Deck 8: Symbolic Models of the Mind and Neural Networks

Question 1

In which of the following ways has artificial intelligence been described?

A) simulation of intelligent behavior using computers
B) a model of the human brain which supersedes its capabilities
C) a branch of computer science that tries to make computers smarter
D) a clarification of questions about human cognitive processes

Answer 1

a model of the human brain which supersedes its capabilities

Question 2

Which of the following represents correspondence in structure between humans and computers?

A) wetware / hardware
B) hardware / software
C) nervous system / software
D) peripherals / brain

Answer 2

wetware / hardware

Question 3

Which of the following represents correspondence in function between humans and computers?

A) stimuli / keyboard
B) sense organs / software
C) wetware / software
D) cognitive acts / software

Answer 3

cognitive acts / software

Question 4

The human brain exceeds the computer in which of the following ways?

A) speed of processing information
B) retrieval of information
C) information storage
D) speed of mathematical calculations

Question 5

In comparing the human brain to a computer, Allman concludes

A) though what has been learned has been limited, that knowledge has been extremely valuable.
B) the comparison is a poor one and unlikely to be productive since there is a fundamental difference between the two.
C) a great deal of understanding has been gained with the promise of more to come in the years ahead.
D) though what has been learned has been limited, a continued investigation has great promise for what might be learned.

Question 6

A parallel distributed processing computer (PDP) would seem to be useful in understanding the functioning of the human brain for which of the following reasons?

A) It is a more advanced version of the serial processing digital computer.
B) It has been developed specifically for this purpose.
C) The brain functions more like a PDP computer which performs functions sequentially and with extreme speed.
D) The brain functions more like a PDP computer which performs numerous functions simultaneously.

Question 7

A symbolic model used for computer-based study of human cognitive processes is described by which of the following?

A) All information can be represented in symbols.
B) Learning is implicit.
C) Information processing involves the synthesis of experience.
D) Most information can be represented in symbols.

Question 8

An algorithm, as a problem-solving procedure, may be described by which of the following?

A) the most common subtype of a heuristic approach
B) used only recently in the creation of programs
C) the systematic consideration of alternatives
D) easy application due to the minimal impact on memory storage

Question 9

A heuristic approach to problem solving may be described by which of the following?

A) a more fully developed algorithmic approach
B) involves strategies which eliminate and select among alternatives
C) responsible for the overwhelming success in competition between early chess programs over Chess Masters
D) makes use of sheer brute force in consideration of alternatives

Question 10

Which of the following approaches to learning would be most compatible with a computer-as-cognitive processor metaphor?

A) Skinner's operant conditioning
B) Hebb's psychoneurological approach
C) the Gestalt approach
D) Pavlov's classical conditioning

Question 11

Which of the following would not be associated as a strength of human cognitive processing as opposed to computers?

A) spreadsheets
B) recognizing anger
C) perception of cognitive meaning
D) perception of complex geometric shapes

Question 12

Chess Masters and computers have which of the following in common with regard to chess strategy?

A) They both follow explicit rules.
B) They both employ a formal logic leading to a single correct solution.
C) Neither anticipate consequences of strategies.
D) They both follow implicit rules.

Question 13

Which of the following does not characterize an instance of learning?

A) implicit, non-verbalized rules
B) conscious and rule driven
C) explicit and unconscious
D) explicit and symbol manipulating

Question 14

Which of the following is true with regard to learning and awareness?

A) Learning involves awareness by definition.
B) One may be unaware of what has been learned.
C) Learning may not occur without awareness.
D) Lack of awareness would negate the usefulness of that learned.

Question 15

If a learner cannot explicitly identify rules of learning, and are not conscious of that which has been learned, the following is true.

A) Learning may only take place at a conscious level.
B) Learning that has taken place may be observed in behavior.
C) A conclusion about learning is not possible.
D) A learning disability exists.

Question 16

The most important determinant of whether learning will occur with a parallel distributed processing model is

A) the central processing unit.
B) the set of individual processing units.
C) a majority of the individual processing units.
D) the pattern of connections among separate processing units.

Question 17

The processing units in a PDP computer may be compared with

A) possessing several brains.
B) human sense organs.
C) components of a neural network.
D) mathematical equations.

Question 18

Which of the following is an assumption of a connectionist model?

A) Rules and symbols are used in all human cognitive processing.
B) In learning, rules must be both explicit and conscious.
C) A serial digital computer is an improvement over a PDP computer.
D) The "fuzzy logic" used by humans at times may be modeled.

Question 19

Neural network models

A) are designed to mimic those of the brain's neurons.
B) possess the exact complexity of the human brain.
C) always use rules of logic.
D) are not altered by experience.

Question 20

In which of the following ways is a connectionist model an accurate representation of human thinking?

A) Its approach is always systematic and rational.
B) Use is made of both logic and insight.
C) The sole approach to problem solving is always "fuzzy logic."
D) The rules of logic are always used.

Question 21

The following conclusion may be reached regarding connectionist models.

A) They are less descriptive of learning than symbolic models.
B) They successfully achieve their purpose-the elaboration of symbolic models.
C) They are not only able to learn, they do so like humans.
D) It is questionable whether either connectionist models or symbolic models can be made to learn like humans.

Question 22

The Turing Test provides a good example of

A) the difference between identifying words (decoding) and understanding (comprehension).
B) a computer model that demonstrates that computers can read.
C) a computer can think.
D) the fact that a computer itself is intelligent.

Question 23

Though sometimes referred to as a myth, computers are indeed dependent on (slaves to) their programmers. This is confirmed by

A) connectionist models.
B) symbolic models.
C) the combination of connectionist and symbolic models.
D) It has not been confirmed.

Question 24

To the extent that a computer is capable of using algorithms, the following is true with regard to a computer/Chess Master chess match:

A) The Chess Master will win the match.
B) The Chess Master will lose the match.
C) Computers will sometimes win, and the Chess Master will sometimes win.
D) No prediction can be made with regard to outcome.

Question 25

"Black Box" implies

A) one reaches a dead end.
B) a controlled environment.
C) a maze.
D) the contents of the mind are unknown.

Question 26

Computer simulation is used to

A) improve computer technology.
B) better understand cognitive outcomes.
C) focus on human cognitive process.
D) devise a system that can accomplish the same things as humans.

Question 27

If a machine produces correct responses for complex problems, then

A) it understands those problems.
B) it has the potential to understand the problem.
C) it understands the responses.
D) it produces correct responses to problems.

Question 28

A serial computer

A) carries out a number of acts simultaneously.
B) was based on a connectionist model.
C) does parallel processing.
D) views intellectual function as a sequence of facts.

Question 29

Viewing computer capability as cognitive brute force is

A) accurate.
B) not presently accurate but soon will be.
C) not entirely correct.
D) only true of a PDP model.

Question 30

Walking is an example of

A) implicit learning.
B) symbolic learning.
C) latent learning.
D) explicit learning.

Question 31

Like humans, neural networks can

A) make inferences after being given rules for doing so.
B) make inferences without being given specific rules for doing so.
C) make inferences, but not accurate ones.
D) not make inferences.

Question 32

A type of processing that allows computers to program themselves and respond in unpredictable ways is called

A) artificial intelligence.
B) parallel distributed processing.
C) symbolic model processing.
D) reductio ad absurdum

Question 33

What kind of computer might most closely resemble the human brain?

A) a PC computer
B) an Apple computer
C) an analog computer
D) a digital computer

Question 34

A connectionist model assumes that

A) all information is represented in symbols.
B) learning may be implicit.
C) thinking does not follow identifiable rules.
D) all logic is "fuzzy logic."

Question 35

Neural network models are different from the conventional computer in that

A) neural networks can make inferences without being given specific rules.
B) they are symbol-based models.
C) they need specific rules in order to make inferences.
D) they are perfect analogies for human thought processes.