Cognition: Memory, Attention, and Learning - Foundations for Designing User-Centered Systems - page 136

Fig. 5.6 The conceptual
components of Wicken's
model of the user's
processing, modality, and
places for attention

Stages

Perception Cognition Responding

Spatial

Visual

Verbal

Auditory

is used (visual or auditory), response type (choice of response type, and execution of
that response, spatial or verbal), and the stages of processing (perception, cognition,
and responding). These define resources for processing and holding information,
with some tasks using more of one type of resource than another.
Most of the evidence to support Wickens' model of attentional resources comes
from studies of dual task performance. Evidence for two different types of
resources, for example, can be found by varying the difficulty of responding to one
task and looking at performance on a concurrent (more) perceptual task. The
performance on the perceptual task is (more or less) constant, even though more
resources are needed for the responding task.

5.3.2 An Information Processing Model of Attention

The ACT-R theory (Anderson and Lebiere 1998 ) incorporates a model of attention
that summarizes one of the common theories in this area, that of attention as a
spotlight in the mental world which is directed at what is being thought about.
Attention in ACT-R is represented as activation of concepts in its declarative
memories. People with more attention have more activation available (Lovett et al.
2000 ). This activation of memory creates the spotlight that focuses on the mental
objects that are being used in processing. These objects may include semantic
memories and episodic memories, as well as current goals and processes.
If more activation is available, this allows more objects to be manipulated at the
same time. This saves time re-retrieving them from memory and allows larger,
more complex objects to be created and used. Each of the objects has an associated
strength. Objects that are more familiar have a higher associated strength and, as a
result, need less activation to be matched against procedural knowledge than less
familiar objects.