About the SA Blog Network



Opinion, arguments & analyses from the editors of Scientific American
Observations HomeAboutContact

Will We Accept Eye-Tracking Gadgets?

The views expressed are those of the author and are not necessarily those of Scientific American.

Email   PrintPrint

It seems like biometric-enabled gadgets should be a hard sell in the post-Snowden marketplace. When Apple announced the Touch ID fingerprint sensor on the iPhone 5s, Twitter lit up with a blaze of NSA jokes. And yet Apple sold some 4 million 5s’s in the first weekend.

This is no surprise, of course. History suggests we’ll learn to accept any new and cool technology the industry throws at us, no matter how initially creepy. In all likelihood, our devices will soon read our prints and recognize our voices and irises. “The integration of biometrics in consumer electronics is really a forgone conclusion,” says Steve Koenig, director of industry analysis for the Consumer Electronics Association.

But I have a theory: If—if—consumers ever draw a line, they will do so when their gadgets start tracking their eyes.

Psychologists and other scientists have long used eye-tracking cameras for research, but the technology is just beginning to move into consumer products. The company MindFlash is beta-testing FocusAssist, an eye-tracking feature that, paired with corporate-training software, can let Human Resources know whether you really paid attention to that orientation video. Tobii Technology, a Swedish firm, is developing eye-tracking technology for use in laptops. The new Samsung GS4 smartphone pauses videos when you look away (although this isn’t true eye-tracking technology; the phone does this by tracking your face and eyes). Google has patented gaze-tracking technology that could allow it to charge advertisers literally by the eyeball.

To my mind, these technologies are categorically different from, say, Apple’s Touch ID. Touch ID is designed to increase security. (Set aside the question of whether it works; within days of the iPhone announcement, German hackers announced that they had cracked Touch ID with a fingerprint they lifted from a piece of glass.) Trust Apple with your fingerprint, the idea goes, and Apple can better protect your bank information.

But how important is it that your phone pause videos when you look away? Does the benefit outweigh the potential loss of privacy?

There is something intrinsically creepy about a device—manufactured by a multinational corporation that stands to profit by gathering intimate information about you—that continuously monitors the movements of your eyeballs.

Here’s how eye-tracking works: A diode shines near-infrared light on your eyeball, while a camera continuously takes photos of your eye. “It’s looking for two pieces of information: the shape and the orientation of the pupil, and reflectance from the cornea,” says Michael Hout, a visual cognition researcher at New Mexico State University. A processor then builds a 3D representation of your eye. “If it can figure out how much your eye is rotated, it can figure out where your eye is pointed.” Thus, it can figure out what you are paying attention to.

You can’t feel the infrared light, of course. The whole process is imperceptible. So, in a sense, is the motion of your eyes. “You’re not thinking about where you’re looking—it just comes naturally,” Hout says. “Eye movements are outside of the influence of someone’s overt thought.”

The preconscious nature of this whole process is what I suspect could give people the willies in a way that fingerprint sensors do not. An eye-tracking gadget knows where you’re looking before you know yourself. It seems to have the ability to spy on the interior of your consciousness.

Is there any evidence that people will reject eye tracking? Not a lot, other than snarky reactions online to MindFlash’s Focus Assist. And like I said, it is possible—even likely—that electronics manufacturers will introduce biometrics in such measured, shiny doses that we will gradually become comfortable with them.

Still, I have to think that at some point consumers will draw a line. It might not come until Google asks us all to send in a vial of blood for their Google+ DNA bank. But if it happens with eye tracking, you heard it here first.

Seth Fletcher About the Author: Seth Fletcher is a senior editor. Follow on Twitter @seth_fletcher.

The views expressed are those of the author and are not necessarily those of Scientific American.

Rights & Permissions

Comments 9 Comments

Add Comment
  1. 1. looie496 8:30 pm 09/30/2013

    Maybe it’s because I’ve used eye tracking in monkey experiments, but this doesn’t have any creepyness at all for me. And the possibilities are so cool! Wouldn’t it be great to be able to operate an interface just by looking at the controls? To type a text message just by looking at the letters? And that’s just the start.

    Bill Skaggs

    Link to this
  2. 2. Jerzy v. 3.0. 5:38 am 10/1/2013

    I am curious about health effects of this eye tracking. What is the effect on eyes of a mobile shining IR light for many hours?

    Link to this
  3. 3. dakiro 8:31 am 10/1/2013

    Everything emits ir radiation so I do not see any reason to be too concerned with health effects of it.

    Link to this
  4. 4. Jerzy v. 3.0. 8:41 am 10/1/2013

    Well, everything emits light radiation, too – but would you like to stare into bright light for several hours?

    Link to this
  5. 5. Marai 11:50 am 10/1/2013

    It’s not technology itself that is dangerous but its effects on social stratification. Knowledge and ownership of how to master different processes always increase social inequality. It is true to fire laying and opening fire as well as to producing and distributing electricity and delegating tasks to automatic machines. It is very probable that cheap mass production of these kind of tracking technologies will profit much more for big and refined corporations ergo will contribute to even more concentration of capital and power. The (yet hypothetical) cause is this: eye-tracking and similar technologies have efficiency advantages for every applications, AND they have special derived advantages ONLY for those companies who can use complicated and mass size metadata concerning behaviour, communication patterns and other things not on an individual but on an aggregated level, and these are the real winners of the game. Therefore what we are interested in is a more balanced power (information) distribution not only about how gadgets work but what ELSE can be done with the collected data and its consequences and we will need controlling institutions (NGO-s and gov.).

    Link to this
  6. 6. Marai 11:58 am 10/1/2013

    ..and I think in many ways our human senses and brains are becoming more and more like sensors in a huge network, the Leviathan is becoming self-concious. ;-)

    Link to this
  7. 7. oldfartfox 3:02 pm 10/1/2013

    A few years ago I was a semi-serious tournament poker player. Since I frequently played against opponents previously unknown to me, I knew nothing about any nervous physical tells they might have, so I relied quite heavily on eye tracking. Of course I was relying on my totally unaugmented ability to do so. With modern technology I might have bee a major force to be reckoned with.

    Link to this
  8. 8. EyeTrackingAcademic 2:57 pm 03/25/2015

    We’ve been able to control interfaces and type onscreen for years. Many people with physical disability use eye tracking as their main, daily means of communication and work. We’ve been able to play games, use a computer and communicate with eye trackers since the 1980s, when the first manufacturers started selling eye trackers as a communication aide. There’s also a lot of work done on safety levels of infra red light – see here: You can use eye movements without recording them – just like a mouse. The real question of whether eye tracking will be accepted for everyday computing is whether, given most people can use normal interactive media, anyone wants to control things with their eyes. Clicking and typing is simply easier with fingers. The real use of eye movements will be for *attention* measures – such as where people look when, optimising an interface or operating system for the individual user, or a game ‘knowing’ where you’re looking and acting accordingly. Right now, the quality of the data is the main issue. Eye trackers capable of accurate and precise measures are still expensive. Cheap options give horrible precision – meaning that the really interesting measures – e.g. whether a driver is getting tired, or whether a person is likely to remember what they just looked at given the kinds of scanning movements the eye is making, or even whether clinically relevant interruptions to eye movements are identifiable that mean neurological conditions may be the cause – all of these measures require accurate and precise signal which is way beyond all current consumer-priced systems. The kinds of eye trackers researchers use to diagnose Alzheimer’s before other tests do, or to understand the changing attentional states in a normal person, requires frame rates of 250-300Hz, whereas laptop cameras, even the best ones, struggle to get a good image at 30 or 40Hz.

    Link to this
  9. 9. EyeTrackingAcademic 3:00 pm 03/25/2015

    …and for studies to standardize and report how good data really is from various commercially available eyetrackers, see here:

    Link to this

Add a Comment
You must sign in or register as a member to submit a comment.

More from Scientific American

Email this Article