658 Psychological Experimentation Addressing Practical Concerns
proceedings. What factors contribute to the accuracy (or
inaccuracy) of such testimony? How is the accuracy of testi-
mony influenced by methods of interrogation? What makes
eyewitness or earwitness testimony more or less credible to ju-
rors? What special considerations are necessary when the eye-
witness or earwitness is a young child, and especially when
the child is the alleged victim of abuse? These and many re-
lated questions have been subjects of experimental research.
Lineup procedures have been the focus of many studies
(R. C. L. Lindsay & Wells, 1980; Malpass & Devine, 1984;
Wells & Lindsay, 1980). One question that has received at-
tention is whether sequential lineups are more or less effec-
tive than simultaneous lineups; sequential lineups appear to
be superior to simultaneous lineups at least in the sense that
they are less likely to yield false identifications (R. C. L.
Lindsay & Wells, 1985). How the confidence with which
identifications are made relates to the accuracy of those iden-
tifications has been another question of interest; overconfi-
dence is not an unusual finding (Juslin, Olsson, & Winman,
1996; Loftus, Donders, Hoffman, & Schooler, 1989; Wells &
Bradfield, 1998), and some data show that confidence may
increase as a result of interrogation without a corresponding
increase in accuracy (Shaw, 1996). A related question has to
do with the degree to which the confidence expressed by a
witness determines the credence that is given by jurors to the
witness’s testimony; it appears that more confident witnesses
tend to be seen as more credible (Cutler, Penrod, & Stuve,
1988; R. C. L. Lindsay, Wells, & O’Connor, 1989).
The reliability of testimony of very young children (Ceci &
Bruck, 1993, 1995; Dent & Flin, 1992; Poole & Lindsay, 2001)
and of very elderly people (Bornstein, 1995; Yarmey, 1984;
Yarmey & Kent, 1980) has been studied experimentally. Ex-
perimentation has shown that having children draw pictures
relating to experiences, especially emotional experiences, can
facilitate their verbal recall of those experiences (Butler,
Gross, & Hayne, 1995; Gross & Hayne, 1998, 1999). Espe-
cially relevant to the assessment of the reliability of testimony
is the finding of the possibility of eliciting “memories” of
events in one’s past that did not occur (Hyman & Kleinknecht,
1999; Loftus, 1997; Pezdek, Finger, & Hodge, 1997). Such re-
sults are especially relevant to reports by adults of having re-
covered lost memories of molestation or other forms of abuse
as children. The problem of suggestibility more generally has
motivated some experimentation (Gudjonsson, 1992; D. S.
Lindsay, 1990; Tomes & Katz, 1997), as has interest in the ef-
fects of sleep deprivation on suggestibility in interrogation
procedures (Blagrove, 1996; Blagrove & Akehurst, 2000).
A topic closely related to eyewitness testimony is that of
face recognition, which has also been the focus of much
experimentation. How reliable is the recognition of faces
captured by a high-quality video camera relative to that of
faces in photographs? Some work suggests that recognition
based on video shots is not very reliable (Bruce et al., 1999;
Henderson, Bruce, & Burton, 2001)—an important finding in
view of the widespread use of closed-circuit TV systems for
security surveillance.Human-Computer InteractionThere are few, if any, areas that have stimulated more experi-
mental work in recent years than that of HCI. Interest has
grown sufficiently rapidly to have stimulated the establish-
ment of several new journals focused on the subject. Topics
investigated within this domain include e-mail and other
computer-mediated human communication (Kiesler, Siegel, &
McGuire, 1984; Kiesler, Zubrow, Moses, & Geller, 1985),
computer-supported work by groups or teams (special issues
ofHuman-Computer Interaction,1992, andInteracting with
Computers,1992; Sproull & Kiesler, 1991), interface design
(Fisher, Yungkurth, & Moss, 1990; Norman, 1991; Paap &
Roske-Hofstrand, 1986), and a host of others (Helander,
Landauer, & Prabhu, 1997).
Work in this general area has been spurred by a rapid in-
crease in the number of people who use computers more or
less daily for professional or personal purposes. The first
heavy users of computers, during the middle of the twentieth
century, were for the most part technically oriented people.
Many of them were working on the development of com-
puter technology itself or were specialists who were applying
it to computationally intensive tasks. With the production of
affordable desktop computers and the proliferation of com-
puter networks, more and more people who were not trained
in computer science or related technical areas became com-
puter users, and the need for the design of interfaces and soft-
ware with their requirements in mind became increasingly
important.
Much of the early experimental work focused on the de-
sign of input-output devices. Efforts to design keyboards that
improve on the standard QWERTY layout predate modern
computer technology by many years, but the proliferation of
computer users for whom the keyboard is the main input
device has increased interest not only in the possibility of al-
ternative key arrangements but in other aspects of keyboard
design (e.g., split keyboards and chord keyboards; Lewis,
Potosnak, & Magyar, 1997). Questions of what should ap-
pear on a visual interface and how the display should be laid
out motivated much experimentation on the design of option
menus and icons (Norman, 1991; Paap & Cooke, 1997) and
on the management of objects that sometimes are (at least
partially) visible and sometimes not (Marcus, 1997).