Bollettino del CIC, Numero 3, Anno 1995

Eleonora Bilotta, Pietro Pantano, Centro Interdipartimentale della Comunicazione, Università degli Studi della Calabria, Arcavacata di Rende (CS) Italia

Abstract: In this work the authors present a semiotic model for analysing and classifying icons. This model is based upon three steps:

1) the icon is analysed in relation to itself;

2) the icon is analysed in relation to the object it represents;

3) the icon is analysed in relation to the human interpreter.

The first step is carried out through a perceptual, formal and gestalt analysis. The second step, which refers to the object the icon represents, is carried out through hermeneutic, semantical, functional and communicative analysis. The third step is carried out through a cognitive, experiential, emotional analysis to determine the user's psychological profile.

Keyword: Icon, semiotic approach, GUI

1. Introduction

The graphical design of a user interface presents a very difficult problem when designers utilise icons.They have to be put in communicative environments and usually icons have precise meanings from a syntactical and semantical point of view on one hand. These meanings deal with the real functions within multimedia page . On the other hand icons have functions in the process of human computer interaction. Like natural languages, graphic interfaces have visual sentences which say things about the world; what a visual sentence says about the world (its information domain) is a sum of the icons it contains and the structures into which these are combined. The contents of these visual words in a sentence denote objects, properties, relationships, or sets of these.

Good interaction with an interface requires that users are able to translate the common every day meaning of the icons into the metaphorical and ergonomic meaning [1] which icons have in that context . In this way users must learn conventional meanings of that particular domain, building up a visual language, with a classification of functions, useful for understanding and interacting, according to their intended aims.

To-day many icon collections exist on the World Wide Web (For example, a metaindex is available at http://www.webcom.com/~webcom/power/icons.html) that interface designers use in their work. Too little attention is devoted to icon classification which is useful to define standards in the computer environments. Some attempts of classification are at the following addresses on the World Wide Web. The first project "Graphical Standard IEC Complaint 417" "specifies the basis for the creation of graphical symbols, including their shape and size, and instructions for their applications" (the on-line address is the following http://www.hike.te.chibau.ac.jp/iec417/index.html).

The second project "Tecfa" presents a formal analysis of the interaction among icons within the same multimedia page and a semantic analysis of the iconís formal characteristics (the on-line address is the following http://tecfa.unige.ch/).

Other studies are "The on-line visual literature" (http://www.pomona.claremont.edu/visual-lit/intro/intro.html) and the CHI 95 tutorial on Icons by William Horton [6].

The present authors retain that none of these studies explains the complex relationship between user and the icons in a graphical interface.

2. The Model

In this work the authors present a semiotic model for analysing and classifying icons [7] [5]. This model is based upon three steps:

1) the icon is analysed in relation to itself;

2) the icon is analysed in relation to the object it represents;

3) the icon is analysed in relation to the human interpreter.

Fig. 1 The semiotic model

The first step is carried out through a perceptual, formal and gestalt analysis.

With regard to the perceptual aspects Selfridge's theory of "characteristics" [8] was utilized. According to this author the elaboration of visual material is organised in a hierarchical way, with a first level that verifies the presence, in the input, of various characteristics; and with a second level which activates specified recognition analysers, particular to those specific characteristics.

Formal analysis is carried out, according to the basic elements of visual communication, considering the underlying structures like dot, line, direction, hue, texture, shape, scale, dimension and motion [3]. These structures let us to divide an icon into its components parts; while gestalt analysis allows one to put these parts together again, arriving at the re-construction of the whole icon, in a process that is not derived from the sum of the single structures, but is a constructive process of the mind [4].

The second step, which refers to the object the icon represents, is carried out through hermeneutic, semantical, functional and communicative analysis. Hermeneutic analysis considers the interpretation of icons into the socio-cultural contexts in which they can be found or utilised. This analysis "emphasises those areas of human experience where individual interpretation and intuitive understanding play a central role" [9]. Semantical analysis deals with the relation of an icon (and of an object) to all the meanings it carries, defining not only the ares the object refers to (a network of meanings), but also the possibility of creating a shared-universe of conventional meanings, which can be utilised by a group of users in that particular domain [1]. Like natural languages, icons give the user the possibility to communicate with a large number of alternative modes of behaviour. A visual interface permits users to select one of these alternatives (through a process of decision-making) and to inter-act with the object/icon that represents the function a user wants to activate. At this stage it is possible to list all the functions the interface makes available and all the communicative behaviours an icon provides the users with.

The third step is carried out through a cognitive analysis of the learning mechanisms the user utilises when he/she is interacting with the interface and with a set of functions, meanings, behaviours that an icon allows. Learning does not depend on the retention of an iconís formal structures, but on the user's ability to create links between his mind and the external world which is organised in the interface; on the user's ability to create (given certain conditions) behaviours that satisfy his own aims. It is therefore necessary to establish: a) the user's experience of the icon in that particular context; b) what his emotional responses in relation to that icon are; c) what his aims are (pragmatics) and d) the user's psychological profile.

3. Conclusions

To summarise this model allows one to determine:

1) perceptual-formal-gestalt requirements that an icon must have from a graphical-representative point of view;

2) a non-ambiguous representation and planning of the functions an icon carries in its role within the interface. This helps clarify its meaning and the communication context into which it is inserted;

3) how an icon can be understood from a cognitive and emotional point of view in relation to the aims the users have in mind when they interact with a computer .

The model allows one to build up graphical interfaces with the possibility of creating a shared-universe of conventional meanings and a better human-computer interaction. At the present time we are utilising this model to create a large data base of icons. We plan to realise some interfaces with these icons and make test-beds to evaluate their usability.

References

1. Bertacchini P.A., E. Bilotta e P. Pantano (1993). Contributi per la definizione di un modello in Ergonomia. In Quinto Congresso Nazionale "Ergonomia e Progetto" della Societa' Italiana di Ergonomia (SIE), Palermo

2. Bilotta E., M. Fiorito and P. Pantano (1995). Visual Interface analysis through an ergonomic model. In Sixth International Conference on Human Computer Interaction Yokohama, Giappone.

3. Dondis D. O. (1973). A primer of visual literacy. Cambridge (Mass.), The MIT Press.

4. Dunker K. (1929). A source book of Gestalt Psychology . New York, Humanities Press.

5. Morris C. (1938). Foundations of the theory of signs. Chicago. University of Chicago Press.

6. Horton W. (1995). Designing Icons and Visual Symbols , in Proceedings of the ACM Conference on Human Factors in Computing Systems.

7. Peirce Ch. S. (1931-35). Collected Papers. Cambridge (Mass.), Harvard University Press.

8. Selfridge O. G. (1959). Pandemonium: A paradigm for learning. In The mechanisation of thought processes. London, H. M. Stationery Office.

9. Winograd T. and F. Flores (1986). Understanding computers and cognition: A new foundation for design, Ablex, Norwood, N. J.