Question 1

Connectionist models are sometimes referred to as neural nets. This makes sense because:

Accepted Answer

Connectionist models are inspired by the structure and function of biological neural networks in the brain. Thus, they are often referred to as neural nets because they aim to replicate the behavior of neurons and their connections in the brain. Advocates of the connectionist approach argue that their models capture the key features of neural processing and thus make good biological sense.

Question 2

Which of the following is MOST important in allowing network models to represent specific knowledge, in contrast to more general knowledge?

Accepted Answer

Token nodes are crucial for representing specific knowledge as they denote individual instances, allowing the network to distinguish between different occurrences or examples within a category.

Question 3

Marni knows many facts about cowboy hats. However, while each of these facts is associated with cowboy hats, the various facts are not associated with one another. As a result, Marni may be less able to recall a particular fact about cowboy hats in comparison to someone who knows just that fact and nothing more. This is because:

Accepted Answer

The large degree of fan from Marni's cowboy hat node causes the activation to be spread too thinly over the numerous connected nodes for it to activate the node she needs. This is known as the fan effect, where the more nodes associated with a particular concept (in this case, cowboy hats), the longer it takes to activate a specific node within that concept network. As a result, Marni may have difficulty recalling a particular fact about cowboy hats, as the activation may be spread too thinly over the many associated nodes.

Question 4

Which of the following is a disadvantage of a winner-takes-all system?

Accepted Answer

In a winner-takes-all system, if a distracting node is strongly activated, it can inhibit the target node, preventing it from being activated, which is a disadvantage.

Question 5

Connectionist or parallel distributed processing (PDP) models differ from classic associative networks in that connectionism:

Accepted Answer

Connectionist or parallel distributed processing (PDP) models employ distributed processes, which means that information is processed in parallel across multiple nodes or units, rather than in a serial manner. This is different from classic associative networks, which rely on a central executive to coordinate processing and may use local representations rather than distributed ones.

Question 6

In a winner-takes-all system:

Accepted Answer

In a winner-takes-all system, a stronger node inhibits a weaker node, preventing the weaker node from activating its associated nodes.

Question 7

A search through memory can be launched in several different ways. Which of the following is NOT one of these ways?

Accepted Answer

The memory executive directing activation toward the address of the target material is not a recognized method of launching a search through memory. Memory retrieval typically involves activation spreading from nodes related to stimuli, ideas, or events, rather than a centralized executive directing the process.

Question 8

In connectionist networks, beliefs and memories are contained within:

Accepted Answer

In connectionist networks, beliefs and memories are stored in the pattern of connections between nodes, rather than in the nodes themselves or their current activation states.

Question 9

The term "connectionism" refers to:

Accepted Answer

Connectionism is a theoretical approach to understanding cognitive processes that involves the idea that complex phenomena are best explained by the parallel operation of many local representations (neurons) that are connected to one another. This approach is often contrasted with more symbolic computational models that involve serial processing of abstract symbols. Connectionism is often associated with artificial neural networks, which are models composed of simple processing units (artificial neurons) that are interconnected and can learn from experience.

Question 10

The tip-of-the-tongue (TOT) state can be accounted for if the incorrectly activated node inhibits the correct node. This is commonly known as:

Accepted Answer

The phenomenon of TOT state is explained by retrieval blockage, meaning the temporarily inability to retrieve a known word due to interference from competing words that were activated but not selected. This interference can inhibit the correct node, causing the TOT state. False activation and associative decline are related but not the exact explanation for TOT state. Memory inactivation is not a commonly used term in this context.

Question 11

In a system using distributed representations, each idea or mental content is represented by:

Accepted Answer

Distributed representations involve patterns of activation across many different nodes, rather than a single node or small group of nodes. This allows for more flexible and robust mental representations, as multiple patterns of activation can represent the same content and similar patterns can represent different contents.

Question 12

In Anderson's network account of propositional knowledge:

Accepted Answer

The answer of In Anderson's network account of propositional knowledge:
...

Question 13

In order for a connectionist or parallel distributed processing (PDP) network to function, each of the nodes must function:

Accepted Answer

Parallel processing is the key feature of connectionist or PDP networks, where all nodes perform computations simultaneously without any central control. This allows for massive information processing and pattern recognition capabilities.

Question 14

In connectionist theorizing, the strength of the association between two nodes is referred to as the association's:

Accepted Answer

In connectionist theorizing, the strength of the connection between two nodes is represented by the connection weight. The connection weight is a numerical value that reflects the strength of the association between two nodes.

Question 15

At this moment, Enriqueta is thinking about her garden. In network terms, this "thinking about garden" is directly reflected in:

Accepted Answer

In network terms, Enriqueta's "thinking about garden" can be represented by the activation levels of the nodes representing "garden". Activation level is the measure of the amount of information available in a particular node at any given time, and it is a direct reflection of the current state of the network. Therefore, the activation levels of the nodes representing "garden" would increase when Enriqueta is thinking about it.

Question 16

Which of the following is an advantage of a winner-takes-all system?

Accepted Answer

A winner-takes-all system provides a selective mechanism by allowing only the strongest signal to dominate, which helps in focusing attention and preventing distracting thoughts from triggering other distracting thoughts.

Question 17

Often mental processes (such as the hunt through memory or the search for a problem solution) must obey several requirements at the same time. According to connectionists, this requirement is achieved through:

Accepted Answer

Simultaneous multiple constraint satisfaction allows for considering and satisfying multiple requirements simultaneously, which is crucial for mental processes that involve multiple constraints. This approach is one of the strengths of connectionist models. Multiple distributed processing (A) and converging associations (D) are related concepts but do not directly address the issue of multiple constraints. The connection weight principle (C) is a general principle in connectionist models, but it does not specifically address the issue of multiple constraints in mental processes.

Question 18

In a connectionist model, "knowledge" is BEST defined as:

Accepted Answer

In a connectionist model, knowledge is represented by the potential for activation to flow in a certain way through the network, which is determined by the weights of the connections between nodes.

Question 19

If the associative links between all nodes were the same type, with no markers for how the nodes are related:

Accepted Answer

Without markers for how nodes are related, we would not be able to differentiate between a statement about someone owning a dog versus someone being a dog themselves, leading to potential confusion and erroneous associations.

Question 20

Which of the following is NOT a proposition?

Accepted Answer

"Too much candy" is not a complete statement that can be either true or false; it lacks a subject and a verb. The other options are complete statements that can either be true or false, and therefore are propositions.

Quiz 8: Associative Theories of Long-Term Memory