This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Exploring Eye Tracking for Games User Research! A case study of lessons learned
THQ Inc. (NASDAQ: THQI) is a leading worldwide developer and publisher of interactive entertainment software. The company develops its products for all popular game systems, personal computers and wireless devices. Headquartered in Los Angeles County, California, THQ sells product through its global network of offices located in North America, Europe and Asia Pacific.
Issues With Think Aloud for Player Research in 3D Games
Many gameplay issues are related to player awareness and understanding of in-game cues and objects
However, player observation and think-aloud techniques do not always provide adequate information regarding player awareness and understanding of these elements
Players may notice details but fail to understand or act on the information (making observation difficult)
Players may not mention details they noticed while they play
Players may not remember details if they are queried after play
Take a 3D game that has already had been tested with Think Aloud Protocol and rerun the study with Eye-Tracking
Gather additional insights into player gaze and awareness, such as: What objects in the 3D world are noticed? When are they noticed? Do
players comment on or remember these objects? What objects in the 3D world are missed? What elements in the HUD are noticed? At what points are they noticed?
Do players comment on or remember these HUD elements? What elements in the HUD are missed? Do overall gaze patterns change in different areas of the game?
You need to network the eye-tracker to a separate PC or gaming console.
Pre-release games may not have been tuned for performance, so high-end hardware is typically required. If testing on a PC, consider: CPU – A high-end recent CPU (e.g. Intel Core 2 Quad) Video Card – Dedicated video card (laptops with integrated video or low-
performance cards will likely not be sufficient) Video output – make sure that your hardware supports the necessary game
output (e.g. HDMI, VGA, etc.) and is compatible with the eye-tracker hardware
In a game study, some players may not volunteer much information using just a Think Aloud.
The game itself may demand their attention at times, and describing the experience can introduce additional cognitive load or interruptions to game play
Our test scenarios included a natural stopping point midway through the study
Our suggestion: Use ‘debrief’ rather than Think Aloud because: Ability to observe what the participants would naturally do
without disrupting how they would play if they were on their own Adding Think Aloud to this experience adds more cognitive load
Clearly stated test goals and plans are important for any domain
Fleeting game events and unique game terminology makes prior planning and preparation even more important
Our suggestion: review test goals with project sponsors in front of the game Game events may appear quickly – observers need to know what to
look for and at what time Game terminology can often be arcane and very specific. For
example, an observer may need to know the visual differences between: Space Marines and Chaos Marines M-16s and AK-47s Atomic Drops and a Facebreaker DDTs
Fixations are generally an inadequate way of measuring subject attention when viewing dynamic media. This is because a subject, when tracking a moving target (like a
soccer ball), is actually performing smooth pursuit. This is similar to fixation, at least cognitively, but it is not detectable via normal fixation detection algorithms.
Instead, we recommend dwell time. This is the summary of time spent within an area whether in fixation, saccade, or smooth pursuit, and is a better way to quantify attention on a moving object.
Glance time is a related and equally useful measure.
Dwell Time: Dwell time starts at the moment the AOI is fixated and ends at the moment the last fixation on the AOI ends(equal to the sum of durations from all fixations and saccades that hit the AOI)
Glance Duration = Saccade duration for entering the object + sum of all fixation durations and saccade durations before the eyes begin to leave the AOI(equal to the dwell time + duration of saccade entering AOI)