Lesson 8 - Questions Flashcards
Define “semantic memory,” and provide some examples. How is it related to top-down processing? Define “categories” and “concepts.” Explain how categories and concepts are related to semantic memory.
Comment: As indicated in the overview to this unit, the idea of defining a concept by necessary features is also known as the classical view. Although certain human-made concepts can be defined by a set of necessary and sufficient features (e.g., a square), many natural concepts cannot be so defined. Consider the concept of “game.” What are the defining features of this concept? The famous philosopher Wittgenstein (1953) posed this question, and concluded that natural categories could not be defined by a set of necessary features. In the quote below, Wittgenstein considers whether the concept of game has a defining feature or a feature common to all games.
Consider for example the proceedings we call “games.” I mean board-games, card-games, ball-games, Olympic games, and so on. What is common to them all?—Don’t say: “There must be something common, or they would not be called ‘games’”—but look and see whether there is anything common to all.—For if you look at them you will not see something that is common to all, but similarities, relationships, and a whole series of them at that. To repeat: don’t think, but look!—Look for example at board-games, with their multifarious relationships. Now pass to card-games; here you find many correspondences with the first group, but many common features drop out, and others appear. When we pass next to ball-games, much that is common is retained, but much is lost.—Are they all ‘amusing’? Compare chess with noughts and crosses [Xs and Os]. Or is there always winning and losing, or competition between players? Think of patience. In ball-games there is winning and losing; but when a child throws his ball at the wall and catches it again, this feature has disappeared. Look at the parts played by skill and luck; and at the difference between skill in chess and skill in tennis. Think now of games like ring-a-ring-a-roses; here is the element of amusement, but how many other characteristic features have disappeared! And we can go through the many, many other groups of games in the same way; we can see how similarities crop up and disappear.
(Wittgenstein, 1953, pp. 31–32)
Wittgenstein concluded that these concepts were best described as a family with shared features. That is, no concept necessarily possessed all family features, and all concepts possessed some family features. This alternative to the classical view, where concepts are defined probabilistically, is also known as the prototype approach, which is described in the text.
Semantic memory refers to our organized knowledge of the world. Ex: Apples have seeds, cats are soft, the grass is green.
The contrast between semantic and episodic memory; semantic memory usually refers to knowledge or info, but does not specify how we acquired that knowledge. In contrast, episodic memory implies a personal experience, because it usually emphasizes when, where or how this event happened to us.
Semantics refers to the meaning of individual words, but psychologists use it more broadly to refer to general knowledge (Martin Luther was born in Georgia), language knowledge (equality is related to the word justice) and conceptual knowledge (a square has four sides). Semantic memory influences most of our cognitive activities. It helps us determine locations, read, and make decisions.
Categories and concepts are essential components of semantic memory.
Category is a set of objects that belong together. For example, the category “fruit”, tells us something useful about its members. If you hear of a fruit that you don’t recognize, you at least know that you probably shouldn’t fry it with onions.
Concept is a term used to refer to mental representations of a category. The physical category called fruit is stored as a mental representation within your cerebral cortex.
Semantic memory allows us to organize objects we encounter in our immediate environment. Even though objects are not identical, we combine together a wide variety of similar objects by using single, one-word concepts. This reduces the space required for storage. Concepts also allow us to make inferences when we encounter new examples from a category. Ex: “fruit”, means “you can eat it.” These inferences allow us to go beyond the given info and expand our knowledge.
How do we decide which objects are similar? Three theories: 1) the prototype approach, 2) the exemplar approach, 3) network models.
What is a prototype? Describe the prototype approach. Define “prototypicality” and “graded structure,” and explain how they are related. Describe the three characteristics of prototypes that differentiate them from nonprototypes. Your descriptions must include definitions of semantic priming, typicality, and family resemblance.
Comment: The priming effect is a robust finding in cognitive psychology. As a general definition, semantic priming refers to the facilitation of any response when preceded by a semantically related event. In one of the first demonstrations of semantic priming, Meyer and Schvaneveldt (1971) had participants complete a lexical decision task or word decision task. In this task, participants were presented with letter strings on a tachistoscope, a mechanical device that precisely controls the duration and location of the letter string presentation. Some of the letter strings were actual words and others were not. The participants had to decide, as quickly as possible, whether the letter string was a word or not by pressing the appropriate button. The researchers found that responses to a word were faster when they were preceded by a related word than when they were preceded by an unrelated word. That is, participants were quicker to respond to the word NURSE when it was preceded by DOCTOR than if it was preceded by BUTTER. Although there have been many variations on this task and related tasks, the semantic priming effect is stable, and reveals interesting aspects about the organization of semantic memory.
Prototype: the item that is the best, most typical example of a category; a prototype is therefore, the ideal representative of this category.
According to this prototype approach, you decide whether a particular item belongs to a category by comparing this item with a prototype. If the item is similar to the prototype, you include the item within the category. IE: a robin is a bird because it matches my ideal prototype for a bird.
Members of a category differ in their prototypicality, or the degree to which they are representative of a category. Robins and sparrows are similar, therefore prototypical, but an ostrich or a penguin are nonprototypes, unless in the context of a zoo for example.
All members of a category are not really equal, instead a category tends to have a graded structure. A graded structure begins with the most representative, or prototypical members, and continues on through the category’s nonprototypical members.
Characteristics of Prototypes: 1) Prototypes are supplied as examples of a category. People judge some items to be better examples of a concept than other items. The typicality effect occurs when people judge typical items (prototypes) faster than items that are not typical (nonprototypes). Research shows that people often supply prototypes as examples, more frequently than they supply nonprototypes. Further, people make quicker judgement about category membership when assessing prototypes rather than nonprototypes.
2) Prototypes are judged more quickly than nonprototypes, after semantic priming. Semantic priming means that people respond faster to an item if it was preceded by an item with similar meaning.
3) Prototypes share attributes in a family resemblance category. Family resemblance means that no single attribute is shared by all examples of a concept, however, each example has at least one attribute in common with some other example of the concept.
Describe the three different levels of categorization according to Rosch’s theory. Provide examples of each. What are the three special characteristics of basic level categories?
Levels of categorization:
Superordinate-level categories are higher level, or more general categories. Ex:Furniture, animal, tool.
Basic-level categories are moderately specific. ex: chair, dog, screwdriver.
Subordinate-level categories refer to lower-level or more specific categories. Ex: desk-chait, collie, phillips screwdriver.
Basic-Level categories are more useful than either of the other two categories-generally. Why?
- Basic-level names are used to identify objects
- Basic-level names are more likely to produce the semantic priming effect.
- Different levels of categorization activate different regions of the brain.
Applying the prototype approach to social relationships: Cognitive explanations are not limited to impersonal concepts such as clothing, vegetables and birds. Semantic memory also processes our concepts about social interactions. Specifically, the prototype approach helps us understand two kinds of love relationships.
1) The prototype of compassionate love: Studies show that feeling protective towards a person, sacrifice and pure were least prototypical. In contrast, features of trust, honestly and caring were most prototypical characteristics.
2) The prototype of “being there” for a romantic partner:
Describe the exemplar approach. Do Demonstration 8.2. Summarize the findings from Heit and Barsalou (1996). Explain how the exemplar approach differs from the prototype approach.
The exemplar approach argues that we first learn information about specific examples of a concept, then we classify each new stimulus by deciding how closely it resembles all of the specific examples. Each of those examples stored in memory is called an exemplar.
The exemplar approach emphasizes that your concept of dog would include info about numerous examples of dogs you have known. In contrast, the prototype approach would argue that your prototype of a dog would be an idealized representation of a dog, with average size and features, but not necessarily like any particular dog you’ve known.
Other words, the prototype approach proposes that your stored representation is a typical member of the categoty. The exemplar approach proposes that your stored representation is a collection of numerous specific members of a category.
Describe the network model of semantic memory by Anderson known as ACT-R. Your description must include the definitions for propositional network, declarative knowledge, and proposition. What are some of the features of Anderson’s model?
Comment: The priming effect described in Question 2 is easily explained using network models of semantic memory and the basic concepts of activation and spreading activation. As a quick review, the priming effect occurs when people respond more quickly when a word is preceded by a related word as opposed to an unrelated word. In terms of the network model, when the first word (or prime word) is presented, it becomes activated and sends spreading activation to the related words through the links. When the target word is presented, it has already been partially activated through spreading activation, and thus requires less time to be completely activated. Semantic priming effects have often been used to test the predictions of network models of semantic memory.
Comment: Anderson’s complex model is only briefly described by Matlin and Farmer, which might leave the reader with more questions than answers. Unlike the previous network model, Anderson’s model is far more ambitious, and attempts to account for all aspects of cognition. For our purposes, we are primarily interested in the representation of declarative knowledge and the idea of a propositional network. The key to understanding a propositional network is to recognize that the basic unit is the proposition, or the smallest unit of knowledge that can be judged as true or false. In general, propositions involve at least two concepts and some relation. This stands in contrast to the earlier network model, where the basic unit of meaning was a single node or concept. The interested reader is encouraged to consult Anderson (1996) for a more complete description of his model.
Anderson and colleagues have constructed a series of network models, which they call ACT-R. ACT-R is an acronym for Adaptive Control of Thought-Rational. This approach attempts to account for a wide variety of tasks, including memory, learning, spatial cognition, language, reasoning, problem solving and decision making. The theory is very complex.
We focus on the theory’s more specific view of declarative knowledge, or knowledge about facts and things.
According to Anderson, the meaning of a sentence can be represented by a propositional networks, which is a pattern of interconnected propositions. He describes proposition as the smallest unit of knowledge that people can judge to be either true or false.
Anderson also suggests that each of the concepts in a proposition can be represented by its own individual network. Further, the links between nodes become stronger as they are used more often.
Do Demonstration 8.3. Describe the parallel distributed processing (PDP) approach as a network model of semantic memory. Your description must include the alternative names for this approach and the four characteristics and definitions for spontaneous generalization and default assignment.
Comment: The PDP approach can also be extremely complex. The two major features of this approach are succinctly described in the label. The “parallel” part of PDP reflects the assumption that processing of information occurs in parallel, as opposed to serially. The “distributed” part of PDP reflects the idea that memories are distributed throughout the network and not in any one specific location. It is the pattern of activation across the network as reflected in the connections between processing units. This stands in contrast to traditional network models where memories were extremely localized (i.e., each node represented a single memory). These two features distinguish the PDP approach from traditional network models.
The parallel distributed processing approach proposes that cognitive processes can be represented by a model in which activation flows through networks that link together a large number of simple, neuron-like units. This approach emphasizes that we should represent these concepts in terms of networks. rather than specific locations in the brain. The theorists who designed this approach tried to construct their model by taking into account the physiological and structural properties of human neurons.
AKA: connectionism and neural networks.
Four general characteristics of the PDP approach:
1- As suggested by the name, parallel distributed processing, cognitive processes are based on parallel operations, rather than serial operations. Therefore, many patterns of activation may be proceeding simultaneously.
2- A network contains neuron-like nodes, which as connected together so that one node is linked to other nodes. PDP theorosits argue that most cognitive processes can be explained by the activation of these networks.
3- The process of spreading information from one node to another is called spreading activation. As the name PDP suggests, a concept is represented by the pattern of activity distributed throughout a set of nodes. This view is very different from the commonsense idea that all the info you know about a person or object is stored in one specific location in the brain.
4- Consistent with the concept of situated cognition, the current context often activates only certain components of a concept’s meaning. Walking in the meat section won’t necessarily connect with animals that clucks and lays eggs.
One advantage of the PDP model is that it allows us to explain how human memory can help us when some info is missing. Specifically, people can make a spontaneous generalization by using individual cases to draw inferences about general info. Spontaneous generalization can help explain stereotyping.
Spontaneous generalization means that we draw a conclusion about a general category and default assignment means that we draw a conclusion about a specific member of a category.
What are the four theoretical features that distinguish the PDP approach? Describe the current status of the PDP approach.
1) The connections between these neuron-like units are weighted and these connection weights determine how much activation one unit can pass on to another unit. As you learn new info, the values of these weights change.
2) When a unit reaches a critical level of activation, it may affect another unit either by exciting it, or inhibiting it.
3) Each new experience with a particular item will change the strength of connections among relevant units by adjusting connection weights.
4) Sometimes, we have only partial memory for some info, rather than complete, perfect memory. The brain’s ability to provide partial memory is called graceful degradation. This explains why the brain continues to work somewhat accurately, even after an accident, stroke, or dimentia has destroyed portions of the cortex.
Define “schemas” and “scripts.” Provide original examples of each. How do they differ? Define “heuristics.” Explain how schemas and scripts are related to heuristics. Do Demonstration 8.4. What are life scripts? Why are they important?
Comment: Many students have trouble distinguishing between scripts and schemas. As Matlin and Farmer indicate, scripts are knowledge structures of situations that involve consistent sequences of events that occur over time. For example, our knowledge of restaurants is likely to be represented as a script because, for most restaurants, a reliable sequence of basic events unfolds over time (e.g., enter, be seated, order, eat, pay, leave). Our knowledge of circuses, however, is more likely to be represented as a schema because the events are not as fixed. Once you enter, there could be clowns, or animal acts, or acrobatic acts, or high wire acts, or any combination of these. As Matlin and Farmer note, however, the two terms are often used interchangeably, although scripts is the narrower term.
Schema: Generalized, well-integrated knowledge about a situation, an event, or a person. They are the basic building blocks for representing our thoughts about people.
Ex: I expect to find produce, bread and meat in a grocery store, not car parts.
Schemas help us make predictions about a situation that are generally correct.
Script: One kind of schema is a script. A script is a simple, well-structured sequence of events in a specified order; this script is associated with a highly familiar activity. A script is a prototype of a series of events that share an underlying similarity.
Script is a more narrow term than schema- referring to a sequence of events that unfold in a specified order, like the standard sequence of events that a customer might expect in a restaurant: sitting down, menu, ordering, eating, paying…..
Schemas are one kind of heuristic, which is a general rule that is typically accurate.
In Schema therapy, the clinician and the client may work together in order to explore the client’s core beliefs and create appropriate, new, more helpful strategies.
A life script is a list of events that a person believes would be important throughout his or her lifetime.
Do Demonstration 8.5. Describe the research by Brewer and Treyens (1981) and what they found regarding memory for schema-consistent material. Summarize the research on memory of schema-inconsistent material, including the four trends.
Brewer and Treyens asked paticipants to wait in a room that the researcher called his office, then were asked to move to a different room and told to recall all they could about the contents of the office. The results show that people were likely to remember items that are consistent with an office schema. Some people also “remembered” items that were not even in the room, but consistent with the office schema.
We sometimes show better recall for material that violates our expectations. People are more likely to recall schema-inconsistent material when that material is especially vivid and surprising. Research by Davidson demonstrates this. He asked people to read stories that described well known schemas (going to movies). The results showed that people were especially likely to recall schema-inconsistent events when those events interrupted the normal expected story.
The results about schemas and memory seem inconsistent, however, the outcome may depend on factors such as the details of the study and length of the specific episode. In general, results show the following trends:
1) If the info describes a minor event and time is limited, people tend to remember information consistent with a schema.
2) If the info describes a minor event, and time is limited, people do not remember info that is inconsistent with the schema.
3) People seldom create a completely false memory for a lengthy event that did not occur.
4) When the info describes a major event that is inconsistent with the standard schema, people ARE likely to remember that event.
Do Demonstration 8.6. What is boundary extension, and how is it related to schemas?
Boundary extension refers to our tendency to remember having viewed a greater portion of a scene than was actually show; our cognitive processes fill in the incomplete objects.
Out schemas for complete objects help us fill in missing material during a memory task.
Based on our expectations, we create perceptual schemas that extend beyond the edges of the photograph and beyond the scope of our retinas.
Do Demonstration 8.7. What is abstraction? How is it related to schemas? How does abstraction differ from verbatim memory? What is the constructive model of memory? Describe the evidence that supports this model. What is the pragmatic view of memory? Describe the evidence that supports this view. What is the current status of schemas and memory abstraction?
Abstraction is the memory process that stores the meaning of a message, rather than the exact words.
Research shows that people usually have poor word for word recall, or verbatim memory even a few minutes after a passage has been presented.
Do we need verbatim memory in our lives? There are two approached to the abstraction issue:
1) The constructive approach: Bransford and Franks asked participants to listen to sentences from several stories. They were then given recognition tests that included some new items and combinations of the earlier sentences. People were convinced that they had seen these new items before.
In memory research, a false alarm occurs when people “remember” an item that was not originally presented.
The research showed that people are likely to make false alarms when a complex sentence was consistent with the original schema, and seldom made false alarms for sentences that violated the meaning of the earlier sentences.
According to the constructive model of memory, people integrate info from individual sentences in order to construct larger ideas. Later, they believe that they have already seen those complex sentences because they have combined the various facts in memory. Once sentences are fused in memory, we cannot untangle them into their original components and recall those components verbatim.
The constructive view of memory emphasizes that our cognitive processes are generally accurate. Sentences are not passively stored, but rather are combined into a coherent story, fitting related pieces together. We typically store an abstract of the info, rather then word-for-word representation.
The Pragmatic Approach: The pragmatic view of memory proposes that people pay attention to the aspect of a message that is most relevant to their current goals. In other words:
1- People know that they usually need to accurately recall the gist of a sentence.
2- They also know that they usually do not need to remember the specific wording of the sentences.
3- However, in those cases where they do need to pay attention to the specific wording, then they know that their verbatim memory needs to be highly accurate.
Murphy and Shapiro speculated that people were most likely to remember the exact wording of a sentence if the words are part of a criticism or insult. Participants read letters that were relatively bland and included neutral sentences. A second group read the same letters, but some of the sentences were written in a sarcastic context. The participants then completed a recognition test. Correct recognition was higher for the sarcastic condition than the bland condition.
Current Status: The constructive and pragmatic approaches are compatible. We integrate info from individual sentences so that we can construct a schema, especially when we don’t need to remember the exact words. However, in some cases we know that specific words matter, so we pay close attention.
What is memory integration? What does the research of Bartlett (1932) reveal about memory integration?
Comment: As the preceding questions indicate, schemas influence our perception and understanding of the events that happen to us on a daily basis. The same events that are perceived and understood through schemas are also ultimately integrated into the schemas. Our schemas are constantly changing as our current experiences are integrated with past experiences.
In memory integration, our background knowledge encourages us to take in new info in a schema-consistent fashion. As a result, many people remember this schema consistent info, even though it was not part of the original stimulus material.
Bartlett asked students to read a Native American story, then they were asked to recall the story after 15 minutes. Bartlett found that students tended to omit info that did not make sense from the viewpoint of British students. They also tended to shape the story so that it was more similar to British fairy tales.
The students were asked to recall the story again after several days delay. The participants borrowed more heavily from their previous knowledge and included less info fro the original story.
Subsequent research confirms that schemas can influence our memory when we are reading unclear material. The research demonstrates how our cognitive processes actively work to make sense out of puzzling info. Specifically , our top down processes often shape our memory for complex material. Also, schemas can mislead us, so that we make systemic errors and “remember” info that was not stated.
Gender stereotypes are beliefs and opinions that we associate with females and males.
Research- 287
The research in gender stereotypes shows that people make schema consistent inferences in explicit and implicit memory. Countries that show the highest scores on tests on implicit gender-stereotype tests, are countries where boys earn higher scores than girls on tests that measure science and math performance.
