CARS VOL2 passage 22 116-120

The classical view of concepts is that their members share conditions that are singly necessary and jointly sufficient to define the concept. For example, someone who is a grandmother has the two criterial conditions of being female and being a parent of a parent. Singly, each condition is necessary—someone who is a man cannot literally be a grandmother, nor can someone who did not produce a child who also produced a child. Jointly, the two conditions are also sufficient—someone who is both female and the parent of a parent is necessarily a grandmother.

The most severe problem for the classical theory is that decades of research have failed to determine such clear-cut definitions for many concepts. Categorization studies suggest an alternative way of understanding concepts. People can reliably order concept members by the extent to which they represent or typify the concept. Apple and peach are judged to be typical fruits, raisin and fig less typical, and pumpkin and olive atypical. The ratings obtained in such studies are relatively uncorrelated with the frequency of occurrence or familiarity of the instances.

The most important aspect of typicality ratings is that they predict performance on a variety of tasks. Research subjects asked to answer yes or no as quickly as possible to the question of whether something belongs in a given category ("Is a fig a fruit?") respond more quickly, the more typical the instance. Furthermore, if asked to generate from memory all instances of a concept, people recall typical before atypical instances. And children learning the names of concept members master typical before atypical ones.

These typicality effects show that the instances of concepts are not always equal. Yet equality would be expected if, as previously assumed, concepts could be defined in a way that was true of every instance. It seems that most natural concepts are defined not by criterial conditions but by properties that are perceptually salient. These properties occur in various numbers and combinations and predominate in only some instances—those considered the "best" examples of the concept. Whether, or how readily, a particular object is classified as a concept member thus depends on its similarity to a hypothetical prototype incorporating all of these properties. (For research purposes, similarity is quantified as the number of properties shared by an object and the prototype.)

Clearly, a prototype model of concepts accommodates the phenomena that challenged prior assumptions. For example, in a yes-no classification task with items that exceed some critical level of similarity to a prototype, the more similar the item and the prototype, the more quickly it should be accepted as an instance of the concept. Conversely, similarity to a prototype would delay the rejection of a noninstance.

Yet recent findings indicate that in ordinary life, categorization is more a process of inductive inference than of comparison with a prototype. In one study with U.S. college students, items were described only by a value on a single dimension and the task was to decide to which of two concepts each item belonged. One item was "an object three inches in diameter," and the concept possibilities were pizzas and quarters. Although a three-inch diameter is objectively closer to that of a quarter than to that of a typical pizza, the object was almost always classified as a pizza. The apparent reason is the greater constraint on the size of coins than on the size of pizzas. This evidence indicates that people consider permissible variability from a prototype as well as similarity to it in classifying objects.

null

According to the passage, the demonstration that an unknown object closer in size to a prototypical quarter than to a prototypical pizza was called a pizza challenges the prototype model by suggesting that the judgment was an inductive inference. The most reasonable objection to this conclusion would be that:

A

the author offers no other evidence to support it.

B

the judgment indicates a lack of knowledge about the concepts.

C

the construction of a prototype is also an inductive process.

D

the diameter of flattened quarters can be unusually large.

Solution: The correct answer is C.

1. The passage author supports this conclusion as follows: “Although a three-inch diameter is objectively closer to that of a quarter than to that of a typical pizza, the object was almost always classified as a pizza. The apparent reason is the greater constraint on the size of coins than on the size of pizzas.”

2. The judgment indicates a full knowledge and understanding of the concepts: “The apparent reason [for this judgment] is the greater constraint on the size of the coins than on the size of pizzas. This evidence indicates that people consider permissible variability from a prototype as well as similarity to it in classifying objects.”

3. A prototype could be the product of an inductive inference, in which the conditions defining a particular example lead to a general conclusion: “Whether, or how readily, a particular object is classified as a concept member thus depends on its similarity to a hypothetical prototype incorporating all these properties.” Thus a particular example embodying a requisite number of conditions could be held up as a prototype.

4. It is not likely, given the definition of a prototype, that a flattened quarter is a frame of reference, especially since, as the passage author points out, there is a “greater constraint on the size of coins than on the size of pizzas.” See rationale C.

What does the author mean to imply by pointing out that the typicality ratings of concept instances are relatively uncorrelated with their familiarity?

A

Experience with the instances does correlate positively with their familiarity.

B

Typical instances are usually encountered more often than atypical instances.

C

The ratings may reflect differential knowledge rather than agreement among the raters.

D

The ratings cannot be explained by the raters' personal experiences with the instances.

Solution: The correct answer is D.

1. Typicality is largely unrelated to familiarity. See rationale D.

2. The passage author does not discuss the relative frequency with which concept instances get rated typical or atypical.

3. The ratings are not reflected in differences between the concept instances but in “the extent to which [the concept members] represent or typify the concept.”

4. The phrases “frequency of occurrence” and “familiarity of the instances” are ways of saying that the raters have personal experiences with the concept instances. Personal experiences do not determine how people typify the concept as much as how something gets more or less typified; for example, an apple is a typical fruit, and an olive is atypical.

Suppose that preschoolers in England are more apt to say raisin than peach when asked to name a fruit. According to the author's reasoning, this phenomenon suggests that:

A

the classical explanation is sometimes best.

B

the properties of prototypes are not universal.

Answer choice eliminated

C

categorization is sometimes a deductive process.

D

no current model can fully explain categorization.

Solution: The correct answer is B.

1. The classical explanation would be inadequate in this instance, since it would fail “to determine such clear-cut definitions” for these particular concepts.

2. The passage author argues that classical theory fails to always determine clear-cut definitions and that typicality effects are not always equal. If preschoolers in England were more apt to say raisin than peach when asked to name a fruit, then, given the passage information that peaches generally tend to be classified as more typical and raisins less typical fruit, it follows that the properties of prototypes are not always universal. The passage author states, “It seems that most natural concepts are defined not by criterial conditions but by properties that are perceptually salient. These properties occur in various numbers and combinations and predominate in only some instances.”

3. Deductive reasoning would begin with the universal in order to determine the particular, but a universal would be indeterminate in this instance. See rationale B.

4. The passage author apparently concludes otherwise, stating that “a prototype model of concepts accommodates the phenomena that challenged prior assumptions.”

Which of the following theoretical constructs is most analogous to the concept of a prototype?

A

The linguistic theory that language is a grammatical structure plus sets of interchangeable words

B

The physiological theory that the brain is compartmentalized by psychological function

C

The geological theory that planets are formed by the accretion of smaller bodies

D

The evolutionary theory that species diverge from an original common form

Solution: The correct answer is D.

1. Linguistic theory is more analogous to defining concepts by criterial conditions rather than “by properties that are perceptually salient.” While its parts are interchangeable, the grammatical structure does not vary its basic form; whereas concept members vary the prototypical form while retaining some of the characteristics of the prototype.

2. The concept of a prototype depends not on the static compartmentalization of functions but on the more dynamic notion that properties are “perceptually salient” and occur in various combinations in the member concepts.

3. This analogy is, arguably, the opposite of the prototype concept. A planet takes on new properties and, ultimately, new form itself through the accretion of other forms, whereas a prototype retains its form while its properties occur in various combinations in member concepts.

4. The prototype incorporates all the properties found in different combinations of various concept members. The passage author elaborates as follows: “It seems that most natural concepts are defined not by criterial conditions but by properties that are perceptually salient. These properties occur in various numbers and combinations and predominate in only some instances—those considered the ‘best’ examples of the concept. Whether, or how readily, a particular object is classified as a concept member thus depends on its similarity to a hypothetical prototype incorporating all of these properties.” Thus concepts that are carrying the properties of a prototype in various combinations can then be said to diverge from a prototype much as species diverge from a common form. In this regard, the new species are much like natural concepts, retaining characteristics of the common form while becoming a variation of that form.

Assume that adult subjects are asked to respond as quickly as possible to the question “Is X a bird?” What does passage information suggest about relative response speeds when X is NOT a bird?

A

A bat will be rejected more quickly than a mule.

B

A mule will be rejected more quickly than a bat.

C

A bat and a mule will be rejected with equal speed.

D

A bat will be rejected more quickly than a turkey is accepted.

Solution: The correct answer is B.

1. The opposite would be true. See rationale B.

2. The mule would be rejected because the bat would be more quickly accepted for its greater similarity to the bird prototype. The passage author explains as follows: “in a yes-no classification task with items that exceed some critical level of similarity to a prototype, the more similar the item and the prototype, the more quickly it should be accepted as an instance of the concept.”

3. The mule would be rejected more quickly because it is less similar to the prototype. See rationale B.

4. Because the turkey is a bird, it would be quickly accepted and the bat simultaneously rejected. Moreover, this response is inconsistent with the item question, which asks about response time when X is NOT a bird.