Response Time and Display Rate
Introduction
Time is precious. When externally imposed delays impede progress on a
task, many people become frustrated, annoyed, and eventually angry. Lengthy
or unexpected system response times and slow display rates produce these
reactions from computer users, leading to more frequent errors and lower
satisfaction. Some users accept the situation with a shrug of their shoulders,
but most users prefer to work more quickly than the computer allows.
There is also a danger in working too quickly. As users pick up the pace
of a rapid interaction sequence, they may learn less, read with lower comprehension,
make more ill-considered decisions, and commit more data-entry errors. Stress
can build in this situation if errors are hard to recover from, or if they
destroy data, damage equipment, or imperil human life (for example, in air-traffic
control or medical systems) (Emurian, 1989; Kuhman, 1989).
The computer system's response time is the number of seconds it takes
from the moment users initiate an activity (usually by pressing an ENTER
key or mouse button) until the computer begins to present results on the
display or printer (Figure 10.1). When the response is completely displayed,
users begin formulating the next action. The user think time is
the number of seconds during which users think before entering the next
action. In this simple model, users initiate, wait for the computer to respond,
watch while the results appear, think for a while, and initiate again.
In a more realistic model (Figure 10.2), users plan while reading results,
while typing, and while the computer is generating results or retrieving
information across the network. Most people will use whatever time they
have to plan ahead; thus, precise measurements of user think time are difficult
to obtain. The computer's response is usually more precisely defined and
measurable, but there are problems here as well. Some systems respond with
distracting messages, informative feedback, or a simple prompt immediately
after a command is initiated, but actual results may not appear for a few
seconds.
Designers who specify response times and display rates in humanp;computer
interactions have to consider the complex interaction of technical feasibility,
costs, task complexity, user expectations, speed of task performance, error
rates, and error-handling procedures. These decisions are further complicated
by the influence of personality differences, time of day, fatigue, familiarity
with computers, experience with the task, and motivation (Carbonell et al.,
1968; Shneiderman, 1980).
Although some people are content with a slower system for some tasks, the
overwhelming majority prefer rapid interactions. Overall productivity depends
not only on the speed of the system, but also on the rate of human error
and the ease of recovery from those errors. Lengthy response times (longer
than 15 seconds) are generally detrimental to productivity, increasing error
rates and decreasing satisfaction. More rapid interactions (less than 1
second) are generally preferred and can increase productivity, but may increase
errors for complex tasks. The high cost of providing rapid response times
or display rates and the loss from increased errors must be evaluated in
choosing an optimum pace.
For alphanumeric displays, the display rate is the speed, in characters
per second (cps), at which characters appear for the user to read. The rate
may be limited by inexpensive modems to 120-1200 cps (characters per second),
or the display may fill instantaneously (typical for many personal computers
and workstations). In World Wide Web applications, the display rate of a
page may be limited by network transmission speed or server performance.
Portions of images or fragments of a page may appear with interspersed delays
of several seconds. Display rates for graphics are measured in bytes/second
Reading textual information from a screen is a challenging cognitive and
perceptual task-it is more difficult than reading from a book. If the display
rate can be made so fast that the screen appears to fill instantly (beyond
the speed at which someone might feel compelled to keep up), subjects seem
to relax, to pace themselves, and to work productively. Since users often
scan a web page looking for highlights or links, rather than reading the
full text, it is useful to display text first, leaving space for the graphical
elements that are slower to display.
This review begins with a model of short-term human memory and the sources
of human error (Section 10.2). Section 10.3 focuses on the role of users'
expectations and attitudes in shaping their subjective reactions to the
computer-system response time. Section 10.4 concentrates on productivity
as a function of response time; Section 10.5 reviews the research on the
influence of variable response times.
 |
Please send
comments and suggestions to the Booksite
Director
Last Updated:
11 December 2002
|
