Please briefly describe the Semantic Feature-Comparison Model.

A modification of the hierarchical network model: it replaces hierarchical structure with semantic links; iii. Set-theoretic model; the basic semantic unit is still the concept; predicate-crossover model: a concrete example of the set-theoretic model, which is used to; validation and evaluation: it can account for category-size effects, but not for familiarity; iv. feature-comparison model; the idea that concepts are represented in long term memory by a set of attributes or features to characterize; semantic space: semantic features, especially defining features, that are ***identical*** between concepts; validation of the model: typicality effects, negative judgments; evaluation: can account for category size

Modification of the hierarchical network model: it replaces hierarchical structure with semantic links, and argues that concepts can be characterized either at the same level or not at the same level. The hierarchical network model was originally proposed for computer simulation, with a strictly logical nature, while the activation diffusion model is more suitable for people, with more flexibility to accommodate more uncertainty and ambiguity.

Three, set-theoretic models

The basic semantic unit remains the concept. Each concept is characterized by a set (set) of information or elements. This information can be categorized into sample sets and attribute sets or feature sets.

Predicate intersection model: is a specific instance of a set-theoretic model that is used to explain some allomorphic and allomorphic statements.

Validation and Evaluation: can account for category size effects, but not for familiarity and typicality effects. Presents the idea of nonpre-existence, where connections between concepts or certain knowledge need to be obtained by comparison or computation. Includes the possibility of more reasoning.

IV. Feature Comparison Model

Thinks that concepts are characterized by a set of attributes or features in long term memory. The various semantic features of a concept are divided into two categories: 1. Definitive features, which are necessary to define a concept. 2. Specific features, which are not necessary to define a concept, but have some descriptive function.

Semantic space: the more semantic features, especially defining features, that are ****similar between concepts, the more closely they are linked. Regardless of the category, the distance between any two points in space reflects the mental or semantic distance between two concepts. The closer the two points are, the closer the two concepts are. Subjects made ratings of inter-conceptual connections, relying on semantic features stored in their long term memory.

Validation of the model: typicality effect, negative judgment

Evaluation: can explain the category size effect, typicality effect, and negative judgment; explained by the similarity of semantic features, concise and effective.

Problems: how to distinguish between defining features and specificity features; there are counterexamples that question the model.

Chapter 5:

1. Why is it necessary to change the full-report method to the partial-report method in sensory memory research? What are the advantages of the latter?

The difference between the experimental design of the partial-report method and the full-report method is that:subjects were not asked to report all the letters, and the first, second, and third of the three rows of letters were paired with high, medium, and low cues, respectively, to tell subjects about this pairing in advance. After the cards were presented, a random cue was played and subjects were asked to report the letter in the corresponding row.

Results:Subjects were able to report all the letters in the rows matched to the cues with 100% recall. Since almost all of the letters in any row were reported, the subject must have memorized all three rows of letters. This indicates that the subject remembered all the letters. This is a partial inference of the whole, similar to a school test.

The experimental design of the partial reporting method showed that subjects did see more letters than they reported, but in the full reporting experiment, some of the letters were quickly forgotten when subjects reported them. Thus, through the partial reporting method, it was confirmed that there is indeed a sensory memory with a considerable capacity, and that the information is retained for an extremely short period of time and is quickly lost.

2. How to verify experimentally that short-term memory and long-term memory are two different memories (require experimental purpose, design, results, conclusions)? (Introduce two experiments)

(1), Baddeley and Warrington et al.'s (1970) comparison experiment Experimental purpose: to verify that short-term memory and long-term memory are two different systems. Experimental design: to compare the memory recall of normal people with that of amnesic patients:

Please adopt if you are satisfied