Chapter 7

Categories - refer to a class of similar things that share an essential core (why all science courses are considered "science") or some perceptual, biological, or functional properties. Categories are more often used to describe objects that objectively exist in the world.
Concepts, on the other hand, more often are used to refer to mental representations of those existing categories. So, concepts are mental representations of objects, events, etc.
5 theories regarding nature of Concepts (Also find the problems with each approach):
1. Classical view of concepts - views concepts as consisting of a list of features that are both individual necessary (the features must be present) and collectively sufficient (if all the features are present, then the object is a member of the concept).
3 implications for the classical view of concepts (Also find the problems with each implication).
First - the assumption that concepts consist of a list of features. Thus, (the implication is that) there are no representations of specific examples for a concept.
Second - membership in a category/concept is clear-cut, that is, there should be no ambiguity concerning category or concept membership.
Third - all members in a category are considered equal.
2. Prototype vies of concepts. - sees concepts as consisting of an idealized representation of a class of objects or events. In the prototype view of concepts, when researchers are talking about instances that belong to a concept they usually talk about the family resemblance of concepts. The prototype view can explain typicality effects.
Concepts can exist at various levels of a hierarchy, and by that we merely mean concepts can be very specific (e.g., a pair of jeans) or very broad (e.g., clothing), or in between.
Rosch and colleagues have found that one level is psychologically basic. This is the basic level of categorization. The broader concepts are called superordinate levels of categories. The more specific concepts are called subordinate levels of categories.
3. exemplar view of concepts - according to this approach, concepts include representations of individual instances of the concept. Researchers think that typical instances of a concept are more likely to be stored as exemplars.
4. schemata/scripts view of concepts - shares similarities with both the prototype view and the exemplar view, because the schemata are conceptualized as storing abstract information about a concept (such as the typical characteristics) and also as storing actual examples (exemplars) of the concept. Schemata can exist on different levels, having subschemata and superschemata.
5. knowledge-based view of concepts. According to this approach, concepts are related to an individual's knowledge and worldview. Concepts and examples of concepts are analogous to theories and data supporting the theories; so concepts are not static, but are constantly being tested and updated/changed.
These five approaches to concepts that we've mentioned can be split into two categories themselves (Also find the problems with each of these & which of the 5 theories/views goes with each category).
First, there are the approaches that envision concepts as being similarity based. According to these approaches categorization is based on the similarity of an instance to the concept. The similarity based approaches tend to emphasize a focus on superficial, perceptual information about an object when forming concepts.
Second, there are the approaches that envision concepts as being explanation based. Here people are conceived of as forming concepts based on meaningful relationships among instances and categories. The explanation based approaches tend to emphasize a focus on deeper, knowledge-derived information about an object's function.
Stereotypes - occupational concepts can influence our memories of a stranger's behavior. Individuals will form stereotypes even when the standard stereotypes of gender and ethnicity are unavailable to them.
Some research suggests that the basic way we categorize human beings is on the basis of gender.
In-group bias - an evaluation of one's own group as better than other groups.
Concept formation refers to the discernment of the properties common to a class of objects or ideas (Bruner et al., 1956, study). Participants used one of three strategies when forming their concepts.
First, conservative focusing is a very systematic approach. The participant would test only one feature change at a time. Second, a more mentally demanding strategy is called simultaneous scanning. Simultaneous scanning involves an individual figuring out multiple hypotheses (concepts) and then choosing specific, optimal cards that will rule out or rule in the hypotheses. Third, the successive scanning strategy tests only one hypothesis or concept at a time. Successive scanning is not as demanding, because participants are just testing one hypothesis at a time, then coming up with the next hypothesis and testing it. For clear cut stimuli people form more precise concepts partially in line with the classical view of concepts. Category membership in the concept is clear cut.
Feature positive effect - you learn a concept better/faster if you see instances of it, rather than if you see examples that are not instances of the concept.
Prototype formation - for more ambiguous stimuli individuals seem to be able to form concepts based on prototypes. And, learning about variability enables people to create larger, more inclusive concepts that include more stimuli (as shown in Posner & Keele experiment).
If individuals use prototypes, then do individuals use exemplars?
Implicit concept learning - Reber (1967, 1976) found that participants who tried to learn letter strings by learning the underlying grammar/rules performed worse than participants who simply memorized letter strings. In these findings, people performed better by memorizing exemplars and, therefore, by engaging in nonanalytic concept formation (i.e., implicit concept learning). Individuals probably engage in this approach (nonanalytic) to concept formation daily.
Brooks (1978) used paired-associates of hieroglyphic symbols with English words. Later, when participants answered questions about the hieroglyphic symbols they reported that they thought of examples/exemplars to give their answers.
So, it seems that individuals take an analytic approach, a prototype approach, & an exemplar approach to concept formation. Simple laboratory tasks may lend themselves to an analytical, hypothesis-testing approach to concept formation, however, more complex, real-world stimuli may lead people to other concept formation approaches.
Brooks argues that there are 5 factors that encourage people to use exemplars: (1) when one needs to be able to learn information that distinguishes among individual instances of the concept, (2) when a person gets to know certain instances of a concept very well (is repeatedly exposed to certain instances), (3) when instances of a concept vary in many ways that are not necessarily apparent to a novice, (4) when real-life concepts belong to a number of categories at the same time, (5) when we have to learn about a concept without knowing how we will have to use the information later.
Nelson (1984) argues that nonanalytic concept formation is more likely with materials that have strong family resemblance structures.
Using & Forming Scripts - research has found that when individuals are asked to write down their scripts for an event (such as a "McDonald's script") a high degree of overlap is found in what people mention, and in their description of information level (eating the food, instead of pick up spoon and dip into ice cream). In addition, when information is presented to participants in a scrambled order, individuals tend to recall the information in a correct scripted order. When only some information is presented to participants, they tend to add information in their recall that comes from a relevant script. And, information more consistent with scripts is remembered better. So, the overall data and findings support the conclusion that individuals do form and use scripts.
Psychological Essentialism - Others have suggested that part of a concept includes information regarding the underlying nature of the category (e.g., molecular or chromosomal structure). In other words, people act as if objects, etc. have underlying essences that make them what they are. It is because objects share this essence they belong to the same concept.
However, individuals do not always make judgments based on these essences. Instead, people may often base their judgments regarding concept membership based on superficial characteristics. Nevertheless, making judgments of category membership based on superficial factors seldomly leads to errors, although it can (e.g., seeing a person from behind with long hair and assuming it is a female, but "it is a man man!" - Austin proclaims!).
So, novices tend to base judgments of category membership on superficial factors, however, as one gains more experience and becomes an expert s/he will be more likely to base his or her judgments of category membership on deeper principles (i.e., underlying essences).
Exactly what factors are stored as part of a concept in part depends on the kind of concept. Some researchers argue that there are three kinds of concepts: (1) nominal-kind concepts - have clear definitions that include information on necessary and sufficient features (classical view); (2) natural-kind concepts - are things that naturally occur in an environment, include more information about essential features (essentialism) and have a family resemblance structure (prototype view), can also fit into the knowledge-based view; & (3) artifact concepts - things constructed to serve some function (i.e., tools & other usually human creations), information about the object's purpose or function is stored in the concept (fits with the knowledge-based view). Some research seems to support these kinds of concepts.