SMI vs. EyeLink

If you have suggestions on how to improve this document, or find mistakes, please send them to ilslabs@nulluu.nl

Introduction

This how-to discusses when to use the SMI eye-tracker, and when to opt for the EyeLink.

Reading experiments

If you can run participants in the lab, you’ll want to use our SR Research EyeLink 1000 or babyEyeLink eye-tracker – it has a higher sampling rate than the SMI and is more precise.

Web/usability research

For web and usability experiments, use the SMI eye tracker (using their own software).

Infant experiments

Use the babyEyeLink.

Visual world paradigm

For VWP experiments, you can use either our Tobii eye tracker (using either the Studio or ClearView software), or the EyeLink, using ZEP. There are advantages and disadvantages to either set-up. A quick overview is given below; consult your eye-tracking lab manager before making a decision.

The SMI set-up: advantages and disadvantages

Advantages

• Portable.
• The SMI software supports more types of stimuli than ZEP does, for instance web stimuli. It can also record the user with a webcam.

Limitations

• Lower sample rate (250Hz – still good enough for most things).
• Measurements may be imprecise, as the software runs on Windows and Windows may decide that it has more important things to do than run your experiment. The data is fairly noisy due to the low sample rate.
• In practice, the SMI software  (needed for web usability experiments) is too slow and inaccurate to analyze decent-sized experiments. You can use Fixation instead, though. Bear in mind that no matter what, analysis will be a lot of work and require creativity and probably some programming on your part.

The EyeLink set-up: advantages and disadvantages

Advantages

• High sample rate (500Hz).
• Accurate measurements.

Limitations

• Less intuitive to use for most users because ZEP runs under Linux.
• Camera set-up and calibration are harder to learn (but easier to control).