Page 392 -

382 CHAPTER 13 Measuring the human

Approaches to collecting data from various parts of the body require different
classes of sensor for measuring responses. Broadly speaking, these sensors fall into
two classes: electrodes, which directly record electrical signals, and transducers,
which convert mechanical or physical measurements into an electrical form (Stern
et al., 2001). In both cases, the resulting analog signals are converted to digital form
by an analog-to-digital converter and stored on computers for filtering and analysis.
Complex physiological responses to different stimuli can make interpretation a chal-
lenge: there is no single, monolithic interpretation of these signals. Although measure-
ments of heart rate, electric conductance of skin, respiration, or brain activity may be
well-defined in terms of the underlying mechanical or biological activity, the meaning
of those phenomena may be much harder to interpret. If an activity causes a person's
heart rate to increase and changes activation patterns of different areas in their brain, is
that because the task was hard? Establishing links between these physiological methods
and concepts of interest to HCI researchers is often difficult. Understanding the limits
of any particular measurements, and any debates over the interpretation of those mea-
surements, is critical for conducting reliable and valid research with physiological data.
Although some of these issues are discussed later, careful researchers will dive into
more recent work in these rapidly evolving areas before rushing into conduct studies.
The sources of physiological data that have been used in HCI research can be
classified according to the type of signal involved, the location on the body, and the
kinds of sensors required (see Table 13.1). The range of data sources and their ap-
plications are likely to continue to expand as researchers find creative applications
for new and evolving technologies.

Table 13.1 Types of Physiological Data Used in HCI Research
Signal Possible
Data Source Technique Type Locations Sensors

Electrodermal Galvanic skin response Electrical Fingers, Surface
activity (GSR) (Scheirer et al., 2002; toes electrodes
Mandryk and Inkpen, 2004)
Cardiovascular Blood-volume pressure Light Finger Surface
data (Scheirer et al., 2002) absorption electrodes
Electrocardiography Electrical Chest, Surface
(Mandryk and Inkpen, 2004) abdomen electrodes
Respiration Chest contraction and Physical Thorax Stress
expansion (Mandryk and sensor
Inkpen, 2004)
Muscular Pressure or position sensing Physical or Varied Pressure
and skeletal (Brady et al., 2005; Dunne electrical sensor,
positioning et al., 2006a,b; Dunne and fiber optics,
Smyth, 2007) others
Muscle Electromyography (Mandryk Electrical Jaw, face Surface
tension and Inkpen, 2004) electrodes
Brain activity Electroencephalography Electrical Head Electrodes
(Lee and Tan, 2006) in helmet
Evoked responses (Stern Electrical Head Surface
et al., 2001) electrodes