Examples of Recent Applied Experimental Work 659Making computer technology accessible to people with
various types of disabling conditions represents a special chal-
lenge that has also motivated research (Elkind, Nickerson,
Van Cott, & Williges, 1995; Newell & Gregor, 1997). Experi-
mentation with natural language and speech for communicat-
ing with computer systems has been ongoing for several
years; these technologies are sufficiently mature that they are
beginning to be applied in practical situations (Makhoul,
Jelinek, Rabiner, Weinstein, & Zue, 1990; Ogden & Bernick,
1997). The research that has brought these technologies to
their current state of development has revealed much about
human language and speech understanding (see chapter by
Fowler in this volume).
The short history of computing technology has been one
of a steady increase in the amount of computing power that
can be packaged in a given space and that can be obtained for
a given cost. Although there are limits to what can be accom-
plished by advances in miniaturization, they have not yet
been realized. Already the state of the art provides people
with access (in a physical though not necessarily a psycho-
logical sense) to enormous amounts of information via the
Internet and the World Wide Web, and it makes possible the
embedding of computing power into the instruments and ob-
jects of everyday life. Research challenges for the future are
likely to have less to do with questions of the design of input-
output devices and more with questions of how to help peo-
ple interact effectively with extremely large information
repositories and with objects and environments that have in-
creasingly cognition-like capabilities (Nickerson, 1995).
Part-Task Training
Training of certain types—especially for tasks involving in-
teraction with complex machines that are costly to build and
operate, such as aircraft—is a very expensive undertaking.
For this reason there has long been interest in the possibility
of doing training of some aspects of such tasks with much
less costly devices. Whether such part-task training is effec-
tive in any particular case is an empirical question and is best
answered by experiment. Many years of research on the
topic have yielded mixed results (Lintern & Gopher, 1980;
Stammers, 1982; Wightman & Lintern, 1985).
Illustrative of recent work in the area is that of several in-
vestigators who have been successful in showing that prac-
tice with Space Fortress, a computer game that is intended to
capture some aspects of flying tasks, can facilitate subsequent
training of pilots of both fixed- and rotary-wing aircraft
(Gopher et al., 1994; Hart & Battiste, 1992). Space Fortress
was used in a coordinated set of studies sponsored by the
U.S. Advanced Research Projects Agency to investigate the
relative effectiveness of a variety of training strategies, most
of which involved part-task training. The composite task—
doing well at the Space Fortress game—was the same for all
participants, but the variety of training regimens used re-
flected experimenters’ differing ideas about how best to break
down the composite task and train people on the components.
The set of studies is described in a special volume of Acta
Psychologica(Donchin, Fabiani, & Sanders, 1989).
The use of simulation for training purposes constitutes a
part-task approach to training, inasmuch as any simulator
faithfully represents only some subset of the characteristics
of the real-world situation of interest. A great deal of experi-
mentation has been required to bring the state of the art of
simulation to the point where it can be the primary means of
training people to perform many complex tasks, piloting and
other aviation tasks being perhaps the most notable exam-
ples. How realistic a simulation must be in order to be effec-
tive for training purposes is a perennial question (Hays &
Singer, 1989), and the answer appears to depend on the
specifics of the task that must be learned.Aviation PsychologyAs we have already noted, many of the problems that en-
gaged experimental psychologists during World War II had to
do with military aviation. Much research continued this focus
after the war, but attention began to be given to problems
within commercial and civil aviation as well. Today the prob-
lems encountered in aviation psychology are considerably
broader in range than are those that occupied researchers in
the early days of the field. The development of multifunction
glass-cockpit displays—cathode ray tubes, liquid crystal
plasma displays—that have less resolution but much greater
flexibility than dedicated traditional instruments or paper
maps has raised a host of questions about how to make the
best use of the new technologies. Heads-up displays pro-
jected on an aircraft’s windscreen provide new challenges to
the visual system (Wickens & Long, 1995). They have re-
ceived extensive research attention in the aviation context
and are beginning to be examined for potential use in auto-
motive systems as well (Weintraub, 1992). There remain
unresolved questions regarding how best to match displays
to pilots’ preferred ways of conceptualizing an airspace
(Wickens & Prevett, 1995). Helmet-mounted displays are
also receiving attention from experimenters because of their
potential uses in aviation, especially in nighttime flight
(Seagull & Gopher, 1997).
Over the past 20 years, flying, especially of commercial
and military transport aircraft, has changed from being pre-
dominately a task of perceptual motor control to being one of