Introduction to eye-tracking
Eye tracking is a great tool to study cognition. It is especially suitable for developmental studies, as infants and young children might have advanced cognitive abilities, but little chances to show them (they cannot talk!).
This tutorial will teach you all you need to navigate the huge and often confusing eye-tracking world. First, we will introduce how an experiment can (and should) be built, explaining how to easily record eye-tracking data from Python. Then, it will focus on how to analyse the data, reducing the seemingly overwhelming amount of rows and columns in a few variables of interest (such as saccadic latency, looking time, or pupil dilation).
How to build an eye-tracking experiment
What do you want to measure?
Looking time
It is much easier to start an eye-tracking project if you have a clear idea of what you want to measure. Classic paradigms on infant research rely on looking time (How long are infants attending a given stimulus?) and are often called Violation-of-Expectation tasks. They familiarize infants with a given stimulus or situation (e.g. a cat) and, after a given number of presentations (e.g., 8), they present a different stimulus (e.g., a dog). Do infants look longer at the dog, compared to the cat? If so, they were able to spot that something changed.
Beware! This does not mean that they can distinguish cats and dogs, but more simply that they can spot any difference between the two images. A careful control of the stimuli should be in place if we want to make strong conclusions from looking time.
Saccadic Latency
Another very popular eye-tracking measure is saccadic latency. It measures how quickly infants can direct their gaze onto a stimulus, once it is presented on screen. This is a great measure of learning because infants will be faster at looking at a stimulus if they expect it to appear in a given position on the screen. They can even be so fast that they anticipate the correct location of the stimulus, even before the stimulus appears! This is called an anticipatory look.
Beware! Saccadic latencies are not a perfect measure of learning. Infants might be faster at looking at something just because they are more interested (pick interesting stimuli to keep them engaged!), and they might become slower due to boredom or fatigue (introduce variation in the stimuli, so that they become less boring over time!).
Pupillometry
The fanciest eye-tracking measure is pupil dilation. It allows us to measure arousal (How interested is the infant in the stimulus?), cognitive effort (How difficult is the task?), and - my special favourite - uncertainty (Does the infant know what will happen next?). However, its fame comes at a price: It is not only the fanciest, but also the most persnickety…
Beware! Stimuli should be carefully designed, controlling their luminance (not too high, and ALWAYS the same) to avoid that task-unrelated variations in light will affect your measurements; They have to be presented in the same location on the screen, as pupil size changes depending on screen location; Pupil dilation is very slow, so the stimulus presentation also has to be slow; Often, a fixation cross has to precede the moment in which pupil dilation is measured, so that the pupil size can return to baseline before the event you care about happens. But if you feel confident about your paradigm, go for it!!
Here is a visual summary of what you can measure:
