Page 360 -

350 CHAPTER 12 Automated data collection methods

FITTS' LAW, CHILDREN, AND MOUSE CONTROL—CONT'D
Paul M. Fitts conducted pioneering experiments in this area in the
1950s, leading to the development of Fitts' law, a frequently cited result
2
in HCI research. Originally intended as investigations of the theoretical
limits of human performance in performing tasks of differing amplitudes
of movement, Fitts' experiments involved asking participants to move
between two targets separated by a distance. Fitts found that the information
content of the task was determined by the distance between the targets
and the inverse of the width of the targets (Fitts, 1954). This result was
later generalized to expressive movement time as being a function of
the logarithm of the ratio of the movement amplitude to the target width
(MacKenzie, 1992).
Fitts' law tells us that as the distance between targets increases, or the size
of the targets decreases, the time required to move between them increases.
This has a certain intuitive appeal: it is harder to reach small targets than it
is to reach larger targets, just as we can cover short distances more quickly
than we can cover long distances. As much of our interaction with computers
involves target selection, Fitts' law can help us understand the impact of
design decisions regarding the placement and sizing of icons on a screen or
keys on a keyboard.
Fitts' law is important enough to have spawned follow-on works, with
researchers examining a wide variety of variations on the original task
(MacKenzie, 1992; MacKenzie and Buxton, 1992). Extensions and novel
applications of have confirmed the relevance of Fitts' law to the use of mobile
devices while walking (Lin et al., 2007); developed models for “two-thumb”
text entry on small keyboards (Clarkson et al., 2007); proposed extensions for
nonrectangular targets (Grossman et al., 2007); explored implications for novel
input modalities including multitouch devices (Nguyen et al., 2014) and flexible
displays (Burstyn et al., 2016); added an extra dimension for virtual reality
and 3D displays (Lubos et al., 2014; Teather and Stuerzlinger, 2014; Janzen
et al., 2016); and even extended Fitts’ law to foot input for under-desk devices
(Velloso et al., 2015).
Juan-Pablo Hourcade and his colleagues at the University of Maryland
faced this problem in the course of their work with young children. Faced
with 5-year-old children who had difficulty clicking on computer icons,
they set out to understand how preschool children differed from young
adults in their ability to complete target-selection tasks (Hourcade et al.,
2004). Although several researchers had conducted Fitts' law research with
young children, none had specifically addressed the question of whether
performance differences justified the effort required to build interfaces
specifically for this class of young users.

2
Authors of books on HCI research are contractually obligated to refer to Fitts' law at least once.