Study Guides (248,518)
Canada (121,606)
Psychology (579)
PSYC 340 (9)

psych Lang after mid2 part 2.docx

12 Pages
46 Views
Unlock Document

Department
Psychology
Course
PSYC 340
Professor
Debra Ann Titone
Semester
Winter

Description
Phonological Encoding Retrieving the sound of words and the second part of lexicalisation Put them in an order we don’t know… Sequence sounds… Main way they defer: 1. Assume a separation of structure 2. Separation of content Two Theories: •Retrieving the sounds of words –Second stage of lexicalization •Problem of sound sequencing –Selecting phonemes to pronounce the words –Phonemes must be put in the correct order How to order the sounds? •Sequence structure and content (sounds) retrieved separately: suggest we have phonological frames for each word that we know st nd rd e.g: 3 sounds= 1 slot for a consonant, 2 for a vowel and 3 for the last consonant->hallow frame then retrieve sound separately then put sounds into the frame Structure Phonological Frame [___] [___] [___] C V C onset nucleus coda Content Phonemes /d/ /o/ /g/ Two mechanisms using this idea: –Scan-copier mechanism –Competitive queing •No distinction between structure and content: –Recurrent networks –Lemma model Scan-copier: Start with a phonological frame for a word->1 slot for a consonant, 2 for ad rd vowel and 3 for the last consonant. Then retrieve phoneme separately and then you just put the phonemes into the slots into a right and left serial order. Checks off every phoneme as its used. Explains speech errors 1. Select the wrong sound and incorrectly check it off 2. Check the sound off as used incorrectly e.g: Good at explaining the dif. sound error that we do. Competitive queuing idea: Gives us a way that the sounds could actually go into the frames. Maybe phonemes are “queued up” or lined up based on how strongly they are associated with the beginnings and endings of words. Initiation units and end units that correspond to the beginning of a word or the ending of a word. Sounds have dif. connection stren to these units. e.g: d has a strong connection to the initiation the g a strong connection to the end and the o a weaker connection to these units. O loses this competion for the beginning and the end. Then these go into the frame  Idea gd at explaining errors o Fill in sounds-temporally supressed if that goes wrong you have the sound being used again  Might not be necessary to distinguish between the structure and the content because this kind of does the work for you. All you need is these associations.  Idea that the recurrent network runs with Takes a list of words as its input and takes all of these words and extracts rules and then it outputs phoneme sequences. No structure content distinction-structure based on these rules on how to structure phonemes. Internal and External feedback-> learning from experience Lemma Model •Sounds retrieved as whole, ordered sequences “Horses” -> Lemma: selected before /horse/+/iz/ sounds are retrieved /h, o, r, s/ /i, z/ -> Lexeme: ordered sets of phonemes [hor] [siz] -> Syllables: phonemes strung together to create syllables Phonetic encoding  Don’t need to sequence the sounds they are already sequenced in this model Summary •We may order sounds by inserting them into a separate structural frame •We may order sounds by learning the sound-ordering rules of the language Interactive Processing Lower lvl processing could feed back to the High processes Reverse flow of information •Lexicalization (2stages) 1. –Meaning to word 2. –Word to sound •Feedback –Sound to word? (Info. coming back from sound lvl to word) Sound to Word •Speech errors –Repeated phoneme effect •Phonemes are more likely to exchange when nearby phonemes are the same E.g: “kit to fill” (fit to kill) – more likely (two things that exchange are next to an i) “kit to fall” (fit to call) – less likely (two things that exchange are next to words that are dif.) Idea: word fit-> feeding info. down, activating sounds, “I” sound is going to feed back up to nearby words ->kill and then kill is going to activate its own sets of phoneme Less likely to exchange call because they don’t fit with any of the same phonemes. Sound to Word •Speech errors –Mixed word substitutions: errors and targets both semantically and phonologically similar e.g: “oyster” (lobster) Intending to say lobster but saying oyster “cat” (rat) –Word exchanges (tend to be similar in terms of syntax class): phonological similarity (happens at a rate that is greater than chance) e.g: “He eats yoga and does yogurt.” Spontanious-> could happen more often because leading is actually a word. Experimentally-> every word starts with a b and every second word starts with a d so it biases you and “barn door” are actual words “Bart board”-> Bart is not word outside of ‘The Simpsons’ So
More Less

Related notes for PSYC 340

Log In


OR

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


OR

By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.


Submit