Semiotics & Syringe Pumps

Semiotics can be a useful paradigm in HCI research and yet the cognitive process of semiosis is difficult to
uncover and empirically study. Thus in the domain of HCI semiotics has largely remained a descriptive theory,
able to provide a theoretical basis for the study of interfaces and interaction, but unable to produce empirical
data and generative research.
This work, made up of several studies, aims to investigates human errors motivated by the problems of medical
interfaces. It takes an empirical approach to investigate the interplay between semiotic signs and human error,
attempting to uncover how signs in the interface may affect the use of interactive devices.
Interfaces are created from signs, collections of symbols, icons and indices which form a semiotic scene, a
meaningful whole through which the user may interact with the underlying system. Therefore interaction with
an interface relies heavily on the process of semiosis.
The first study in this thesis was a questionnaire study looking at number pads as indurate signs for calcu-
lators and telephones. The questionnaire was designed to ascertain how users interpreted number-pads and
what features of the number-pad influenced this interpretation. We found that the layout of the numerical
buttons on a number-pad had little to do with how the number-pad was perceived, and that the users based
their assumptions about the use of the interface based entirely upon the extra contextualizing non-numerical
buttons.
The wish to use a semiotic paradigm in an empirical study demanded the exploration of a novel experimental
methodology. The next set of studies were experiments to see whether the interpretation of indurate signs
could be overcome under pressure. Thus we used a computer game based experiments as it was thought that
they would allow for the complete control and manipulation of signs within the experimental environment, and
encourage more natural semiosis that one might expect from participants in a real life task based explicit ex-
periment. In these studies it was found that under pressure participants fell back upon the culturally fossilized
meanings of the indurate signs they encountered, suggesting that indurate signs may cause misinterpretation
in human-machine interaction if used ineffectively.
Overall this thesis makes a contribution to semiotics by exploring the notion of indurate signs and how they
are interpreted, by investigating what features of common interfaces affect semiosis, and by attempting to
further the course of empirical semiotic studies. This thesis also contributes towards the use of computer
games as a research tool by charting the evolution of the game based experimental methodology over the
course of this thesis.