When you are typing away at your computer, you don’t know what your fingers are really doing.
That is the conclusion of a study conducted by a team of cognitive psychologists at Vanderbilt and Kobe universities. It found that skilled typists can’t identify the positions of many of the keys on the QWERTY keyboard and that novice typists don’t appear to learn key locations in the first place.
“This demonstrates that we’re capable of doing extremely complicated things without knowing explicitly what we are doing,” said Vanderbilt University graduate student Kristy Snyder, the first author of the study, which was conducted under the supervision of Centennial Professor of Psychology Gordon Logan.
A description of the research will appear in an upcoming issue of the journal Attention, Perception & Psychophysics, which recently posted it online.
The researchers recruited 100 university students and members from the surrounding community to participate in an experiment. The participants completed a short typing test. Then, they were shown a blank QWERTY keyboard and given 80 seconds to write the letters in the correct location. On average, they typed 72 words per minute, moving their fingers to the correct keys six times per second with 94 percent accuracy. By contrast, they could accurately place an average of only 15 letters on a blank keyboard.
The fact that the typists did so poorly at identifying the position of specific keys didn’t come as a surprise. For more than a century, scientists have recognized the existence of automatism: the ability to perform actions without conscious thought or intention. Automatic behaviors of this type are surprisingly common, ranging from tying shoelaces to making coffee to factory assembly-line work to riding a bicycle and driving a car. So scientists had assumed that typing also fell into this category, but had not tested it.
What did come as a surprise, however, was a finding that conflicts with the basic theory of automatic learning, which suggests that it starts out as a conscious process and gradually becomes unconscious with repetition. According to the widely held theory – primarily developed by studying how people learn to play chess – when you perform a new task for the first time, you are conscious of each action and store the details in working memory. Then, as you repeat the task, it becomes increasingly automatic and your awareness of the details gradually fades away. This allows you to think about other things while you are performing the task.
Given the prevalence of this “use it or lose it” explanation, the researchers were surprised when they found evidence that the typists never appear to memorize the key positions, not even when they are first learning to type.
“It appears that not only don’t we know much about what we are doing, but we can’t know it because we don’t consciously learn how to do it in the first place,” said Logan.
Evidence for this conclusion came from another experiment included in the study. The researchers recruited 24 typists who were skilled on the QWERTY keyboard and had them learn to type on a Dvorak keyboard, which places keys in different locations. After the participants developed a reasonable proficiency with the alternative keyboard, they were asked to identify the placement of the keys on a blank Dvorak keyboard. On average, they could locate only 17 letters correctly, comparable to participants’ performance with the QWERTY keyboard.
According to the researchers, the lack of explicit knowledge of the keyboard may be due to the fact that computers and keyboards have become so ubiquitous that students learn how to use them in an informal, trial-and-error fashion when they are very young.
“When I was a boy, you learned to type by taking a typing class and one of the first assignments was to memorize the keyboard,” Logan recalled.
Yale neuroscientist Gordon Shepherd has studied neurons for decades. But until recently he’d never had a neuron he could grasp with his own two hands: Neurons are much too small.
Now he’s got his very own 3D neuron in all its spidery glory, a vastly enlarged but precise replica that is the latest custom-made anatomical model to emerge from the Yale Center for Engineering Innovation and Design (CEID). The model neuron is believed to be the first made with a 3D printer.
“Brain microcircuits have a very complicated 3D architecture,” said Shepherd, a professor of neurobiology at the Yale School of Medicine and author of “The Synaptic Organization of the Brain,” a classic in the literature of neurobiology. “The model will give us unprecedented appreciation of this architecture. It’s like being with someone versus having just a picture.”
Last spring the CEID produced a model of a diseased human knee, along with the tumor eating it away. Inspired by that project (the brainchild of Yale radiology resident Mark Michalski) Shepherd inquired about the possibility of making a neuron.
Enter Joseph Zinter, the CEID’s assistant director, and Yusuf Chauhan, a full-time design fellow there, who together produced the 3D knee. The neuron’s tendril-like structure seemed to them a shape ideally suited for 3D printing.
“The wild, seemingly haphazard geometry of a neuron, with its cell body, delicate branches of dendrites, and long fibers make it nearly impossible to fabricate by conventional means,” Zinter said. “But 3D printers can easily handle these types of complex geometries. They’re the ideal technology for this kind of project.”
Unlike drilling, cutting, and milling, which strip away raw material to create an object, 3D printers add material — exactly and exclusively where it’s needed to form the desired object.
Shepherd’s lab team prepared 3D digital images of a specific mouse neuron — one among millions. Zinter and Chauhan then converted the data into a language readable by the CEID’s printers and set them to work. Within a day Shepherd beheld a hugely magnified but otherwise precise replica of a murine mitral cell, or mouse olfactory neuron. Made of plastic, it measures 4.25 inches high by five inches wide, thousands of times larger than the real thing.
“We’ve been inspecting it from every angle and comparing it with experimental data, ” said Shepherd, who has already presented it to groups of other scientists. As best he can tell, they seem awed, he said.
“There was a bit of a stunned silence when I pulled the model from its box and held it up for all to see,” Shepherd said of a presentation at Yale. “There definitely seems to be something unexpected about seeing a nerve cell in this new guise for the first time.”
That’s what Zinter likes to hear.
“In addition to being used for the fabrication of models, prototypes, and usable parts, 3D printing allows for the visualization of information in new and exciting ways,” he said. “This neuron is a perfect example. Professor Shepherd’s neuron data is now a tangible three-dimensional object. The ability to interact with information in an additional dimension, whether it’s a microscopic neuron or a patient’s CT scan, will lead to new insights and discoveries. Researchers are still on the cusp of how to best use 3D printing technology.”
Shepherd already has plans for 3D prints of more intricate neural networks. “We see a future in which 3D models of nerve cells will be an integral part of doing research and of teaching neurobiology,” he said.