Abstract
Architectural spaces such as station environment and underground paths which are concealed from the physical information of open air space such as rows and corners along the street, according to the restrictions caused by the functional requirements, building and street networks are getting complex and intensive. These spaces are required signs and marks shown in the sign boards and maps for guiding, conducting and positioning. Although signs and maps are expressing in many ways, signs by letters are most frequently used, because they show the contents most clearly. So important signs such as "EMERGENCY EXIT" are increasing in numbers in order to attract attention, and other information from advertisement also trace the same way of presentation. This causes the increase of both quality and quantity of signs. Accordingly visual perception forces the accepting of excess information, and the complexity of signs causes refusal and mistake. In this paper, from the functional aspect of visual search tasks, sign complexity in the station emvironment and the under ground paths were investigated. Based on the result of the previous report in which probability characters during visual search tasks were reported, the way visual sense reacts to the elements which constitute signs was analysed by the preliminary experiment, and as a result the fact that they react to the contrast between elements was introduced. So by hypothesizing the probability of variable, processing time which is the scale showing the difficult level of the tasks is argued. That is, first of all, physical elements of signs in the models (signs subtended angle, angle between horizontal line of sight to the center of environmental representation, the oriented angle of the surface of a sign, reciprocals of subtended angles between the center of a given sign and each of the center of all other signs) are measured. And with these measured scales, informatinal amounts of each element of signs are to be searched. Second, the theoretical model of the processing time based on the empirical model introduced by analysing the relation between experimental processing time and information are proposed. Also by the correlation analysis between order of fixation and theoretical processing time, the hypothesis of the probability of variables during search process is concluded to be well fitted to the visual search tasks.
