Language, Cognition, and Intelligence

Language, Cognition, and Intelligence

Language

Language production: the ability to use words, phrases, and sentences to convey information.

Language comprehension: the ability to understand the messages conveyed by words, phrases, and sentences.

Key to understanding language is to appreciate that it consists of simple building blocks that can be combined in many ways according to certain rules. There are four distinct types of building blocks:

1)  Phonology

2)  Syntax

3)  Semantics

4)  Pragmatics

Phonology: the structure of sounds of the words in a language; there are about 100 phonemes or basic sounds that humans are capable of making, but no single language uses all 100. English uses about 45. (Boy and toy differ in one phoneme.) Some sounds are accented in English, but in other languages (such as French), each syllable is given equal emphasis.

Syntax: the internal organization of a sentence, determined by a set of rules—called grammar—for combining different parts of speech into an acceptable arrangement. E.g., “Kicked girl ball blue the” has inappropriate syntax.

Semantics: the meaning of a word, phrase or sentence. Morphemes are the smallest unit of meaning in a language. E.g., “ing” or “ed” added to a word. The meanings of many words are determined by more than one morpheme. (Undoing, undone). Puns come into play here. “Energizer bunny arrested and charged with battery.” Morphemes are also subject to syntax (you can’t add “ing” to the beginning of the word.) Specific events in the past contribute to the meanings of words (e.g., “berserk” comes from Vikings wearing bearskin shirts into battle, leading to the old Norwegian word of “berserkr,” meaning the ferocity of Vikings in battle). Different parts of the brain are involved in processing semantics and syntax. Semantics and syntax are largely distinct. We can understand the phrase “Fastly dinner eat, ballgame soon start” (Yoda speak), but the syntactically correct “Colorless green ideas sleep furiously” has no meaning.

Pragmatics: the indirect or implied meaning of words and sentences. “Do you know the time?” means “Tell me what time it is.” You’re not looking for an answer of yes or no. Pragmatics plays a role in understanding metaphors, in which we must actively inhibit or suppress the irrelevant aspect of the meaning. “The lawyer is a shark.” Understanding metaphors uses different parts of the brain than other aspects of language use.

INTERLOCKING MECHANISMS

The four building blocks are intertwined. We usually learn to produce the sounds of words at the same time that we learn to comprehend them, and the mechanisms that produce syntactically correct sentences also help us understand them. We use the principle of pragmatics both when producing speech (guiding our patterns of intonation and use of metaphor and humor) and when comprehending speech.

BILINGUALISM

All people with normal language ability learn language just by being immersed in it. Most people in the world also learn a second language at some point. Kim et al. (1997) used functional MRI to see if different parts of the brain were used in native speakers vs. adult learners of a language and found that language is processed differently when learned as adults than when learned as children. For adult learners, part of the frontal lobe was activated when they used the 2nd language. This part of the brain plays a part in working memory, indicating that adult learners must rely on conscious thought & working memory when using the second language. It doesn’t come automatically the way language learned during childhood does.

Adults pick up vocabulary words in a second language easily, but the same can’t be said for learning grammar and phonology. Most people who learn a 2nd language after puberty make grammatical errors and speak with an accent, although some can speak flawlessly. Musical ability contributes to the differences between the flawless and imperfect speakers; people with good musical ability can learn to pronounce words n another language better than those without good musical skills. The more formal education you have and the younger you are when learning a 2nd language, the better you’ll be at learning a 2nd language.

Cognition

Mental Images—mental contents like those that arise during perception, but they arise from stored information rather than on immediate sensory input. Thinking relies on mentally manipulating information, which is stored in different ways. Both inner speech (self-talk) and metnal images are manipulated in working memory and play a key role in thinking.

Thinking is more than just talking to yourself. Why?

1)  Words are often ambiguous, but thoughts are not. (You don’t confuse yourself by thinking of a “solution” and wondering in which way you meant to use the word.)

2)  If thinking were just talking to yourself, why would you ever have trouble putting thoughts into words?

3)  Animals seem to “think” but they don’t have language. They do have problem-solving ability, which doesn’t seem possible without “thought.”

The idea that language shapes our perceptions and thoughts—and thus, that people with different languages think differently—is known as the linguistic relativity hypothesis. (Benjamin Lee Whorf, 1956). This has been found not to be true; language clearly does not entirely shape our perceptions and thoughts, although words do help. Language can enhance memory (elaborative rehearsal).

Are mental images the basis of thought, or are they like language—used in thinking but not the fundamental form of thoughts?

--Visual mental imagery, the most prevalent form of imagery, relies on most (about 92%) of the same parts of the brain that are used in actually seeing objects.

Three properties of visual mental imagery:

1)  Spatial extent-when we see an object, we see it as extending over a specific portion of space; the object is said to have spatial extent. In the mind, objects in mental images resemble actual objects. Many experiments have shown that more time is required to scan greater distances across visualized objects; when you visualize an object, that object in the image has a definite spatial extent. Occipital and parietal lobes are active during visual imagery with the eyes closed.

2)  Limited field of view-in visual mental imagery, we “see” only a limited field of view, the same way we see it in real life. The larger the object, the farther away it seemed to be in the mental image when it began to overflow. This is because mental imagery relies on many of the same brain systems that visual perception relies on.

3)  Limited resolution-we see objects with limited resolution (level of fine details). In imagery, we can’t see fine details if we’re imagining an object that’s too tiny. People require more time to “see” images that are tiny than those that are larger.

There are several types of imagery—visual, auditory, motor, and spatial (even blind people have spatial images), and we can manipulate objects in our mental images. (Turn “N” 90 degrees to get Z.)

Mental images aren’t the only thing involved in thought. Limitations are:

1)  Abstract concepts—not easily represented as an image (love, justice, mercy…)

2)  Ambiguity—some images are ambiguous. Are you looking at a box from the side or a square piece of cardboard?

3)  Individual differences—there are some people (about 2%) who have poor visual imagery, but they’re not poor thinkers.

Concepts

A concept is an idea that underlies the meaning of a word or image. It may be only a single word or a phrase. A concept is unambiguous and may be concrete or abstract. Concepts may be expressed as words or images, but they are not the same as either. A concept is specified by morphemes, which convey the underlying idea or meaning of the word. Sometimes words mean more than one concept (“solution”), and sometimes the same concept can be expressed in different words (sofa, couch). The same image can sometimes express different concepts (image of apple can be “apple,” “fruit,” or “Golden Delicious”), and more than one image can sometimes express the same concept (bananas and oranges = fruit).

Aristotle said that a given concept is defined by a set of features, but this thinking has been discarded because 1) it often isn’t clear what the features of a particular concept might be; and 2) even when it is clear what the features are for a concept, the theory makes wrong predictions. The theory predicts that either a concept applies to a given object or it doesn’t, period. It’s all or none. However, concepts can actually apply “more or less” to a given object. How well the concept applies depends on the object’s typicality. The more typical the object is, the better it fits the concept. Canary fits “bird” better than “ostrich” does.

Prototypes are the most typical example of the concept. Concepts are specified as a set of features that describe the prototype. Only a percentage of these features need to be present for the concept to apply. This is the advance over Aristotle—it’s not all or none but a “good enough” fit.

Sometimes concepts are stored as a collection of examples, not as prototypes. And sometimes they’re stored by functions (especially verbs).

Concepts are organized in terms of specificity, and when classifying an object or event, we use a concept at what Rosch et al have called the basic level. The basic level is at an intermediate level of specificity; like the middle rung of a ladder with more general concepts above it and more specific ones below it.

Problem-Solving

Problem-solving: consists of devising a way to overcome an obstacle ( a problem) that stands in the way of your present situation and a desired goal.

Reasoning: deciding what follows from an idea or situation. Plays a key role in solving problems but is not only involved in problem-solving.

Steps in solving problems:

• Representation problem—the first step; how to characterize the nature of the problem itself (with description, drawings, or both). E.g., the Hiking Monk problem. Finding the best representation of a problem can be hard because you can get stuck on one interpretation, called functional fixedness, that requires effort to break out of. Extra brain activity is required when you need to inhibit one way of viewing a situation and switch to another.
• Strategy—an approach to solving the problem. Two types of strategies:
• Algorithms—set of steps that, if followed methodically, will guarantee the correct solution to the problem
• Heuristics—a rule-of-thumb strategy that does not guarantee the correct solution but offers a shortcut to it
• Analogy—a type of heuristic that relies on finding points of correspondence with previously solved problems and their solutions (e.g., compare brain surgery to military strategy to best destroy a tumor)
• In order to enhance problem-solving, do the following:

1)  Represent the problem effectively. Sometimes explaining the problem to someone else helps you find a solution or new way of looking at it.

2)  Focus on the problem. Don’t lose sight of what actually constitutes the problem (don’t redefine the problem or add to it).

3)  Don’t restrict the resources (get locked into seeing the problem from only one angle)

4)  Consider alternative types of solutions (don’t get stuck with a certain mental set—an approach to solving a problem that worked for a similar one in the past, which leads to a fixed way of thinking about how to solve the present one.

5)  Take a fresh look. Walk away from the problem for awhile.

LOGIC

All types of reasoning rely on logic, which is a set of rules that determines which conclusions follow from particular assumptions.

Deductive reasoning—going from general to specific. If Jack is a man, and all men are Martians, then Jack is a Martian. It doesn’t matter if the conclusion makes no sense; it only matters that the steps—the form--are followed correctly.

Inductive reasoning—works from the specific to the general. Here you’re using specific, individual examples to figure out a general rule. E.g., bills of the Euro increase in size as they increase in value. Scientific method relies on this type of reasoning—we try to generalize from samples to a population.

Logical errors

1.  Affirming the consequent: assuming that a specific cause is present because a particular result has occurred; we incorrectly work backward, from result to cause. E.g, I’m wearing my church shoes, so it must be Sunday. Not necessarily.

2.  Confirmation bias: a bias to seek information that will confirm a rule but not to seek information that would refute it. Example: Wason and Johnson Laird’s Card Task: A D 4 7. The rule is “If a card has a vowel on one side, then it will have an even number on the other side.” How many and which cards must be flipped over to decide whether this rule is true? Answer: 2. The A and the 7.

Heuristics & biases

1.  Representativeness heuristic: the strategy in which we assume that the more similar something is to a prototype stored in memory, the more likely it is to belong to the category of the prototype. (Tversky & Kahneman’s accountant vs. engineer problem). People ignore the base rate (70% likelihood that he’s an accountant) and say Jack is an engineer because he has characteristics representative of engineers (likes math puzzles, not interested in social & political issues, etc.). Base rate fallacy. The base-rate rule states that if something is chosen at random from a set, the chance that the thing will be of a particular type is directly proportional to the percentage of that type in a set.