Analyzing the perception of water surfaces in urban landscapes using eye tracking

Authors and Affiliations: 

Lien Dupont, Ghent University, Department of Geography, Ghent, Belgium

Johan Pihel, Swedish University of Agricultural Sciences, Department of Landscape Architecture, Alnarp, Sweden

Asa Ode Sang, Swedish University of Agricultural Sciences, Department of Landscape Architecture, Alnarp, Sweden

Veerle Van Eetvelde, Ghent University, Department of Geography, Ghent, Belgium



Landscape is an important public interest as it determines a part of the quality of life for people everywhere. In urban landscapes, rivers, canals, ponds and other water surfaces contribute to the ecological value of the city. However, little is known about how the public perceives these landscape features and how they visually observe this ‘urban ecology’, while this perception could be valuable for urban planning and design. A way to objectively measure how people observe landscapes is offered by the technology of eye tracking. By continuously recording the point-of-regard and the eye movements of an observer while looking at an image, eye tracking makes it possible to reconstruct the whole scan path made by the observer. From these data, the objects which drew much or little attention can be extracted and the general observation pattern can be analyzed. In our study, we use landscape photographs, which have been demonstrated to be valid surrogates for in situ landscapes, as a basis for an eye tracking experiment. In particular, a set of photographs of urban landscapes, in which water surfaces like rivers, canals or ponds are present, are tested to determine to which degree water attracts people’s attention. For comparison, photograph simulations in which these water surfaces were replaced by another land use, are tested as well. Furthermore, we analyze if the degree to which the surrounding landscape is reflected in the water surface influences if and how people observe these water bodies. Photographs containing water surfaces with clear reflections, vague reflections and without any reflection of the surrounding landscape are therefore included in the eye tracking experiment. Finally, we also investigate if the landscape type, in which the water surface is situated, has an effect on the view pattern. To this end, observation patterns made in photographs taken in a rural landscape are compared to the ones occurring in images representing a more urban context. The eye tracking experiment was executed using a non-portable iView X RED eye tracking system with a sample rate of 120Hz meaning that the point-of-regard is recorded 120 times per second. The participants group consists of graduate geographers and landscape architects. During the test, the participants were instructed to attentively observe 63 landscape photographs, without executing specific search tasks. The results of the experiment are statistically analyzed and visualised in two-coloured heat maps. The results of this study may be helpful in how to visually integrate ecological water surfaces in urban areas. In general, eye tracking can be an innovative approach to design urban spaces in such a way that new designs, which increase the ecological value of the city, are also effectively seen by the public. This, in turn, might raise the awareness of the importance of such open spaces.


Convention, Florence, 20.X.2000, ETS No.176.

De Lucio, J. V., Mohamadian, M., Ruiz, J. P., Banayas, J., Bernaldez, F. G., 1996. Visual Landscape Exploration as revealed by Eye Movement Tracking. Landscape and Urban Planning, 34, 135-142

Dupont, L., Antrop, M., Van Eetvelde, V. (2013). Eye tracking analysis in landscape perception research: influence of photograph properties and landscape characteristics. Landscape Research, in press.

Holmqvist, K., Nyström, M., Andersson, R., Dewhurst, R., Jarodzka, H., Van De Weijer, J. (2011) Eye tracking: A comprehensive guide to methods and measures (Oxford: Oxford University Press).

Sevenant, M., Antrop, M. (2011). Landscape Representation Validity: A Comparison between On-site Observations and Photographs with Different Angles of View. Landscape Research, 36 (3), 363-385

Tveit, M., Ode, A., Fry, G. (2006) Key concepts in a framework for analyzing visual landscape character, Landscape Research, 31, pp.229-255.