Towards a semio-cognitive theory of Human-Computer Interaction
CHI '01 extended abstracts on Human factors in computing systems - CHI '01
A B S T R A C T The research here presented is theoretical and introduces a critical analysis of instrumental approaches in Human-Computer Interaction (HCI). From a semiotic point of view interfaces are not "natural" or "neutral" instruments, but rather complex sense production devices. Interaction, in other words, is far f o m being a "transparent" process. In this abstract we present the fundaments of a theoretical model that combines Semiotics with Cognitive Science approaches. K e y w o r d
... s Interfaces, Human-Computer Interaction, semiotics, transparency I N T R O D U C T I O N Along with the diffusion of user-friendly interfaces a new conception of HCI became popular amongst designers and researchers: the interface is an "instrument" or "tool" that must "disappear" during interaction. User-centered design evangelists have expanded this perception to the last consequence: the best interface is the "invisible" one, i.e. the interaction device that users "don't feel" and allows them to concentrate on accomplishing their task . This principle is valid for any interface: interaction devices of videogames, word-processors or web sites must disappear when the user interacts. Only in cases of system breakdown the user will "feel" the interface . An invisible interface is the utopia for any interface designer. But what's good for a designer -and specially for the user-is not necessarily good for a researcher. The apparent naturalization of interactive processes should not imply -as often occurs in instrumentalist approaches-a loss of theoretical complexity: even the most simple example of HCI (like clicking an icon or activating a "roll-over" device) involves an intricate combination of perceptive and interpretative operations. Users, not researchers, need an non-visible technology. Then, to design "invisible" interfaces first we must understand the interaction process and bring out its internal sense production machine. Human-computer interfaces, are not neutral or impartial territories like any other place where semiotic and cognitive processes manifest. The purpose of our research is to dismount the myth of interface transparency, identifying the complex semiotic and cognitive processes behind HCI. S e m i o t i c s and c o g n i t i o n Why a "semio-cognitive" approach and not just a "cognitive" theory? Because interaction is such a complex and uncertain process that one theory is not enough. HCI is a multidisciphnary field where different approaches can (and should) be appfied. Semiotic is not just the "science of signs" but a theory about "sense production and interpretation". It does not involve real people but operates with virtual models and sense production/interpretation strategies. If we consider HCI a semiotic process, the research must focus on the ambiguous game of sense production and interpretation that involves designers and users. The main purpose of semiotic research applied to HCI is not to provide guidelines for interface designers but to create theoretical models of interactive processes. Certainly, semiotic research could provide designers with many models and tips that would optimize the hard art of designing user interfaces.