Information Technology Reference
In-Depth Information
Table 4.1
Some human sensory thresholds (under ideal conditions)
Sight
A candle flame seen from 50 km on a clear dark night (100 quanta to the eye, or
10 quanta absorbed by the rods)
The tick of a watch from 6 m in very quiet conditions (0.0002 dynes/cm 2 )
Sound
Taste
One gram of table salt in 500 L of water (0.0001 M)
One drop of perfume diffused throughout a three room apartment or 1 9 10 -12
Smell
mol/L of
ethyl merchantman
Touch
The wing of a bee falling on your cheek from a height of 1 cm (10 mg force)
thresholds under ideal conditions for the various senses (taken from Galanter 1962 ).
It should be noted that thresholds can vary across tasks and contexts. In a room full of
noisy machinery, for example, you may need to increase the volume of alarm
warning sounds, so that they can be heard by users.
The sensitivity of perception can be measured, based on how small a change
can be detected. This difference is called a just noticeable difference, or JND. In
vision, for example, it is how much brighter a scene has to be in order for someone
to report it as being brighter; and in hearing it is how much louder a sound must be
to be noticeably different. Although JNDs are objectively measured, their mag-
nitude is subjective, varying across users, tasks, contexts, and modality.
4.2.2 Habituation
All living things react to stimuli. If a stimulus occurs repeatedly and is not
regarded as salient they will habituate to it. People who live beside a railroad track,
for example, grow accustomed to the noise of the trains to a point where they no
longer notice it. In other words, people learn which stimuli are not salient (in the
current context), and hence do not require further processing. Habituation effec-
tively frees up cognitive resources, which allows people to use those resources to
deal with new stimuli as they are presented. Constant ''confirm action'' boxes, for
example, will become habituated to.
4.2.3 Signal Detection Theory (SDT)
Accuracy is an important aspect of performance. In many cases, simple measures of
correctness, such as the number of targets that were recognized and the number that
were missed, are sufficient. This is too simple a measure, however, when examining
performance under conditions where the ability to be correct is difficult, and the
types of mistakes are important. Table 4.2 provides a way of summarizing this more
complex situation: if you report seeing something when it is there, it is called a hit;
if you fail to report seeing something when it is there, it is called a miss. Similarly,
Search WWH ::




Custom Search