Page 386 -

376 CHAPTER 13 Measuring the human

MEASURING WORKLOAD—CONT'D
Despite their wide acceptance, these instruments suffer from the same
shortcomings as other surveys. Asking users to rate the workload after they
have completed a task relies on fallible human memory, leading to potentially
inconsistent assessments that fail to account for much of the nuanced workload
requirements inherent in many complex tasks.
These shortcomings have led to the development of a variety of
approaches for using physiological sensors to measure workload. One
possible approach involves the use of eye-gaze tracking to measure pupil
diameter, which has been shown to increase with stress or frustration
(Barreto et al., 2008; Klingner et al., 2008; Jiang et al., 2014). Links between
pupil dilation and mental load have been used to explore user interactions
in contexts such as web content, where relevant content has been associated
with larger pupil dilation than less relevant content, indicating that more
mental effort is involved when content is pertinent (Gwizdka and Zhang,
2015). Other efforts have looked at the use of microsaccades and saccadic
intrusions—deviations from a gaze point followed by a short fixation and
then a return to the original point—to derive similar measures (Tokuda et al.,
2009, 2011).
Other physiological measures—many of which are discussed in
this chapter—have also been used to assess workload. One 2010 study
investigated the utility of several simultaneous measures, including an eye
tracker, an electrocardiogram armband, a wireless electroencephalogram
headset, and a heart-rate monitor, along with NASA-TLX ratings, to
determine which combination of signals best measured workload. The
average of the heat flux (as measured by the armband) and the variability
of the electrocardiogram provided the highest classification accuracy
(Haapalainen et al., 2010), suggesting that combinations of measurements
may be useful in measuring complex phenomena such as workload.
Alternative approaches to assessing workload through physiological signals,
including more direct measures of brain activity, are discussed later in the
chapter.

13.3 MOTION AND POSITION TRACKING
If the study of the motions of our eyes can provide insights into attention, work-
load, and other important processes, what else can we learn from the human
body? Human bodies are constantly moving: even when we are “sitting still,”
our torsos move slightly with each breath. Movements of our hands, arms, heads,
torsos, and even legs and feet can be measured by multiple types of sensors,
providing useful opportunities for studying and changing how we interact with
computers.