1
Cognitive Modeling offers Explanations for Effects found
in Usability Studies.
Sabine Prezenski Nele Russwinkel
Dep. of cognitive Modeling in dynamic HMS,
TU Berlin Marchstr. 23,
10587 Berlin, Germany
sabine.prezenski@tu-berlin.de
ABSTRACT
Two studies evaluate the usability of two versions of an
android shopping list application. ACT-R modeling
approaches and empirical findings are presented. It is
shown that semantic networks have a strong influence on
performance and learning. Effects of version updates are
discussed.
Author Keywords
Usability; cognitive modeling; ACT-R; application; mobile;
semantic network
ACM Classification Keywords
User centered design; ergonomics; theory and methods
design tools and techniques; human factors; human
Information processing; experimental design
INTRODUCTION
These days, life without mobile applications and smart
phones is hard to imagine. The market for applications is
growing rapidly [1]. For an application to be successful,
high usability is compulsory. Conventional usability testing
is time and money consuming. We therefore ask, how can
usability of applications be guaranteed without testing costs
exploding? The following paper argues that cognitive
modeling with ACT-R can serve as substitute for extensive
usability testing. We will present results of two studies of
an application to show how learning in applications
proceeds, why semantic knowledge is important and also
focus on version updates effects.
COGNITIVE MODELING & ACT-R
Cognitive architectures such as ACT-R [2] offer a
computable platform that represents well established
theories about human information processing. With
cognitive architectures it is possible to simulate cognitive
mechanisms and structures such as visual perception or
memory retrieval. ACT-R is a hybrid architecture, which
means that is has symbolic (knowledge representations such
as chunks and rules called productions) and sub symbolic
components (activation of chunks and utility of
productions). The structure of chunks is characterized by
different slots (or attributes), that can be filled with
information. Category membership is represented in slots;
this allows building semantic networks. Furthermore, new
chunks can be learned during a task. The production system
persists of rules defined by an “if” and “then” part. If the
cognitive system with its modules and chunks in the buffers
meet the conditions of the rule it can be selected. Then the
action part is executed.
USABILITY
Standard ISO 9241-11 defines usability as effectiveness,
efficiency and satisfaction. General ergonomic principles
for the design of dialogues between humans and
information system are specified in Standard ISO-9241-
110, seven criteria are outlined (suitability for the task,
suitability for learning, suitability for individualization,
conformity with user expectations, self descriptiveness,
controllability, and error tolerance).
Most usability criteria however can be assessed with
quantitative user tests. Suitability for learning can be
measured via comparison of several runs [3].
It is commonly agreed, that human knowledge is
represented in form of a semantic network [4]. Within these
net categories, associations with subcategories and retrieval
of subcategories succeed best and faster when the category
representations are addressed. To answer the question, what
is the best design for a menu structure, we have designed
two versions of an application. One version has two
subcategories (memory-like, Version A) with the
disadvantages of more required clicks; the other has only
one level of sub-categories (Version B) and therefore
requires fewer clicks.
Permission to make digital or hard copies of all or part of this work for
personal or classroom use is granted without fee provided that copies are
not made or distributed for profit or commercial advantage and that copies
bear this notice and the full citation on the first page. Copyrights for
components of this work owned by others than ACM must be honored.
Abstracting with credit is permitted. To copy otherwise, or republish, to
post on servers or to redistribute to lists, requires prior specific permission
and/or a fee. Request permissions from Permissions@acm.org.
ECCE '14, September 01 - 03, 2014, Vienna, Austria
© 2014 ACM. ISBN 978-1-4503-2874-6/14/09…$15.00
DOI: http://dx.doi.org/10.1145/2637248.2743001
Sabine Prezenski and Nele Russwinkel. 2014. Cognitive Modeling offers Explanations for Effects found in Usability Studies. In Proceedings of
the 2014 European Conference on Cognitive Ergonomics (ECCE '14). ACM, New York, NY, USA, Article 36 , 4 pages. This is a post-print
authors version. The original publication is available at the ACM Digital Library via DOI=http://dx.doi.org/10.1145/2637248.2743001