CH9 Flashcards
Mental workload:
the information
processing demands imposed by the performance of cognitive tasks. Workload depends on
the match between task requirements and capacities, and different people have different
capacities.
Effort:
conscious exertion
Arousal:
physiological sense of readiness for activity.
Resources:
hypothetical processing reserves or sensory, motor, or information processing capacities.
Sometimes these concepts are interrelated. Too much or too little workload impairs
performance, so a balance needs to be found.
Kahneman: the arousal of an individual
determines
the capacity available to engage in different activities.
An indicator of stress is
strategy adjustment:
when demands are high, we switch to using less effortful strategies for
performing a task. An example is attentional narrowing: you restrict your attention to an
inappropriately small set of displays or information sources. Fatigue after-effects: after a
fatiguing task, people show a preference for using low-cost strategies for performing other,
unrelated tasks.
attentional narrowing:
you restrict your attention to an
inappropriately small set of displays or information sources.
Fatigue after-effects:
after a
fatiguing task, people show a preference for using low-cost strategies for performing other,
unrelated tasks.
Difficulty intensitivity is
that allocating more resources to one of two tasks
does not always hurt performance on the other one.
Structural alteration is
changing the
structure of a task without changing its difficulty.
Perfect time-sharing is
when two tasks
shown to interfere with other tasks can nonetheless be performed together without
decrement.
multiple-resources concept (MRC)
(different sorts of tasks draw on separate resources)
Difficulty intensitivity is that allocating more resources to one of two tasks
does not always hurt performance on the other one.
Structural alteration is changing the
structure of a task without changing its difficulty.
Perfect time-sharing is when two tasks
shown to interfere with other tasks can nonetheless be performed together without
decrement.
These three are all evidence for the multiple-resources concept. The MRC is
inadequate as a model of attention but it can serve as a useful framework for predicting
operator performance.
Strategy shifts can be used as
an indicator of
changes in operator mental workload. The skill level of the performer has a large influence
on the methods or strategies employed.
indicator of changes in
operator mental workload. Large influence of skill level and processing resource demands.
Three classes of techniques have been developed to measure mental workload:
physiological, performance-based, and subjective measures.
Physiological measures of mental workload are
direct measurements of various body
or brain responses to task performance. You can either measure general arousal, or
specific brain activity.
Underlying assumption: various bodily systems are activated whenever the demand for mental effort increases. Pupil diameter (pupillometry) and
heart-rate variability are the most suited for examining changes in mental workload.
Event-related potentials have been shown to be effective indicators of mental load,
and can also show which of two tasks is seen as more important. ERPs give an
indication of more specific processing activity. The P3 is linked to workload and it has
a big amplitude when an unexpected stimulus occurs. Measuring the amplitude of
ERP components may be a good indicator of workload even when the event used to
elicit the ERP is totally irrelevant to the task.
Primary-task technique:
simply measuring some aspects of task performance. When
difficulty is below a certain level, primary-task performance will not show any
decrements. This is why it’s better to use a secondary-task methodology, in which a
second task is performed at the same time as the primary task, and the extent to
which performance on one of the tasks suffers is measured. The purpose is to use up
any capacity that might be available so that changes in the difficulty of the primary
task will be reflected in performance changes. In a loading-task paradigm, the focus
is placed on the secondary task, and different aspects of primary-task performance
are measured. In a subsidiary-task paradigm, the emphasis is placed on the primary
task, and performance on the secondary task is measured. Embedded secondary
tasks are tasks that can occur naturally in the task environment under consideration,
but are not part of the primary-task performance
secondary-task methodology
a second task is performed at the same time as the primary task, and the extent to
which performance on one of the tasks suffers is measured.
In a loading-task paradigm, the focus
is
placed on the secondary task, and different aspects of primary-task performance
are measured.
In a subsidiary-task paradigm, the emphasis is placed on
the primary
task, and performance on the secondary task is measured.
Embedded secondary
tasks are
tasks that can occur naturally in the task environment under consideration,
but are not part of the primary-task performance
Performance operating characteristics:
performance on one task is plotted against
performance on a second task. The more difficult a task, the harder it will be to
combine with a second task and the more performance will suffer to the extent that
resources are withdrawn from it and allocated to the other task. Performance at
different allocation policies (e.g. devote 50% to task A and 50% to task B) will fall
along a curve, the POC.
Subjective measures, in which people report how much workload they experience,
are based on judgements and they’re influenced by the range and frequency of
different possible stimuli. Ratings of a task with a relatively restricted range of
difficulty conditions will tend to overestimate workload whereas a high range of task
difficulties will result in an underestimation of task workload. This can be solved by
showing all possible conditions in a practice session. Some researchers have
proposed that unidimensional methods are just as appropriate as multidimensional
method. Dual-task performance is determined by a number of factors, including the
resources invested in the task and competition for common resources.
A workload measure must be
reliable, easy to use, and sensitive.
A measure is diagnostic if
it not only reflects changes in load, but also discriminates between the amounts of load
imposed on various operator capacities or resources, such as perceptual versus memory
load.
