This preview shows page 1. Sign up to view the full content.
Unformatted text preview: Cognitive Neuroscience Where is language? Case Studies Investigation the neural bases of language Language is typically highly lateralized to one hemisphere (typically left) Many areas within the left hemisphere (as well as some within the right) have been found to be active in different language tasks Case studies Method Lexical Peterson et al. (1988) Damasio et al. (1993) Shapiro et al. (2001) ASL Neville, et al. (1998) Hickok et al. (2001) PET Lesion TMS fMRI Lesion Word Knowledge: Localization of function? Luxurious seagoing vessel Noun, plural +s YACHT /y a t/ Peterson, et al. (1988) Experimental conditions Fixation Passive words (visual/aud input) Saying words (visual/aud input) Semantic association Data analysis Stimulation condition Task: Stim: vs. Comparison condition Fixation cross Function revealed Passive words spoken written Say words spoken written Say verb written noun sensory processing, compre spoken words (phono) written words (ortho) Articulatory coding, motor programming Task: Stim: Passive words Task: Stim: Read aloud Semantic association Methods PET 17 subjects: 1849 years, old, native English speakers, strongly righthanded Tracer: water labeled with Oxygen15 Acquisition: 7 parallel slices with centerto center distance of 14.4 mm Tasks presented in blocks of 40 s Presentation: one stimulus/second Results Stimulation condition Task: Stim: vs. Comparison condition Fixation cross Function revealed Passive words written written words (orthography) Results Stimulation condition Task: Stim: vs. Comparison condition Fixation cross Function revealed Passive words spoken spoken words (phonologycomprehension) Results Stimulation condition vs. Comparison condition Function revealed Task: Stim: Say words Passive spoken words spoken Articulatory coding, motor programming (phonologyproduction) Results Stimulation condition vs. Comparison condition Function revealed Task: Stim: Say verb written noun Read aloud Semantic association (verbs) Damasio, et al. (1993) 2 Tasks (1) Noun retrieval: b&w line drawings and photos of objects from the categories animals, fruits/vegetables, tools/utensils (n=400) (2) Verb retrieval: b&w line drawings and photos of actions Nouns and verbs did not differ in the frequency or length Normal subject controls carried out the task to establish correct responses Damasio, et al. (1993) Subject Boswell AN1033 KJ1360 Controls 24 51 91 92 nouns Animals Fruits Tools verbs 76 70 95 94 92 96 53 95 25 54 88 92 Damasio, et al. (1993) Subject Boswell Stimulus duck penguin pineapple zebra ostrich raccoon zebra pumpkin KJ1360 cutting conducting digging Error responses bird bird possibly vegetable horse bird that sticks head in sand animal that washes food horselike animal with black & white stripes melon....use it in Halloween going.....scissoring band director getting ready to move dirt AN1033 Damasio, et al. (1993) Deficits: "double dissociation" of noun and verb production the spoken language system is instantiated in the brain in such a manner that words of one grammatical category may be selectively damaged Lesions? Grammatical distinctions in the left frontal cortex (Shapiro, et al., 2001) Questions: (Expt 1) Can different substrates be identified with TMS for nouns and verbs? (Expt 2) Do the different neural substrates associated with nouns vs. verbs reflect: grammatical category differences (noun/verb) OR semantic differences (object/action)? Shapiro, et al. Background: Practice with a task decreases time to respond (facilitation) Will TMS applied over left prefrontal cortex disrupt this facilitation effect equally for nouns and verbs? 1) Baseline measurement of speed for nouns and verbs 2) Repeat task with sham TMS (coils perpendicular to head) 3) Repeat task with real TMS 3) Observe the size of the "practice" or facilitation effects for nouns/ verbs with sham vs real TMS Methods 8 Righthanded native speakers of English rTMS: 1Hz frequency for 300 sec inferior midfrontal gyrus, anterior and superior to Broca's area 6 cm anterior and 1cm ventral to motor areas-- supposed verb areas (e.g. Damasio et al) Methods Task: a written word (noun or verb) for 250 msec: (e.g.,CHOICE or LOSE) Symbolic cue indicating what to produce: Singular() or plural() (in third person for verbs) Choice Choices (he/she) Loses (we) Lose Measured voice response latency Methods (cont.) Experiment Structure 3 conditions: baseline, sham, real Baseline trials involve the same stimuli (no stimulation) carried out BEFORE stimulation trials Purpose: to determine baseline reaction time before practice Stimulation blocks (sham then real): administer 300 secs of sham or real rTMS Present a block (80 trials) of only nouns or verbs and subjects perform the task 10 minutes of rest between blocks Experiment 1 Determine if the stimulated area is critical for verb vs. noun processing If so, the facilitation effect for verbs (but not nouns) after rTMS should be reduced relative to the sham condition Difference in RT between baseline and experimental condition Findings: the only sig difference is: Verbs: sham/real Conclusion: The effect of TMS is restricted to verbs Experiment 2 Determine if the verbspecific effects are due to their grammatical or semantic properties Ask subjects to the same task with pseudowords (they have no meaning)! E.g. WUG > WUGS Predictions: If grammatical properties are relevant, then pseudoverbs should be affected by rTMS more than pseudonouns Methods Task: a written pseudoword for 250 msec: WUG Blocks of "nouns" or "verbs" Symbolic cue indicating what to produce: Singular() or plural() for nouns rd 3 person singular/plural for verbs Measured voice response latency "WUGS" The effects of TMS were restricted to verbs Conclusions " "at least with respect to....grammatical category, nouns and verbs have distinct neuroanatomical underpinnings, and can be dissociated by targeted suppression of the left prefrontal cortex" Sign Language: fMRI Question: Is the left hemisphere specialized for the processing of spoken language? Or For the grammatical coding of information? Which neural substrates support ASL (American Sign Language)? A fully grammatical language which is NOT a translation of spoken English (different syntax, vocabulary, etc.) ASL has contrastive features: configuration place of articulation movement Recall that: spoken phonology has contrastive features: voicing place of articulation manner Neville et al. (1998) Examine brain activation in: (a) Congenitally deaf, native ASL learned English late and imperfectly and without auditory input children of deaf parents who know no ASL (b) Normally hearing, bilingual ASL and English (c) Normally hearing, monolingual English speakers Neville et al. (1998) Two fMRI scanning conditions: (1) English material (presented one word/string at a time in the center of the screen) sentences consonant strings (2) ASL material (film of a native deaf signer): ASL sentences nonsign gestures physically similar to ASL signs Behavioral task: At the end of each run, yes/no recognition questions Signed sequences vs. Nonsense sequences Written sentences vs. Consonant strings Neville et al. (1998) Results: (1) All Ss processing their native language had sig. activation in classical LH language structures (more robust in anterior vs. posterior regions) (2) Sign language users display RH activation as well (3) Deaf sign language users did not display LH activation in reading English (4) Hearing sign language users displayed LH anterior, but not posterior, activation in reading English. Neville et al. (1998) Conclusions: "The early acquisition of a fully grammatical, natural language is important in the specialization of these systems" " strong biases...render regions of the left hemisphere well suited to process a natural language independently of the form of the language" Hickok et al. (2002) Examined effect of unilateral lesion on ASL comprehension in native signers LH Damage: 11 deaf RH Damage: 7 deaf, 1 hearing Comprehension Tasks in ASL: Single signs (subject points to picture corresponding to sign) Commands carried out in the context of array of tokens with different colors / shapes: Simple sentences ( e.g., "point to the white circle") Complex sentences (e.g., "touch all the circles except the yellow circle") Hickok et al. (2002) Results Conclusion: LH dominance in sign language comprehension. The eloquent brain ...
View Full Document
This note was uploaded on 07/29/2008 for the course NEUROSCIEN 70 taught by Professor Whitney during the Spring '08 term at Johns Hopkins.
- Spring '08