Page 149 - DCAP310_INTRODUCTION_TO_ARTIFICIAL_INTELLIGENCE_AND_EXPERT_SYSTEMS

Unit 7: Probabilistic Reasoning

in the regularization of images (i.e., images are represented as more like pure abstract geometric Notes
images, though they are irregular in shape). There are several ways that humans form and use
cognitive maps. Visual intake is a key part of mapping. The first is by using landmarks. This is
where a person uses a mental image to estimate a relationship, usually distance, between two
objects. Second, is route-road knowledge, and this is generally developed after a person has
performed a task and is relaying the information of that task to another person. Third, is survey.
A person estimates a distance based on a mental image that, to them, might appear like an actual
map. This image is generally created when a person’s brain begins making image corrections.
These are presented in five ways:

1. Right-angle bias is when a person straightens out an image, like mapping an intersection,
and begins to give everything 90-degree angles, when in reality it may not be that way.
2. Symmetry heuristic is when people tend to think of shapes, or buildings, as being more
symmetrical than they really are.
3. Rotation heuristic is when a person takes a naturally (realistically) distorted image and
straightens it out for their mental image.
4. Alignment heuristic is similar to the pervious, where people align objects mentally to
make them straighter than they really are.

5. Relative-position heuristic people do not accurately distance landmarks in their mental
image based on how well they remember that particular item.

Did u know? Another method of creating cognitive maps is by means of auditory intake
based on verbal descriptions. Using the mapping based from a person’s visual intake,
another person can create a mental image, such as directions to a certain location.

7.4.1 Problems with Current Approaches to Uncertainty

One of the main reasons for the problems with current methods of environmental management
is scientific uncertainty not just its existence, but the radically different expectations and modes
of operation that scientists and policymakers have developed to deal with it. To solve this
problem, these differences must be exposed and understood, and better methods to incorporate
uncertainty into policymaking and environmental management must be designed. To
understand the scope of the problem, it is necessary to differentiate between risk, which is an
event with a known probability (sometimes referred to as statistical uncertainty), and true
uncertainty, which is an event with an unknown probability (sometimes referred to as
indeterminacy). For instance, every time you drive your car, you run the risk of having an
accident because the probability of car accidents is known with very high certainty. The risk
involved in driving is well known because there have been many car accidents with which to
calculate the probabilities. These probabilities are known with enough certainty that they are
used by insurance companies, for instance, to set rates that will assure those companies of a
certain profit.
There is little uncertainty about the possibility of car accidents. If you live near the disposal sight
of some newly synthesized toxic chemical, however, your health may be in jeopardy, but no one
knows to what extent. Because no one knows the probability of your getting cancer, for instance,
or some other disease from this exposure, there is true uncertainty. Most important environmental
problems suffer from true uncertainty, not merely risk. Uncertainty may be thought of as a
continuum ranging from zero for certain information to intermediate levels for information
with statistical uncertainty and known probabilities (risk) to high levels for information with

LOVELY PROFESSIONAL UNIVERSITY 143