Spring 00, Exam 2

Psy 5054 ]


Part 1: Multiple Choice. Circle the letter corresponding to the correct answer. Only one answer is correct for each question. (1 point each)

1. Seidenberg & McClelland's (1989) parallel distributed processing model of word recognition and naming predicts that orthographic-phonological regularity will have less effect on naming times for _____ than for _____.
(a) words with large neighborhoods/words with small neighborhoods
(b) high frequency words/low frequency words
(c) non-words with large neighborhoods/non-words with small neighborhoods

2. The acoustic properties of many speech sounds vary with context. For example, the /d/ sound in the word "dig" has very different properties than the /d/ sound in the word "dug". This illustrates the _____ problem that characterizes speech perception.
(a) segmentation
(b) invariance under paraphrase
(c) lack of invariance

3. A _____ is the smallest unit of meaning in language.
(a) proposition
(b) morpheme
(c) phoneme

4. The study of how speech sounds are produced is called _____.
(a) acoustic phonetics
(b) articulatory phonetics
(c) phonology

5. Native speakers of Japanese have difficulty hearing the difference between the English words "lag" and "rag". This illustrates _____.
(a) gang effects
(b) categorical perception
(c) the word superiority effect

6. _____ explicitly assumes that the structures and processes underlying language comprehension are independent of the structures and processes that mediate other cognitive abilities.
(a) Forster's (1981) model of lexical access and sentence comprehension
(b) Seidenberg & McClelland's (1989) model of lexical access and naming
(c) McClelland & Rumelhart’s (1981) Interactive Activation Model

7. If Rumelhart and McClelland (1982) had found that letters are just as difficult to recognize in unpronounceable non-words that share letter combinations with real words (e.g., "SLNT") as they are in unpronounceable non-words that don't share letter combinations with real words (e.g., "XLQJ"), then _____ would have been falsified.
(a) the whole word model of reading
(b) the Interactive Activation Model
(c) PANDEMONIUM

8. McClelland and Rumelhart (1981) found that the difference in activation levels between high- and low-frequency words in their Interactive Activation Model increases over processing cycles when both are equally consistent with the input. They refer to this as the _____ effect.
(a) gang
(b) rich-get-richer
(c) pseudo-word superiority

9. Patients with surface dyslexia have no problem with regular words (e.g., "like" and "cane") but are unable to read irregular words (e.g., "said" and "lose"). This suggests that they have lost the _____ route proposed by _____ models of reading (e.g., Coltheart, 1978).
(a) indirect/whole word
(b) indirect/dual route
(c) direct/dual route

10. Miller & Isard (1963) demonstrated that _____ constraints improve auditory word recognition.
(a) syntactic
(b) semantic
(c) both syntactic and semantic

Part 2: Definitions. In just 1 or 2 sentences, give an operational definition for each of the following concepts. Your definition may come from an experiment you are familiar with or you may make up your own definition (as long as it accurately defines the concept and is operational). (2 points each)

Grading Criteria:

  • 1 pt. for correctly identifying the concept
  • 1 pt. for using a procedural definition

11. The Neighborhood of the Word "Bike"

The neighborhood of the word "bike" can be defined as all the monosyllabic words in the latest edition of Webster's Dictonary that end in "ike". This would include words like "hike," "mike" and "trike."

12. Naming Time

I would present participants with words on a computer screen and ask them to quickly and accurately pronounce each word as it appeared. Naming Time could then be defined a the time (in msec.) that elapsed between the presentation of the letters and the detection of a response by a voice-activated relay.

13. Sentence Verification Time

As in a typical sentence verification experiment, I would ask show participants some sentences on a computer screen and ask them to indicate whether each is true or false by pressing a YES or NO button as quickly as possible. Sentence Verification Time could then be defined as the time that elapses between the presentation of a sentence, and a correct button press.

14. Lexical Decision Reaction Time

I would present participants with letter strings on a computer screen and ask them to press one key if the letters form a word, or another key if they do not. Lexical decision reaction time could then be defined a the time between the presentation of the letters and the pressing of a key (as measured by the computer).

15. The Dominant Meaning of an Ambiguous Word

To determine the dominant meaning of an ambiguous word such as "bug" I would ask 50 people to use it in a sentence, then I would count the number of sentences that (in my judgement) made use of each meaning listed in the latest edition of Webster's Dictonary (e.g., "insect" and "listening device").  If one meaning was used in at least 30 of the sentences (60% of the total) I would take that to be the dominant meaning, otherwise there would be no dominant meaning for the word.

Part 3: Short Essay. Answer each of the following questions using no more than half of a page for each. (5 points each)

16. In a form of dyslexia called "letter-by-letter" reading, patients are said to read words one letter at a time, rather than recognizing words all at once. Design an experiment to test this claim. Be sure to specify, using operational definitions and/or examples where they are appropriate, what your independent and dependent variables would be. What pattern of results would you expect to see if this claim is true? What pattern would you expect to see if it is false?

Grading Criteria:

  • 1 pt. for operational definition of I.V.
  • 1 pt. for operational definition of D.V.
  • 1 pt. for correct prediction if claim is true
  • 1 pt for correct prediction if claim is false
  • 1 pt. for overall coherence of the answer

Example Answer:

To test this claim I would ask patients diagnosed with letter-by-letter reading to read a short narrative (about one page long) and while crossing out all ofthe t's. I would then ask them questions about the narrative just to make sure that they had understood it. Next, I would use word frequency norms todetermine the frequency of each word (occurrences per million words) in the English language. I would then divide all of these words into two groups:  those with frequencies above the median for the text (High Frequency Words) and those with frequencies below the median (Low Frequency Words). Thiswould be my independent variable. For each participant I would then calculate the percentage of t's detected in High and Low Frequency Wordsrespectively. This would be my dependent variable. If the claim is true, I would expect to find no differences in t-detection between High and Low Frequency Words. If the claim is false, I would expect to find that participants detect a higher percentage of t's in low frequency words than in high frequency words. This follows from Healy's (1976) finding that the probability of overlooking a target letter in a word increases as the process of recognizing that word becomes more automatic.   

17. Some psychologists have argued that morphologically complex words (such as "unhappy") are not represented in the lexicon. They argue that such words are recognized by accessing and then combining their constituent parts (e.g. "un" and "happy"). Design an experiment to test this hypothesis. Be sure to specify (using operational definitions) what your independent and dependent variables would be. What pattern of results would you expect to see if the hypothesis is true? What pattern would you expect to see if it is false? (Hint: You may wish to consider frequency effects.)

Grading Criteria:

  • 1 pt. for identification of I.V.
  • 1 pt. for identification of D.V.
  • 1 pt. for correct prediction if hypothesis is true
  • 1 pt. for correct prediction if hypothesis is false
  • 1 pt. for overall coherence of the answer

Example Answer:

To test this hypothesis I would first generate two lists of morphologically complex words. One list would include only high frequency words (among the 2,000 most common words in the English language according to published norms) while the other list would include only low frequency words (not among the 2,000 most common English words). The frequency of the morphemes out of which the words are constructed would be identical for the two lists. I would then present the words to participants, one at a time, on a computer monitor and ask them to pronounce each word out loud as quickly as possible. I would use the computer to measure the time that elapses between the presentation of each word, and the detection of a naming response. This would be my dependent variable. The independent variable would be word frequency (high versus low). If the hypothesis is true, the average naming time should be the same for both groups of words (because of the equal frequencies of their parts). If the hypothesis is false, high frequency words should exhibit shorter naming times than low frequency words.

 

Psy 5054 ]

The views and opinions expressed in this page are strictly those of the page author. The contents of this page have not been reviewed or approved by the University of Minnesota

This page was last updated on 04/13/00.