@phdthesis{Ungewiss2015, type = {Master Thesis}, author = {Judith Ungewiss}, title = {Parameters for Vigilance, Attention and Cognitive Workload within Eye Tracking Recordings}, url = {https://nbn-resolving.org/urn:nbn:de:bsz:944-opus4-1396}, pages = {XVIII, 95 S.}, year = {2015}, abstract = {Purpose: Usually, a theory of attention upon gazed-at locations is applied. More parameters than gaze location can be derived to improve the theory of attention allocation. The aim of this study was to identify parameters related to eye tracking, that are suitable indicators of attention. Methods: Binocular eye tracking data was collected with the Eye Tribe tracking system (The Eye Tribe Aps, Copenhagen, Denmark) for the task of visual exploration of the painting “Unexpected Visitors” by Ilya Repin. 20 subjects (valid data: 19/20) had to look at this painting for about two and a half minutes in order to generate fatigue and inattention. In a second step, suitable parameters of attention were transferred to a data set (8 subjects, valid data: 6/8) on a perimetric task executed with the OCTOPUS 900 perimeter (Haag-Streit, K{\"o}niz, Switzerland). Monocular parameters could be applied on the perimetric task, the error rate (false positive and false negative catch trials, 5 \% each) were taken as additional parameter. Results: For the image viewing task, the only parameter showing significance was the average level (a10) of fatigue waves (p = 0.00024, ANOVA, ∆ = -0.8316 px). Blink duration (∆ =-270.4 ms), pupil variability (∆ = -0.17868), saccade length (∆ =-0.3135 px) and fixation duration (∆ = 186.5 ms) did not change significantly, but showed relevant trends by differences ∆ of their median between the first and last tenth of the recording time. Blink rate and the Index of Cognitive Activity (ICA) did neither show significant changes nor relevant trends. Vergence accuracy failed to indicate fatigue due to variability between subjects and comparatively small effect size. For the perimetric data, in 3 of 6 subjects fatigue waves over a limited time window could be observed. Only for one subject, a relevant increase in false negative responses to catch trials (50 percentage points) could be observed. Conclusion: Pupil diameter variability, saccade length, fixation duration and fatigue waves were the parameters indicating fatigue. Only the latter parameter has the potential to be applied to perimetric data.}, language = {en} }