ICC 2013 Pre-Conference Workshop - Eye tracking: why, when, and how?
Two types of participants took part in the user study: experts and novices. The participants of the expert group were at the moment of the study employed at the Department of Geography at Ghent University. They obtained at least a Master degree in Geography or Geomatics and received, both theoretical and practical, cartographic training. In their daily job they use paper and digital maps on a regular basis. Consequently, the expertise of this group is two-fold. On the one hand, they have a substantial level of background knowledge of cartographic syntax and semiotics, on the other hand, they are highly experienced in working with paper and digital maps. The group of novices were Bachelor students of the Faculty of Psychology and Educational Sciences at Ghent University. None of them received any previous training in cartography. The group of expert users counted 14 participants, whereas the group of novices 15. The data of one novice participant was omitted because this participant was familiar with a number of regions in the stimuli. All participants co-operated on a voluntary basis and were Dutch-speaking (the language in which the test was presented).
- Stimuli & assignment
Twenty demo-maps, with a basic and controlled design, were presented to each participant in a random order on a screen. The background of all demo-maps is identical: three polygons depicted in a pastel colour. The foreground of the demo-maps is only populated with point objects and their associated labels.
Two types of demo-maps were used in the user study, which used a different label placement algorithm. In the 'Total'-type maps, all labels are positioned in their most optimal position, also after the interaction. In the 'Border-type' maps, not all labels are placed in their most optimal position. Only labels that will become unreadable after an interaction (because they will be cutt-off at the border of the image) will be relocated.
Each demo-map was depicted for 50s, after which a pan-operation, over a fixed distance to the right, was simulated. The duration of this simulated pan-operation was 1s (from 50-51s).
On the initial view, the participants had to locate five names, which were depicted on the right side of the map. They could choose in which order they wanted to locate these names. During the simulated interaction, the list with five names also changed: three names remained in a list (but in a different order) and two were new names. After the interaction, the users had to locate these names again on the map image. The trail ended when the participants had located the latter five names.This assignment corresponds to a visual search for points of interest on a map.
An example of a demo-map is presented below (view before and after the simulated interaction)
The study was conducted in a controlled environment: the Eye Tracking Laboratory in the Faculty of Psychology and Educational Sciences (Ghent University). An EyeLink 1000 eye tracking device (SR Research, Ontario, Canada) was used to record the participant’s eye movements during the study. This system can sample the participant’s Point of Regard once every millisecond. The stimuli were presented on a 21 inch monitor with a resolution of 1280 by 1024 pixels.
More information regarding the specifications of this equipment and software can be found here: http://www.sr-research.com/EL_1000.html
The provided dataset lists the fixations which were identified by and exported from the DataViewer software package (SR Research). Each line in the csv-file corresponds to a single fixation and the following characteristics are available:
- trID - trial ID: a combination of the code for the person and the map;
- trNr - trial number: Ta unique number for each trial. Novices have a trial number 1xxx; experts have a trial number 2xxx;
- pIdx - point ID: a unique number for every fixation with one trial;
- X: x-position of the fixation (in screen coordinates, pixels);
- Y: y-positon of the fixation (in screen coordinates, pixels);
- time: timestamp of the fixation (in ms);
- partCode: code of the participant (novices: P1-P15 without P7| experts: P1-P14)
- partType: novice or expert; -map: demo-map that was used (name of the region);
- mapType: type of map that was used (Total or Border)
Description data challenges
- Identify patterns in the users' search behaviour over time
- Are there differences in this due to expertise, label placement, before/after the interaction?
- Are participants who search for the five names in a random order more effcicient than participants who search in order?
- Is the Border-label placement more efficient than the Total-label placement?
- Can participant locate the three labels that remained in the list after the interaction more efficiently?
- How does the layout of the map (point objects & labels) influence the users' search behaviour?
Link to related articles
- Ooms, K., De Maeyer, P., Fack, V., Van Assche, E., & Witlox, F. (2012a). Interpreting maps through the eyes of expert and novice users. International Journal of Geographical Information Science, 26(10), 1773-1788. (http://www.tandfonline.com/doi/abs/10.1080/13658816.2011.642801)
- Ooms, K., Andrienko, G., Andrienko, N., De Maeyer, P., & Fack, V. (2012b). Analysing the spatial dimension of eye movement data using a visual analytic approach. Expert Systems With Applications, 39(1), 1324-1332. (http://www.sciencedirect.com/science/article/pii/S0957417411011249)
- Ooms, K., De Maeyer, P., Fack, V., Van Assche, E., & Witlox, F. (2012c). Investigation the effectiveness of an efficient label placement method using eye movement data. The Cartographic Journal, 49(3), 234-246. (http://www.ingentaconnect.com/content/maney/caj/2012/00000049/00000003/art00005)
- The eye movement data can be downloaded here
- The stimuli can be downloaded here:
Two other datasets are distributed: