Lexicon 2 Review Sheet Answers

Answer the following questions in the space provided.

1. How was the tip-of-the-tongue phenomenon proved to be a more phonologically based retrieval problem than meaning based retrieval problem? What does this tell us about retrieval of words in our mental lexicon?

TOTs occur when we want to say a word, but the word doesn't come to us. Brown and McNeill presented definitions of infrequent words, such as sampan, and asked subjects to produce the defined word. Subjects were more likely to approximate the target words with similar sounding words than with similar meaning words. In this case, subjects were more likely to say sarong for sampan than they would say "houseboat" or "yacht" which have related meanings. Hence, this suggests that we sometimes activate words by their sounds.

2. Describe a confrontational naming task.

Individuals see a picture of an object and have to come up with the name. In such a task, the experimenter will measure either the reaction time or the success rate (or both).

3. Define lexical ambiguity. How is phoneme monitoring used to examine the processing of ambiguous words?

Lexical ambiguity occurs when a single word may be interpreted to have more than one meaning. E.g. righting and writing.

In a phoneme monitoring task, subjects are given a phoneme to look for and they have to respond any time they come across it. Foss (1970) presented subjects with sentences containing ambiguous words. E.g The man started to drill before the truck arrived. In this case he asked them to monitor for /b/ as in before which came after the ambiguous word drill. He observed that the reaction time increased and attributed this result to the process of activating more than one meaning of an ambiguous word.

However, in a later study the time between the ambiguous word and the phoneme that was monitored was varied which found that the increased processing load with lexical ambiguity was very short lived. If the phoneme was delayed by as long as two syllables, the increased processing time for ambiguous words disappeared. These results suggest that although multiple meanings of an ambiguous word are briefly entertained, the ambiguity is quickly resolved.

4. Describe how context could possibly affect the activation of multiple word meanings for an ambiguous word making reference to bottom-up, top-down processing and cross-modal lexical decision tasks.

Can the context (top-down) that is biased toward one or another meaning of an ambiguous word selectively activate the appropriate meaning (bottom-down)? Do we activate inappropriate word meanings even when there is a contextual reason not to do so? Swinney (1979) examined the question with a cross-modal lexical decision task. A cross-modal task is one where the stimulus and the prime came in different modalities. So people heard a sentence, and at the same time they saw either a real word or a nonsense word come up on a computer screen. They had to decide whether the item on the computer screen was a real word or not.

Swinney found that decision times for visual words related to either meaning of the ambiguous word were shorter than for unrelated words when the visual words immediately followed the ambiguity. When the visual words were presented four syllables after the ambiguity, however, only the contextually appropriate meaning was facilitated. This suggests that even in the presence of a strong biasing context, multiple meanings of ambiguous words are briefly activated.

Work by Hogaboam and Perfetti (they asked subjects to compare sentences with dominant vs. secondary meanings and asked them to decide whether final word was ambiguous) indicate that dominant meaning is always activated and secondary meaning is activated when necessary, or when it is very similar in strength to dominant. When there is a strong frequency bias towards one word, and a strong sentential bias, the secondary meaning may not be activated at all.

5. Define recoding and chunking. Thinking about recoding, how does the use of long term memory fascilitate a short term memory task?

Chunking involves reducing the size of an array by grouping information into units that are easier to code. Recoding involves assigning these units codes that facilitate recovery of position as well as item information. When we recode, we can probably remember long strings of numbers or words but we need access to LTM to do it. E.g. you need to know what the College Park zip code is to start with your recoding process.

6. Describe some of the limits to recalling things from working memory.

There are limits in terms of amounts of information, limits in terms of time (the number of words people could recall back correlates with their reading rate), and limits in terms of processing resources. If a task is more difficult (uses more processing resources), then there is less space available for storage, if the task is easy, you can store more.

7. What is the difference between surface and deep structure?

The surface structure of a sentence is represented by the words you actually hear spoken or read. Deep structure is the underlying structure of a sentence that conveys the meaning of a sentence.

8. Define parsing. Explain how parsing relates to modular and interactive models of processing.

Assigning words in a sentence to their relevant linguistic categories is called parsing. There are two views on how we parse sentences. One is the modular view. In this case, we use the syntactic module to parse and we are not influenced by higher order contextual variables such as meaning of sentence. The other view is the interactive view of parsing. In this case we use all available information including syntax, lexical, discourse and nonlinguistic info such as context.

9. What is the minimal attachment principle?

We prefer to attach new items into the phrase marker being constructed using the fewest syntactic nodes consistent with the rules of the language.

Answer the following multiple-choice questions.

1. a

2. b

3. c

4. b

5. c