Introducing psycholinguistics –

Psycholinguistics

• Speech is one of the natural activities that we, as humans, engage in. Unlike many other
learned activities, virtually all humans learn a language and any human baby can learn
any human language; not dependent on what language parents spoke.
• Only humans learn language. It is one of the unique abilities that distinguish us from all
other animals.
• In Understanding more about how the linguistic system functions and how it is related to
other cognitive systems, be they other characteristics unique to humans or traits shared by
virtually every animal, furthers our understand how we function.

Psycholinguistics is the study of the mental aspects of language and speech. It is primarily

concerned with the ways in which language is represented and processed in the brain.

A branch of both linguistics and psychology, psycholinguistics is part of the field of cognitive

science. Adjective: psycholinguistic.

The term psycholinguistics was introduced by American psychologist Jacob Robert Kantor in his

1936 book, "An Objective Psychology of Grammar." The term was popularized by one of
Kantor's students, Nicholas Henry Pronko, in a 1946 article "Language and Psycholinguistics:
Pronunciation: si-ko-lin-GWIS-tiks

Also known as: Psychology of language

Etymology: From the Greek, "mind" + the Latin, "tongue"

On Psycholinguistics

"Psycholinguistics is the study of the mental mechanisms that make it possible for people to use
language. It is a scientific discipline whose goal is a coherent theory of the way in which
language is produced and understood," says Alan Garnham in his book, "Psycholinguistics:
Central Topics."

Two Key Questions

According to David Carrol in "Psychology of Language," "At its heart, psycholinguistic work
consists of two questions. One is, what knowledge of language is needed for us to use language?
In a sense, we must know a language to use it, but we are not always fully aware of this
knowledge.... The other primary psycholinguistic question is what cognitive processes are
involved in the ordinary use of language? By 'ordinary use of language,' I mean such things as
understanding a lecture, reading a book, writing a letter, and holding a conversation. By
'cognitive processes,' I mean processes such as perception, memory, and thinking. Although we
do few things as often or as easily as speaking and listening, we will find that considerable
cognitive processing is going on during those activities."

Psycholinguistics
Ratner, J.B. Gleason, in Encyclopedia of Neuroscience, 2004
Psycholinguistics is the discipline that investigates and describes the psychological processes
that make it possible for humans to master and use language. Psycholinguists conduct research
on speech development and language development and how individuals of all ages comprehend
and produce language. For descriptions of language, the field relies on the findings of linguistics,
which is the discipline that describes the structure of language. Although the acquisition,
comprehension, and production of language have been at the core of psycholinguistic research,
the field has expanded considerably since its inception: The neurology of language functioning is
of current interest to psycholinguists, particularly to those studying sex differences, aphasia,
language after congenital or acquired injury to the immature brain, and developmental disorders
of language (dysphasia). Some psycholinguists have also extended their interests to experiments
in nonhuman language learning (e.g., gorillas and chimpanzees) to discover if language as we
know it is a uniquely human phenomenon.
Psycholinguistics
Psycholinguistics combines methods and theories from psychology and linguistics. It attempts to
evaluate the psychological reality and underpinnings of linguistic rules and processes. It also
seeks to link word and sentence processing to the deeper expressive processes of message
construction and interpretation. Psycholinguistic experiments typically use reaction-time
methodology to examine language comprehension and production as online processes. This
research has shown that words and sentences are constructed in an overlapping, incremental
fashion. This incrementalism allows us to begin producing words – even before we have
formulated the details of later parts of our utterances. In order to construct formal models of this
incremental processing, psycholinguists have used both modular accounts that emphasize back-
tracking and neural network accounts that emphasize smooth ongoing interactions between
constraints. These models are then further elaborated to account for the process of language
acquisition, as well as variations in language processing that arise between different languages.
Modular theories typically propose a fixed core of processing strategies that apply across
languages and are available even to young children. Interactive theories tend to emphasize the
extent to which the structures of particular language emerge during the process of learning, as a
response to specific structural cues provided by the native language.
How Language Is Done

In the book, "Contemporary Linguistics," linguistics expert William O'Grady explains,

"Psycholinguists study how word meaning, sentence meaning, and discourse meaning are
computed and represented in the mind. They study how complex words and sentences are
composed in speech and how they are broken down into their constituents in the acts of listening
and reading. In short, psycholinguists seek to understand how language is done... In general,
psycholinguistic studies have revealed that many of the concepts employed in the analysis of
sound structure, word structure, and sentence structure also play a role in language processing.
However, an account of language processing also requires that we understand how these
linguistic concepts interact with other aspects of human processing to enable language
production and comprehension."

An Interdisciplinary Field
"Psycholinguistics... draws on ideas and knowledge from a number of associated areas, such
as phonetics, semantics, and pure linguistics. There is a constant exchange of information
between psycholinguists and those working in Neurolinguistics, who study how language is
represented in the brain. There are also close links with studies in artificial intelligence. Indeed,
much of the early interest in language processing derived from the AI goals of designing
computer programs that can turn speech into writing and programs that can recognize the human
voice," says John Field in "Psycholinguistics: A Resource Book for Students."

Natural language processing is the use of computers for processing natural language text or
speech. Machine translation (the automatic translation of text or speech from one language to
another) began with the very earliest computers (Kay et al. 1994). Natural language interfaces
permit computers to interact with humans using natural language, for example, to query
databases. Coupled with speech recognition and speech synthesis, these capabilities will become
more important with the growing popularity of portable computers that lack keyboards and large
display screens. Other applications include spell and grammar checking and document
summarization. Applications outside of natural language include compilers, which translate
source code into lower-level machine code, and computer vision (Fu 1974, 1982).
Most natural language processing systems are based on formal grammars. The development and
study of formal grammars is known as computational linguistics. A grammar is a description of a
language; usually it identifies the sentences of the language and provides descriptions of them,
for example, by defining the phrases of a sentence, their interrelationships, and perhaps also
aspects of their meanings. Parsing is the process of recovering a sentence's description from its
words, while generation is the process of translating a meaning or some other part of a sentence's
description into a grammatical or well-formed sentence. Parsing and generation are major
research topics in their own right. Evidently, human use of language involves some kind of
parsing and generation process, as do many natural language processing applications. For
example, a machine translation program may parse an input language sentence into a (partial)
representation of its meaning, and then generate an output language sentence from that
representation.
Modern computational linguistics began with Chomsky (1957), and was initially dominated by
the study of his ‘transformational’ grammars. These grammars involved two levels of analyses, a
‘deep structure’ meant to capture more-or-less simply the meaning of a sentence, and a ‘surface
structure’ which reflects the actual way in which the sentence was constructed. The deep
structure might be a clause in the active voice, ‘Sandy saw Sam,’ whereas the surface structure
might involve the more complex passive voice, ‘Sam was seen by Sandy.’
Transformational grammars are computationally complex, and in the 1980s
several linguists came to the conclusion that much simpler kinds of grammars could describe
most syntactic phenomena, developing Generalized Phrase-Structure Grammars (Gazdar et
al. 1985) and Unification-based Grammars (Kaplan and Bresnan 1982, Pollard and Sag 1987,
Shieber 1986). These grammars generate surface structures directly; there is no separate deep
structure and therefore no transformations. These kinds of grammars can provide very detailed
syntactic and semantic analyses of sentences, but even today there are no comprehensive
grammars of this kind that fully accommodate English or any other natural language.
Natural language processing using handcrafted grammars suffers from two major drawbacks.
First, the syntactic coverage offered by any available grammar is incomplete, reflecting both our
lack of understanding of even relatively frequently occuring syntactic constructions and the
organizational difficulty of manually constructing any artifact as complex as a grammar of a
natural language. Second, such grammars almost always permit a large number of spurious
ambiguities, that is, parses which are permitted by the rules of syntax but have unusual or
unlikely semantic interpretations. For example, in the sentence ‘I saw the boat with the
telescope,’ the prepositional phrase ‘with the telescope’ is most easily interpreted as the
instrument used in seeing, while in ‘I saw the policeman with the rifle,’ the prepositional phrase
usually receives a different interpretation in which the policeman has the rifle. Note that the
corresponding alternative interpretation is marginally accessible for each of these sentences: in
the first sentence one can imagine that the telescope is on the boat, and in the second, that the
rifle (say, with a viewing scope) was used to view the policeman.
In effect, there is a dilemma of coverage. A grammar rich enough to accommodate natural
language, including rare and sometimes even ‘ungrammatical’ constructions, fails to distinguish
natural from unnatural interpretations. But a grammar sufficiently restricted so as to exclude
what is unnatural fails to accommodate the scope of real language. These observations led, in the
1980s, to a growing interest in stochastic approaches to natural language, particularly to speech.
Stochastic grammars became the basis of speech recognition systems by outperforming the best
of the systems based on deterministic handcrafted grammars. Largely inspired by these
successes, computational linguists began applying stochastic approaches to other natural
language processing applications. Usually, the architecture of such a stochastic model is
specified manually, while the model's parameters are estimated from a training corpus, that is, a
large representative sample of sentences.
As explained in the body of this article, stochastic approaches replace the binary distinctions
(grammatical vs. ungrammatical) of nonstochastic approaches with probability distributions.
This provides a way of dealing with the two drawbacks of nonstochastic approaches. Ill-formed
alternatives can be characterized as extremely low probability rather than ruled out as
impossible, so even ungrammatical strings can be provided with an interpretation. Similarly, a
stochastic model of possible interpretations of a sentence provides a method for distinguishing
more plausible interpretations from less plausible ones.

Questions

1. What is the psycholinguistic theory?

2. What is the scope of psycholinguistics?

3. What are the branches of psycholinguistics?
4. What is the importance of psycholinguistics?
5. What are the 5 stages of language acquisition?

6. What are the two main theories of language acquisition?

7. What are the 3 theories of language acquisition?