Eye tracking can alter decision-making

Eye-tracking technology can be used to influence an individual's decision-making...
24 March 2015

Interview with 

Daniel Richardson, UCL


Person looking ahead


Killing people is bad, right? And surely lying is wrong? We all like to think that we make good judgements based on our inner moral beliefs, and that these are strongly held and unshakeable. But it turns out that our morals may be more wobbly than we'd like to think, according to new research from Daniel Richardson at UCL, where he's been using eye-tracking devices to manipulate peoples' choices, as he explains to Kat Arney...

Daniel - We were interested in a process that people go through when they're making difficult choices. In this case, we looked at moral decisions because they are things that we think are permanent about ourselves. There a permanent things about our values, and our principles, and our ethics, and even our religious beliefs. We wanted to see if you make those decisions in the same way that you make a simple choice like whether or not to have a salad or a hamburger for lunch.

Kat - How do you go about figuring what's going on when someone's making a decision like this?

Daniel - Well, what we did is we looked first at some of the previous research on eye movements and show that if we asked people - if I show you two pictures of people and say, "Who's the more attractive one? Who would you like to date?" what we find is that people look back and forth between those two alternatives. But their gaze will centre around the one that they're going to eventually decide. We can look at people's eye movements and that gives us a little window into the process of their decision. So, before they actually decide, we can tell which one they're leaning towards.

Kat - Their eyes are giving it away, maybe before they've even really figured it out.

Daniel - Yeah. The eye movements are a window into this cognitive process. Well, if we have this window into people's decision making process, can we use that information to change their minds? Can we use it as a tool to persuasion, given that we've got this magic insight into what people are thinking?

Kat - So, how do you go about doing that then using this information about where people look to even manipulate what they're going to say, what they're going to think?

Daniel - So, what we do is, a phrase pops up on a computer, our moral dilemma. Something like, "Is lying always wrong?" And on screen there are two options. You can only have two choices, yes or no or sometimes or always. And then we eye-track people as they look back and forth between these two choices. But what we did was randomly pick one of those beforehand that we're going to try and bias people towards. So, say I wanted you to say, "Yes, murder sometimes is justifiable." What the eye-tracker does is as you're looking back and forth between these two, it waits until you're looking at the one that we want to bias you towards. So, the system waits until you're looking at the word 'yes' and then suddenly, a thing pops up in the screen saying, "Decide now!" What we found is if you make that decision at the moment that we tell you to, at the moment that you're currently considering that choice, you're more likely to pick that choice.

Kat - How do you know that you are actually influencing someone's choices? They're not just picking the thing that they would've picked anyway, that you can actually change what someone's thinking.

Daniel - Well, that's a very good question and sometimes of course, maybe you do have a very strong belief about this particular moral question. So, what we do is we randomly choose these targets. We're going to bias you towards a yes, or a no for every person. We do this with lots of people on lots of different trials. So sometimes we try and bias you towards saying, "Yes, murder is justifiable." Sometimes we bias you towards saying, "No." For any one person, perhaps they did have a strong decision one way or another. When we look over lots of people we average, we find that on average, we get these increases on people picking the target that we bias them towards. So, we know that we're having this cause and effect on people's decisions because we can change them one way or the other, and we can do it at the toss of a coin.


Add a comment