Response Time and Display Rate

1
Chapter 10

• Response Time and Display Rate
• February 3, 2004

2
10.1 Introduction

• Response time seconds from user ativity to
  computer response
• User think time seconds during which user
  thinks before action
• Display rate characters per second of display
• Must balance fast response with accuracy

3
10.2 Memory

• 72 chunks
• Working memory is used to generate and implement
  solutions
• More efficient if you can cluster into larger
  chunks
• Complex problems often dealt with by developing
  higher-level concepts that group several lower
  level concepts into a single chunk
• Easy to disrupt short term memory (highly
  volatile)

4
Error

• Long delays can be distacting and lead to wasted
  effort
• Short delays can lead to ill thought out plans
• There might be a preferred response time to
  maxmize performace

5
Optimum Problem Solving

• Driving analogy
• Graphical indicators make happy users
• Users have knowledge
• The solution plan has little delay
• Distractions are low
• Anxiety is low
• Feedback about progress
• Errors can be avoided and handeled easily

6
10.3 Expectations and Attitudes

• How long will users wait for the computer to
  respond before they become annoyed???
• For many tasks 2 seconds, sometimes within 0.1
  second- turning the wheel of a car- pressing a
  key on the keyboard, piano or telephone 30
  seconds, red traffic light to turn green, two
  days for a letter to arrive or a month for
  flowers to grow

7

• If the task is completed much more quickly or
  much more slower than expected, users become
  concerned or frustrated. Even though people can
  detect 8 changes in 2-4 second response time
  users do not become concerned until the change is
  much greater.
• Individuals tolerance for delays also impacts
  the response-time expectations- personality,
  costs, age, mood, time of the day , pressure to
  complete the work, .

8

• Less than 100 milliseconds wait user
  controlled three-dimensional animations, flight
  simulations, graphic design, .
• Three primary factors influences users
  expectations
• 1. Previous experiences
• 2. Individual personality differences
• 3. Task differences

9

• Three conjectures emerge
• Individual differences are large and users are
  adaptive. work faster as they gain experience.
• For repetitive tasks, users prefer and will work
  more rapidly with short response times.
• For complex tasks, users can adopt to work with
  slow responses time.

10
10.4 User Productivity

• Shorter system response time usually lead to
  higher productivity. But sometimes working too
  quickly may lead to errors that reduce
  productivity.
• Just like driving there is no general rules ,
  just as a new route evaluation must be done for
  each trip

11
Repetitive Tasks

• The nature of the task has a strong influence on
  whether changes in response time alter user
  productivity.
• Operators may learn to use the system more
  quickly with short system response times because
  they can explore alternatives more easily.

12
Problem Solving Tasks

• Users will adapt their work style to the response
  time.
• Study shows that the time to solution was
  invariant with respect to response time!

13
Summary

• User pick up the pace of the system to work more
  quickly with shorter response time and that they
  consistently prefer a faster pace.
• The optimal response time may be longer than the
  minimum possible response time.

14
10.5 Variability

• Predictability peace of mind
• Insurance industry based on the idea that most
  people are willing to sacrifice a reduction of
  immediate pleasures in exchange for peace of mind
• Extreme variation in response time should be
  prevented or addressed by the system (i.e.
  through some progress indicator)
• As variability increases, performance may
  decrease
• Dissatisfaction is most noticeable if delays are
  unusually long (i.e. 2x anticipated response time)

15
10.6 Practitioners Summary

• Computer system response time closely linked with
  user productivity, error rates, satisfaction,
  working style, etc (refer to Box 10.1, pg. 367)
• In general, user satisfaction is inversely
  proportional to response time
• Optimal response time can be established by
  measuring ? productivity associated with cost of
  errors
• More informative the feedback the better

16
10.7 Researchers Agenda

• Study error rates as a function of response time
  for a range of tasks and users?
• How else are user work styles affected, other
  than through response time?
• Effectiveness of diversionary tasks vs. progress
  reports when technical feasibility prevents short
  response?
• Strive for specification of upper/lower limits
  for response time for respective commands