology was the science of mental life, so it is not surprising that Small described his maze study in "mentalistic" terms, rather than in the kind of language one might expect to read in a more modern "learning" study. So instead of reporting results in terms of error rates and time to completion, Small tried to infer what the rats were doing as they made their way through the maze. Although Small was criticized by Thorndike for being overly anthropomorphic, his results make for fascinating reading. For example, describing a rat almost making a wrong turn, he wrote that the rat "hesitated as if ‘scratching his head,' then entered this [dead end path] slowly and doubtfully—only a few steps, however; then with a sudden turn and a triumphant flick of his tail he returned to the correct path" (Small, 1901).
Despite the anthropomorphism, Small made important observations that were verified by subsequent studies. For instance, two of his rats were blind, yet they learned the maze just as easily as their sighted compatriots. Small's conclusions that vision was not needed to learn the maze, and that learning resulted from "the gradual establishment of direct associations" between maze stimuli and motor responses (Small, 1901), were supported a few years later in a famous series of studies by behaviorism's founder, John Watson.
Maze learning turned out to be more complex than either Small or Watson thought, but Small's work is less important for its conclusions than for the fact that it initiated a flood of research using mazes. Over the next few decades, versions of the Hampton Court Maze and many variations of it appeared throughout the academic landscape, as research psychologists used the maze to explore the basic processes of learning. Rats weren't the only subjects making their way through the twists and turns; Human maze studies began appearing, ranging from simple table-top devices that blindfolded humans tried to learn by running a stylus through grooves cut into wood, to human-size mazes in the spirit of Hampton Court.
One such study was attempted by E.G. Boring, psychology's famous historian. As a graduate student at Cornell, the center of Titchenerian introspective psychology, Boring constructed an outdoor maze similar to the Hampton maze but with a circular design ("walls" were created with wooden stakes and wire). Blindfolded, Boring and several fellow grad students wound their way through the maze, rattling off introspective reports about the experience as they went along. They even tied sacks of flour to their backs with holes piercing the bottom so they could later trace their progress (if only they had taken pictures). The results were inconclusive. Boring later said that the main outcome of his only maze study was that he fell in love with one of his fellow maze-running colleagues, Lucy Day, whom he eventually married.
Mazes reached their experimental heyday in the 1930s and 1940s, when Tolman could claim that rat behavior at a choice point was the key to psychological knowledge and not be laughed off the stage. In those days, mazes were the apparatus of choice in the battles among competing learning theorists (e.g., Tolman, Hull). Today, mazes continue to be used by experimental psychologists. The goal is no longer to understand maze learning per se; rather, the maze is just another useful tool for examining such topics as drug effects and spatial memory.
Yet the maze holds an exalted position in psychology's history, thanks to Kline's comments about rats burrowing under a porch, Sanford's knowledge of Hampton Court, and Small's landmark study.
C. James Goodwin, PhD, is professor emeritus at Western Carolina University. Katharine S. Milar, PhD, of Earlham College, is historical editor for "Time Capsule."
- Miles, W.R. (1930). On the history of research with rats and mazes: A collection of notes. Journal of General Psychology, 3, 324–337.
- Olton, D.S. (1979). Mazes, maps, and memory. American Psychologist®, 34, 583–596.
- Small, W.S. (1901). Experimental study of the mental processes of the rat. American Journal of Psychology®, 12, 206–239.
- Tolman, E.C. (1948). Cognitive maps in rats and men. Psychological Review®, 55, 189–208.
Letters to the Editor
- Send us a letter