Dogs, for instance, are usually associated with eagerness and reward-seeking behaviors. The researchers found that those associations meant that consumers were more likely to respond to more-promotional advertisements that featured dogs. By contrast, cats are usually associated with vigilance, which meant that consumers were more likely to respond to prevention-related ads that featured cats.

For example, in one part of the study 157 participants were shown a video of either a dog or cat. Then both groups were asked which brand of vitamin they preferred. Brand A contained vitamin C and iron and promoted high energy, while brand B was high in antioxidants and reduced the risk of cancer and heart disease.

Participants shown a dog video had a higher preference for brand A than participants who saw a cat video.

This pattern was repeated in several experiments, including one in which pet traits were directly incorporated into the advertising text. In this experiment, one group saw a sneaker ad with this message: “Be a dog person! Reach your health goal with eagerness. Our sneakers feature N-Energy synthetic material, which improves breathability of the shoes and promotes strong support for your feet.” Another group saw the same message, but with the first sentence changed to “Be a cat person!” rather than a dog person.

The participants were then told the sneakers retailed for $50 and asked to bid on the shoes. Participants in the dog group bid on average $33.74, while participants in the cat group bid $28.23.

In the same experiment, a third group was shown a more preventive message: “Be a cat person! Reach your health goal with caution. Our sneakers feature N-Ergy synthetic material, which is antiskid and reduces the possibility of foot pain.” A fourth group was shown the same message, but with a dog substituted for a cat in the first sentence. In this instance the cat group bid an average of $32.45, compared with an average $28.40 for the dog group.

“When the advertisement didn’t conform to stereotypical traits, the positive effect of using a dog or cat was muted,” says

Lei Jia,

now an assistant professor at Kent State University and one of the study’s co-authors.

One of the implications of the study, Mr. Jia says, is that while pets can be used as an effective way to manipulate consumer behavior, marketers need to be sensitive to preconceived notions of dog and cat behavior. For that reason, insights from this study may not apply to consumers in other countries where notions of pet traits differ, he says.

Ms. Ward is a writer in Vermont. She can be reached at reports@wsj.com.