User:Morbidpoet

Introduction
When dealing with the issue of how children learn language, most psycholinguists seem to agree on that there are two basic tasks facing infants and young children (Hall, 2004):
 * 1) Pulling out words and other linguistic elements (such as prosodic factors) out of continuous speech. To get a basic idea of how hard this is, think of the last time you heard a foreign language spoken around you. Unless you were familiar with a small part of the language, it is likely that you would think they are speaking very fast, and you would probably have trouble picking out individual words in a sentence.
 * 2) two Having broken up the speech into individual words, children must then decide what each of the words refer to in their environment (or outside their environment). Simply put, they must discern the meaning of the word.

For the first one, the obvious problem for infants learning a language is the speed at which native speakers speak, as well as the relatively few pauses they use. To get a basic idea of how hard this is, think of the last time you heard a foreign language spoken around you. Unless you were familiar with a small part of the language, it is likely that you thought they are speaking very fast, and you would probably have had trouble picking out individual words in a sentence. The second task (and the main focus of this website), is complicated by just how many possible referents exist in the world that a speaker could be referring to, as most humans do not limit themselves to talking about only the here and now. For example, someone inside might say “it’s raining today” and unless this person was looking and pointing outside, the infant would try and apply that verb to something in their immediate environment.

New Model for Learning
The problem with many of the older models of word learning is that they tend to restrict their focus to merely one aspect of word learning; the speaker’s intentions, the environment of the speaker or the probabilities of words. However, these theories can seem very simplistic when one considers the problem mentioned in the intro; most humans do not talk about what is only in their immediate environment. Given these issues, Frank, Goodman and Tenenbaum (2009) have created a new model of the learning of word meaning, which they called the intentional model. Their model was created to take into consideration both the cross-situational and social theories of word learning. The theory rests on several variables- what the speaker intends to refer to, the learner’s lexicon of the language, the actual words of the speaker and where the physical situation of the learner- all of which are related by probabilities based on two assumptions: first, the intention of the speaker comes from their environment (the world around them) and second, that the words of the speaker are the translation from intention (or thoughts) into words of their language. The diagram of the model looks like this: