Technology-driven behavioral research at Smith

In the olden days—say, ten years ago—researchers studied human behavior using one-way mirrors and paper-and-pencil questionnaires. In the Smith School’s state-of-the-art Netcentric Behavioral Laboratory, researchers use sophisticated computer software to record the responses of study participants through computer keyboards, joysticks, and even special monitors designed to track human eye movements.

The Behavioral Lab was opened in 2003 to help the Smith School’s marketing, management and organization, and decision and information technologies departments conduct research on human behavior. The lab has 18 networked workstations with monitors set in individual carrels. This keeps participants from seeing—and being influenced by—the behavior of those around them. Video cameras and a one-way mirror permit recording and monitoring of participants’ interactions. Eye-trackers that look like standard computer monitors allow researchers to see exactly what people are looking at on a screen. The school’s Media Lab software permits researchers to administer questionnaires via computer and capture clickstream data as participants use the Internet.

These tools make it possible to conduct studies that would be impossible without the lab’s specialized equipment. One such study was conducted by Roland Rust, David Bruce Smith Chair in Marketing, Rebecca Hamilton, assistant professor of marketing, and former Smith PhD Debora Thompson. In this experiment participants sat at the lab’s computers to either use or evaluate one of two virtual digital video players: one loaded with features, and the other relatively simple. The results were surprising. Most people who evaluated the digital video players without using them said they would rather have a digital video player with more features than with less. But when study participants actually used the virtual video players, the majority found the feature-loaded version frustrating and hard to use. What people said they wanted turned out to be quite different than what they actually liked when they had a chance to use the product.

Being able to actually use the products was a key part of the study’s design because participants could not imagine how their product preferences would change before they had a chance to use it. They had to actually use the virtual video player for the “feature fatigue” effect to occur.

“Because we don’t have all the distractions of the real world, the Behavioral Lab gives us both a more accurate way to evaluate behavior, and a way to capture behaviors we couldn’t observe in any other setting,” says Hamilton, who is chair of the committee that manages the Behavioral Lab. “For example, when participants use virtual products in the lab, we can record their entire interaction with the product and then ask them how they feel about the product after they’ve used it.”

In another recent study Rosellina Ferraro, assistant professor of marketing, looked at priming and suggestibility among consumers. Ferraro wondered if seeing a product repeatedly in real-life situations would affect consumer’s choices. In her study, participants were shown a series of 20 photos, some of which displayed a bottle of Dasani water being used by people in ordinary situations. Each photo was on the screen for exactly two seconds. After viewing the photos, they were offered a bottle of water as a thank-you for their time. Several brands of water were present, including the Dasani brand. Ferraro found that participants who were exposed more often to the Dasani photos, and did not know or remember that they had seen the photos, were more likely to choose the Dasani brand.

Practice to Theory For researchers in the school’s management and organization department, the lab is an excellent environment to test behavioral effects that they observe in the actions of managers and leaders in real-world situations.

“In a field study we may see an effect, but we can’t necessarily explain why that effect exists, or what conditions led to that effect,” says Paul Tesluk, associate professor of management and organization. “In real life, you can’t say to one CEO ‘Okay, go out and be charismatic for us’ and to another CEO ‘Don’t be charismatic.’ And you can’t then go out and stop employees after the CEO’s speech and ask ‘How do you feel? What is your level of commitment after this speech?’”

That kind of human interaction may not be possible in a field study, but it can happen in the Behavioral Lab. The lab’s capabilities allowed Tesluk and colleagues Ken G. Smith, Dean’s Chaired Professor of Business Strategy, and former Smith PhD student Rita Cotilla to examine the roles CEO behavior plays in shaping an environment conducive to constructive conflict, the kind that fosters debate and discussion and results in the effective synthesis of diverse information. The study was funded in part by a $325,000 grant from the National Science Foundation.

In the study, student participants worked as teams using a business simulation. The students were given a complicated decision-making task which required them to combine their information and skills. In addition to ‘winning’ the game by completing the task, the students assigned as CEOs were instructed to behave in a way that encouraged fairness, so that each member of their teams had equal input. Researchers then videotaped participants’ interactions throughout the process of the simulation, stopping students periodically to ask them about the CEO’s behavior.

The study also had a field component. This combination of field study and the controlled experiments in the Behavioral Lab produce much stronger and more reliable results than a field study alone, says Tesluk.