Skip to Content

Brain Sensor for Market Research

A startup claims to read people’s minds while they view ads.
December 7, 2007

Market researchers have long sought people’s assessments of not-yet-released advertisements and products. But when people recall how they felt during a commercial, for instance, they often can’t accurately describe what their reactions were at each moment in the 30-second spot. Now a San Francisco startup called Emsense claims that it has the tools needed to monitor a person’s true reactions during an entire commercial or video game.

Mind the marketing: Emsense has developed a sensor-laden headset (above) that testers in market-research studies wear. It tracks brain activity using a single electroencephalography sensor (EEG) at the forehead, as well as other sensors that monitor breathing rate, head motion, heart rate, blink rate, and skin temperature.

The company has developed a sensor-laden headset that tracks brain activity using a single electroencephalography sensor (EEG) at the forehead, and other sensors that monitor breathing rate, head motion, heart rate, blink rate, and skin temperature–all of which can be indicators of whether a person is engaged or excited. In addition, says Hans Lee, chief technology officer at Emsense, his team has built proprietary algorithms that find meaning from the data collected by the sensors. Founded in 2004, the company was originally developed to build an EEG-based video-game controller. (See “Connecting Your Brain to the Game.”) Recently, though, the team found that using its technology for market research is more lucrative. And, as the political season ramps up, the company is testing its system internally on campaign ads.

“Our technology allows us to collect moment-by-moment metrics while avoiding the cognitive bias that can interfere with self-reporting and focus groups,” says Lee. He explains that Emsense spent about four years collecting data on how viewers reacted to specific events in commercials and games, such as an intense battle scene, or a joke or a sales pitch in an ad. The company used this data to build mathematical models describing how physiological signals change in response to specific events. The technology, he claims, can let a game maker know the point at which people get sucked into a game and the point at which they lose interest. An advertiser can learn if its sales pitch comes at a time when commercial watchers have a positive or a negative feeling about the ad, Lee says.

To be sure, the idea of using biometrics to provide marketing insight isn’t new. In fact, there is a fairly large body of published scientific work that correlates people’s physiological responses toadvertisements by measuring skin conductance, heart rate and eye movements, says Alan Gevins, executive director of the San Francisco Brain Research Institute and president of SAM Technology, a company funded by Federal grants since 1986 to do brain research and develop technology that tests attention and memory using EEG. There are also some studies using EEG to evaluate advertising but it is a much more difficult problem technically and scientifically and the results are not that robust. “Use of EEG for this purpose is not at all ready for commercial prime time,” he says. “It’s really difficult.”

Gevins says that he has no way of evaluating the Emsense technology because there is no technical information on their website about what they are actually doing or any data or references to peer-reviewed research they’ve published using their technology. But he notes that Emsense isn’t the first company that has tried to use EEG sensors to track people’s emotional responses to advertisements. “There have been a number of start-ups that have made EEG headsets of unknown quality and tried to market devices or studies for evaluating advertising and other analogous uses,” Gevins says. “They have a lot of press releases, but ultimately they fail … The bottom line is, they couldn’t do it.” For example, in the 1990s, a company called Capita Research tried to use EEG technology licensed from NASA for marketing purposes. Gevins, who has been working on correlating EEG to attention and memory for decades, says that he’s skeptical of companies that lack scientific depth in this area. “If someone has the capability of actually doing this,” he says, “then there would be objective indicators that it’s real, for instance publications that have gone through peer-review.”

Even so, Emsense, which was founded by a handful of MIT graduates, claims that its technology is superior to that which has come before it. While it hasn’t published in peer-reviewed journals, it has 22 patents that cover topics that range from sensing using a dry EEG sensor to processing the data and analyzing it for its customers, says Lee. And he claims that customers, such as video-game maker THQ, creator of Frontlines, are coming back.

Emsense contacts potential testers in its database of about 5,000 people. These testers go to Emsense testing centers, scattered throughout the country, where they put on the headset and sit down to watch television or play a game. Lee explains that from the headset’s data, the company’s algorithms deduce a handful of coarse assessments about a tester’s experience: when she has positive or negative feelings, whether or not she is concentrating, if she is excited, and, for games, how much she is engaged with it. Emsense then produces charts that qualitatively show these responses throughout the experience.

For instance, in a commercial for a detergent, an advertiser shows a pregnant woman, who is wearing a pink shirt, eating ice cream. A few seconds into the ad, she drops ice cream on her shirt, scoops it up with her spoon, and continues to eat. Toward the end of the commercial, the product is introduced, and the stain removed. According to Emsense’s data, women tend to respond negatively to the first part of the ad, apparently growing distressed as the woman drops the ice cream and scoops it up. When the product name is introduced and the stain removed, women’s responses turn more positive. Men, however, respond positively when the woman drops ice cream, but they remain neutral throughout the rest of the commercial. This is evidence, says Lee, that the commercial works well with its intended audience–namely, women.

But positive emotions, in particular, are difficult to measure using EEG, says Brian Knutson, professor of neuroscience at Stanford University. His group uses functional magnetic resonance imaging to peer deep within the brain, where the electrical signals that correspond positively are strong, he says. But by the time they make their way to the scalp to be measured by EEG, the signals are much weaker, and it’s difficult to get a clean reading. But ultimately, Knutson says, the important question to ask is whether or not Emsense’s approach is better than self-reporting, which is “a hell of a lot cheaper.”

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.