Real-time sharing of gaze data between multiple eye trackers: evaluation, tools, and advice

Research output: Contribution to journalArticle

Standard

Real-time sharing of gaze data between multiple eye trackers : evaluation, tools, and advice. / Nyström, Marcus; Niehorster, Diederick C; Cornelissen, Tim; Garde, Henrik.

In: Behavior Research Methods, Vol. 49, No. 4, 08.2017, p. 1310-1322.

Research output: Contribution to journalArticle

Harvard

APA

CBE

MLA

Vancouver

Author

RIS

TY - JOUR

T1 - Real-time sharing of gaze data between multiple eye trackers

T2 - evaluation, tools, and advice

AU - Nyström, Marcus

AU - Niehorster, Diederick C

AU - Cornelissen, Tim

AU - Garde, Henrik

PY - 2017/8

Y1 - 2017/8

N2 - Technological advancements in combination with significant reductions in price have made it practically feasible to run experiments with multiple eye trackers. This enables new types of experiments with simultaneous recordings of eye movement data from several participants, which is of interest for researchers in, e.g., social and educational psychology. The Lund University Humanities Laboratory recently acquired 25 remote eye trackers, which are connected over a local wireless network. As a first step toward running experiments with this setup, demanding situations with real time sharing of gaze data were investigated in terms of network performance as well as clock and screen synchronization. Results show that data can be shared with a sufficiently low packet loss (0.1 %) and latency (M = 3 ms, M A D = 2 ms) across 8 eye trackers at a rate of 60 Hz. For a similar performance using 24 computers, the send rate needs to be reduced to 20 Hz. To help researchers conduct similar measurements on their own multi-eye-tracker setup, open source software written in Python and PsychoPy are provided. Part of the software contains a minimal working example to help researchers kick-start experiments with two or more eye trackers.

AB - Technological advancements in combination with significant reductions in price have made it practically feasible to run experiments with multiple eye trackers. This enables new types of experiments with simultaneous recordings of eye movement data from several participants, which is of interest for researchers in, e.g., social and educational psychology. The Lund University Humanities Laboratory recently acquired 25 remote eye trackers, which are connected over a local wireless network. As a first step toward running experiments with this setup, demanding situations with real time sharing of gaze data were investigated in terms of network performance as well as clock and screen synchronization. Results show that data can be shared with a sufficiently low packet loss (0.1 %) and latency (M = 3 ms, M A D = 2 ms) across 8 eye trackers at a rate of 60 Hz. For a similar performance using 24 computers, the send rate needs to be reduced to 20 Hz. To help researchers conduct similar measurements on their own multi-eye-tracker setup, open source software written in Python and PsychoPy are provided. Part of the software contains a minimal working example to help researchers kick-start experiments with two or more eye trackers.

U2 - 10.3758/s13428-016-0806-1

DO - 10.3758/s13428-016-0806-1

M3 - Article

VL - 49

SP - 1310

EP - 1322

JO - Behavior Research Methods & Instrumentation

JF - Behavior Research Methods & Instrumentation

SN - 1554-351X

IS - 4

ER -