Language and Thought - Psy102
Language and Thought
Learning Goals
Evaluate the linguistic relativity hypothesis.
Describe the structure of language and its basic units.
Explain the role of shared knowledge in language comprehension (pragmatics).
Describe the major milestones of language development.
Evaluate whether there is such a thing as ‘non-human language’.
Evaluate why the ability to form categories is useful.
Explain why psychologists have rejected the idea of “defining features” and think instead of the “family resemblance” of category members.
Describe how all things can be thought of as members of categories at many different levels and what basic level categories are.
Explain the differences between ill-defined problems and well-defined problems.
Explain the differences between algorithms and heuristics.
Describe how the availability heuristic can sometimes lead to incorrect judgments about probability.
Linguistic Relativity Hypothesis
Definitions
Weak Version: The language that you speak influences how you think.
Strong Version (Linguistic Determinism): The language that you speak determines how you think.
Evidence for the Weak and Strong Versions
Weak LR Hypothesis
Instances where spoken language appears to influence thought have been observed.
Strong LR Hypothesis
Key Question: Does language determine what you think? Specifically, if your language has no way of describing something, do you face problems thinking about it?
Evidence:
Example from literature: In George Orwell's novel "1984," a totalitarian regime alters language to eliminate discussions about revolution, suggesting that limiting language can limit thought.
Real-World Example: The Hopi language lacks past tense forms of verbs. The question arises: Do Hopi speakers therefore think about the past differently?
Contrast examples:
English: "I walk to school" versus "I walked to school".
Hopi: "I walk to school" versus phrases describing past actions without requiring tense forms.
Influence of Language on Thought
Flexibility of language allows expression of concepts even in the absence of specific vocabulary.
Examples:
The Dani people have only two words for colors—“mola” for warm tones and “mili” for cooler tones. Question raised: Does this limit their perception of color compared to English speakers?
The Hanunoo people possess 92 names for rice, potentially leading to a different perceptual experience compared to English speakers.
Debate: Is the variability in perception influenced by language, or does it reflect cultural importance?
Concluding Points
Language does not strictly determine thought but can influence it significantly.
Research by Carroll & Casagrande (1958) discusses strong existing correlations, but evidence suggests that cultural thinking precedes linguistic development.
Structure of Language
What Makes Language Distinct?
Animals communicate, but what differentiates animal communication from human language?
Hierarchical Structure and Generative Grammar
Human language is characterized by a hierarchical structure, meaning it is organized into multiple levels, from the smallest units of sound to complex sentences and meaning.
Phonemes: The smallest units of sound in the language; e.g., in "cat," phonemes include /k/, /æ/, and /t/.
Phonology: Rules dictating how phonemes are combined to form words.
Morphemes: The smallest meaningful units in language; can be whole words or prefixes/suffixes (e.g., "un-" for negation, "-ed" for past tense).
Syntax: Rules for combining words into correct sentences.
Semantics: Rules for understanding meaning within sentences.
Pragmatics: The social aspects and contextual understanding of language.
English Language Complexity
Lack of one-to-one correspondence between phonemes and letters creates challenges in reading and writing English.
Example: The letter ‘g’ has different sounds, as in "gun" versus "giraffe."
Example: Complexity in the spelling of the same sound using different letters (e.g., the /s/ sound can be represented by ‘s’, ‘ss’, ‘c’, ‘ce’, ‘se’, ‘ps’, etc.). This inconsistency makes decoding words and spelling difficult for learners.
Language Units
Morphemes: The smallest meaningful units in language; can be whole words or prefixes/suffixes (e.g., "un-" for negation, "-ed" for past tense).
Syntax: Rules for combining words into correct sentences.
Example of incorrect syntax: "Craig London in his wife met" versus proper English structure: "Craig met his wife in London."
Semantics: Rules for understanding meaning within sentences.
Example: “Colourless green ideas sleep furiously” highlights semantics in thought process; syntactically correct but semantically meaningless, demonstrating distinct rules for structure and meaning.
Pragmatics
Understanding language goes beyond semantics and syntax to include pragmatics, which involves the social rules of communication and how context influences meaning. Pragmatic rules help extract deep structure from surface structure.
Grice's Maxims (1975):
Quantity: Provide as much information as necessary but not more.
Quality: Strive to speak the truth.
Relevance: Stay relevant and informative.
Manner: Be clear, brief, and orderly.
Distinctions in Understanding Language
Surface Structure: The literal order of words as they appear in a sentence.
Deep Structure: The underlying, intended meaning behind the words.
Pragmatic rules help navigate sentences with multiple possible meanings or ambiguous surface structures, revealing contextual importance and allowing us to infer the speaker's true intent (deep structure). For example, "Can you pass the salt?" is a question about ability (surface structure) but is understood as a request (deep structure) due to pragmatic rules.
Language Development
Stages of Language Acquisition in Children
Human children develop language through a consistent pattern of stages across cultures:
3-5 Weeks: Cooing begins, producing vowel-like sounds (e.g., "ooooh," "aaaaah").
4-6 Months: Babbling emerges, involving the repetition of consonant-vowel combinations (e.g., "ba-ba-ba," "ma-ma-ma").
9-12 Months: Expressive jargon (or protowords) develops, where infants produce speech-like sounds that have the intonation and rhythm of real speech but lack actual words.
10-18 Months: First words appear, typically nouns that refer to concrete objects or people (e.g., "mama," "dada," "ball").
18-24 Months: Telegraphic speech begins, characterized by two-word phrases that contain only essential words, much like a telegram (e.g., "want milk," "daddy go").
Over-generalisations
Over-generalisations (or overregularization): Children apply grammatical rules too broadly, often to irregular forms (e.g., saying "goed" instead of "went," "sheeps" instead of "sheep," "mouses" instead of "mice").
Significance: These errors reveal that children are not merely imitating what they hear. Instead, they are actively acquiring and applying underlying grammatical rules, even when those rules don't apply universally to all words in the language. This suggests an innate capacity for language learning beyond simple mimicry.
Non-Human Language
Communication Systems in Wild Environments
Non-human primates in their wild environments possess sophisticated communication systems, but these are generally not as complex as human languages.
Limited Referential Scope: Their calls are often tied to specific, immediate contexts (e.g., alarm calls for particular predators).
Lack of Arbitrariness: The connection between a sound and its meaning is often fixed and not arbitrary.
Limited Generativity and Syntax: They typically lack the ability to combine sounds or gestures in novel ways to create new meanings (generativity) or to follow complex grammatical rules (syntax).
Absence of Displacement: They cannot easily communicate about things that are not physically present or about past/future events.
Language Abilities in Captivity
Some non-human primates, particularly chimpanzees, bonobos, and gorillas in captivity, have demonstrated impressive language-like abilities, primarily through sign language or symbol boards.
Examples:
Washoe (chimpanzee): Learned American Sign Language (ASL) and could combine signs creatively (e.g., "water bird" for a swan).
Koko (gorilla): Reportedly learned over 1,000 ASL signs and understood spoken English.
Kanzi (bonobo): Learned to communicate using a lexigram board, demonstrating understanding of spoken English and combining symbols in ways that suggested some understanding of syntax.
Limitations: While these primates can acquire a significant vocabulary and combine symbols, their grammatical structures remain relatively rudimentary compared to human language. They rarely demonstrate human-level generativity or complex syntactic understanding, and their acquisition process differs from spontaneous human language learning. The extent to which this constitutes "language" rather than advanced communication remains a subject of debate.
Categorization
Usefulness of Forming Categories
The ability to form categories is fundamental to human cognition and is incredibly useful for several reasons:
Organizing Knowledge: It allows us to structure and systematize the vast amount of information we encounter, making the world more predictable.
Efficient Processing: By categorizing objects, we don't need to treat every new instance as unique, reducing cognitive effort. We can apply knowledge gained from one member to others in the same category.
Making Predictions: Categories enable us to make inferences and predictions about unfamiliar objects based on their category membership (e.g., if it's a "bird," it probably flies).
Communication: Categories provide a shared framework for communication, allowing us to refer to groups of objects with single terms.
Learning: It facilitates learning new concepts and interacting effectively with our environment.
Rejection of "Defining Features" and Acceptance of "Family Resemblance"
Psychologists have largely rejected the idea that categories are defined by a strict set of "defining features" (necessary and sufficient conditions that every member must possess).
Reason for Rejection: Many real-world categories lack clear-cut boundaries. For example, it's difficult to find a single set of features that defines "game" (some have players, some don't; some are competitive, some aren't).
Instead, the concept of "family resemblance", proposed by Ludwig Wittgenstein, is favored.
Family Resemblance: Members of a category share a set of characteristic features, but no single feature is shared by all members. Instead, members overlap in various ways, much like members of a family might share some features (e.g., nose shape, hair color) but no single feature is common to everyone. This approach better accounts for the graded structure and fuzzy boundaries of most natural categories.
Levels of Categorization and Basic Level Categories
Things can be thought of as members of categories at many different levels of specificity, forming a hierarchical structure:
Superordinate Level: Very general categories (e.g., "furniture," "vehicle," "animal"). These provide less specific information.
Basic Level: The most natural, intuitive, and cognitively privileged level of categorization (e.g., "chair," "car," "dog").
Subordinate Level: Very specific categories (e.g., "dining chair," "sports car," "golden retriever"). These provide a lot of specific information but might be less frequently used in everyday language.
Basic Level Categories:
Are the first categories children learn.
Are typically named first and most frequently by adults.
Maximize within-category similarity (members look very similar) while minimizing between-category similarity (members of different basic categories look quite distinct).
Provide an optimal balance between informativeness and distinctiveness, making them the most efficient level for everyday interaction and communication.
Problem Solving
Ill-defined vs. Well-defined Problems
Problems can be classified based on the clarity of their components:
Well-defined Problems:
Have a clearly specified initial state (where you are now).
Have a clearly specified goal state (what you want to achieve).
Have a set of clearly defined operators or rules (what actions you can take to move from the initial to the goal state).
Examples: Solving a crossword puzzle, completing a Sudoku, solving a chess problem, or an algebra equation. The path to a solution, even if difficult, is constrained by known rules.
Ill-defined Problems:
Lack a clear initial state, goal state, or a complete set of operators/rules. The problem itself might be ambiguous.
Requires more creativity, insight, and defining the problem elements yourself.
Examples: Writing a novel, choosing a career, finding happiness, designing a new product, or addressing climate change. The "solution" is often open to interpretation.
Algorithms vs. Heuristics
Humans use different strategies to solve problems:
Algorithms:
Are step-by-step procedures that, if followed correctly, guarantee a solution to a problem.
They involve a systematic search through all possible options or calculations.
Examples: A mathematical formula, a recipe for baking a cake, or systematically trying every possible combination for a lock.
Advantage: Guaranteed correctness.
Disadvantage: Can be time-consuming or computationally intensive, especially for complex problems.
Heuristics:
Are mental shortcuts or "rules of thumb" that provide a quick and efficient way to arrive at a solution, but do not guarantee optimality or correctness.
They reduce the number of possible solutions to consider, saving cognitive effort.
Examples: Trial and error, working backward from the goal, or breaking a large problem into smaller sub-problems.
Advantage: Faster and less effortful.
Disadvantage: Can lead to errors or biased judgments.
The Availability Heuristic
Availability Heuristic: A cognitive bias where people estimate the probability or frequency of an event based on how easily examples or instances come to mind (i.e., their "availability" in memory).
How it Leads to Incorrect Judgments:
If vivid, dramatic, or recently encountered events are more easily recalled, people may overestimate their actual likelihood, even if they are statistically rare.
Example: After seeing extensive news coverage of plane crashes, people might overestimate the risk of flying and underestimate the risk of driving, even though statistically, driving is far more dangerous. The vividness of the plane crash images makes them more "available" in memory.
Example: People might judge words starting with ‘r’ to be more common than words with ‘r’ as the third letter because it's easier to retrieve examples of the former, despite the latter being statistically more frequent.