1

Li2 Language VariationLi2 Language Variation

NeurolinguisticsNeurolinguistics

TodayToday’’s topicss topicsSurvey of ways of seeing where in the brain a given linguistic activity is carried outEvidence for most language functions generally being in left brainWhere we believe phonology, syntax, semantics, lexical entries, different languages, etc. are processed

2

L R

Language in the brainLanguage in the brain

R side of body, R visual fieldAnalysisLanguage

eyescorpus callosum

L side of body, L visual fieldvision, spatial relationsmelody, facial recognition, gesture, joke, metaphor

Evidence for language in L brainEvidence for language in L brainFinding: Language normally controlled in L brain for righties

Lefties often bilateralSmall % of righties are bilateralOne member of mirror twin pairs is often right-lateralLateralized damage before critical age ⇒ relateralizationSchizophrenics significantly less lateralized (Sommer et al. 2001)Women less strongly lateralized than men

How do we know?R brain damage/removal doesn’t affect language; L brain damage doesLoss of language often accompanied with paralysis of R sideMore…

3

The Wada testThe Wada test

Used before brain surgery to see if it will leave patient aphasicInject sodium amytol or sodium methohexital to paralyze one or the other brain hemisphereAsk patient to count or name pictures on a screenRasmussen and Milner 1977: survey of Wada results for 262 individuals

96% L hemisphere, 4% RFor lefties, 70% L, 15% R, 15% in both hemispheres

Brain mappingBrain mappingPenfield and Roberts 1959, Ojemann 1983, Cairns 1999:60Done pre-surgery to determine which areas should not be cut, so as to avoid aphasic outcome

4

Dichotic listening testsDichotic listening tests

R ear wired to L brain; L ear wired to R brainInfo can then pass to other half through corpus callosum, but reaches opposite hemisphere first

Experimental paradigm:two different speech stimuli (typically CV) presented simultaneously, one in each earSubjects asked (i) which stimulus they hear better, (ii) what they heard in each ear, or (iii) identification response timed

PREDICTIONS: L hemisphere dominance for speech perception ⇒ preference for R ear stimulus (“Right Ear Advantage”)R hemisphere dominance for speech perception ⇒ preference for L ear stimulusNo hemispheric dominance for speech perception ⇒ no preference for one ear vs. the other

5

FINDINGS:Normals: accuracy of detecting target words was greater when stimuli were presented to the R ear, indicating L hemisphere specialization (Berman et al. 2003)Dyslexics: Many studies have shown that dyslexics have less R-ear asymmetryLesioned patients:

right lesioned patients and the control subjects had normal and expected right ear advantageleft lesioned patients performed almost at random with regard to the right and left ear stimulus (Hugdahl et al. 2003)

Dichotic listening testsDichotic listening tests

PET scanPET scan

Technique:Radioactive material (carbon, nitrogen, etc.) injected or inhaledWhen this material gets into the bloodstream, it goes to areas of the brain that use itBreakdown of radioactive material ⇒ release of gamma raysGamma ray detectors record the brain area where the gamma rays are emitted

Detection of words produces increased blood flow in a left frontal area associated with motor and phonetic processing, and a left temporal area associated with semantic memory (Berman et al. 2003)

PET scan PET + MRI

PET = positron emission tomography

6

TranscranialTranscranial Magnetic StimulationMagnetic StimulationPascual-Leone et al. 1991,Knecht et al. 2002

Normal subjects with range of languagelateralization from left- to right-hemisphere dominanceWhile the subjects performed a task that required language skills, the authors delivered transcranial magnetic stimulation (TMS) to one side of the brain or the otherSuch ‘virtual lesions’ disrupt the function of a targeted brain region for several minutes. In subjects with language dominance on one side, TMS over the dominant side slowed verbal processing, but TMS over the other side did notCf. effects of electrical stimulation on left brain when patients undergoing brain surgery are speaking

TMS and idiomsTMS and idiomsMassimiliano et al. 2004

Hypothesis: figurative language, including idioms, is controlled by the right hemisphere.Experiment:

repetitive transcranial magnetic stimulation (rTMS)40 opaque, nonambiguous idioms, 15 participants5 test blocks: {disrupt L/R x frontal/temporal} + baseline

In each block: 8 trials w/ idioms, 8 trials w/ literal sentencesIn each trial subject is presented with:

written sentence for 2000 msec, then:pair of pictures for 2500 msec

(i) corresponds to the sentence(ii) corresponds to literal meaning of idiom or (for

literal sentences) a sentence differing in a detailS has to press button corresponding to the picture matching the string.

7

TMS and idiomsTMS and idiomsMassimiliano et al. 2004

Results:Reaction times:

increased following left temporal rTMSunaffected by right hemisphere rTMSno difference between idiomatic and literal sentences in either condition

Conclusion:Opaque idiom and literal sentence comprehension depend on the left temporal cortex.

Conclusions thus farConclusions thus farThere is ample evidence that language is normally located in the left brain.

Note: not all parts of the left brain are used for language―PhineasGage:

8

PhineasPhineas GageGage

Locating linguistic activities Locating linguistic activities within the L brainwithin the L brain

Neural areas associated with:Phonology

Secondary Auditory Cortex, including Wernicke’sarea, esp. LH

OrthographyVisual word form area (inferior occipito-temporal), esp. LH

Words (lexicon)Extent of LH inferior temporal lobe

9

Locating phonological Locating phonological abilitiesabilities

ImagingImaging

10

Dyslexia Dyslexia ←← difficulties in difficulties in phonological processing?phonological processing?

General features of dyslexiaDyslexics have trouble reading despite normal-high intelligence and exposure to sufficient language instruction. Reversal of words and letters, difficulty in pronouncing new words, difficulty in making a distinction between similarities and differences in words (on for no), and difficulty in discerning differences in letter sounds (ten, tin).The distinctive pattern of spelling and memory problems that characterizes dyslexia has a strong genetic basis

Though all humans are equally likely to have dyslexia, the symptoms manifest significantly more often in speakers of languages with more opaque orthographies (English, French vs. Italian)

Numerous studies have shown that phonemic awareness (ability to count # of phonemes in a word, etc.) is a powerful predictor of reading success (Alegria, Pignot, and Morais 1982, etc.)Involves (in)ability to access the underlying sound structure of words (Liberman and Shankweiler 1991;Shaywitz 1996, 1998; Wagner and Torgesen 1987)

Georgiewa et al. 2002 study of silent readingDifficulties in phonological processing are currently considered one of the major causes for dyslexia.fMRI showed a significant difference in the activation in the left inferior frontal gyrus between the dyslexic and control groups, resulting from a hyperactivation in the dyslexics.ERP scalp distribution showed a significant distinction between the two groups concerning the topographic difference for left frontal electrodes in a time window 250–600 ms after stimulus onset for non-word reading.

Locating syntactic abilitiesLocating syntactic abilities

11

BrocaBroca’’ss aphasia (aphasia (agrammatismagrammatism))Results from damage to Broca’sarea, in front of the parietal lobe.

a.Me…build-ing…chairs, no, no cab-in-ets. One, saw…then, cutting wood…working…b.Cookie jar…fall over… chair…water… empty… ov…ov…[Examiner: ‘overflow’] Yeah.

halting speech, reading, and writingomission of function words and inflectionsdifficulty with grammaticalityjudgements

He showed her baby the picturesHe showed her the baby pictures

Mr. Tan’s brain

A brain region for Universal Grammar?A brain region for Universal Grammar?Musso et al. 2003Broca's area, a region of the brain involved in language, becomes increasingly active as subjects acquire rules from a foreign language, but not as they acquire comparable rules that are inconsistent with real languages.

German-speaking adults learned a small vocabulary in a foreign language (Italian or Japanese). Subjects were then taught rules either consistent or inconsistent with universal grammar.While they were in an fMRI scanner, subjects were asked to judge whether test sentences followed or violated these rules. A specific region of Broca's area became more active over time as participants became more adept with the real rules, and less active over time in response to rules that did not follow universal grammar.

12

Locating lexical knowledgeLocating lexical knowledge

Localization of the LexiconLocalization of the LexiconLesion foci associated with category-specific anomias

Cf. Shelton, J., Fouch, E., &Caramazza, A. The selective sparing of body part knowledge: A case study. Neurocase. 4, 339-350.

13

From a PET study of naming

Localization of the LexiconLocalization of the Lexicon

CategoryCategory--specific processingspecific processing

Pulvermüller, Friedemann and Olaf Hauk. 2006. Category-specific Conceptual Processing of Color and Form in Left Fronto-temporal Cortex. Cerebral Cortex 16.8:1193-1201.

14

WernickeWernicke’’ss aphasiaaphasiaResults from damage to Wernicke’s area, below the parietal lobe.

Examiner: What kind of work have you done?We, the kids, all of us, and I, we were working for a long time in the…you know…it's a kind of space, I mean place rear to the spedwan…Examiner: Excuse me, but I wanted to know what work you have been doing.If you had said that, we had said that,poomer, near the fortunate, forpunate,tamppoo, all around the fourth of Martz. Oh, I get all confused.

Syntax and intonation OK, but faulty semanticsProduce nonsense words (poomer,tarripoi, trebbin, repuceration)Poor comprehension

•Patient developed aphasia (especially problems producing word-initial consonants, I.e. a type of apraxiaof speech) following hemorrhage of basal ganglia

•(Examiner: What’s happening with her?)

•With her? She’s overflowing the sink . . . And . . . he’s /s/ . . . (Examiner: How ‘bout over there?) ‘‘Over there? He /kæraher/ it in /e/ /ki/ it’s a . . . (expletive) . . . /fo fo folling/ it’s cookie jar . . . he’s about to fall off . . . and hurt hisself.

the “cookie theft picture”

15

ConclusionsConclusionsMost language functions are (normally) located in the left brain.Language itself is modular, with different subcomponents of the grammar and lexicon being located in different parts of the left brain.

ReferencesReferencesBaynes, K. et al. 1998. Interhemispheric

integration masks modular organization of cognitive systems. Science 280:902-905.

Berman et al. 2003. Complementary hemispheric specialization for word and accent detection.Neuroimage 19.2.1:319-31.

Hugdahl et al. 2003. Dichotic listening performance and frontal lobe function. Brain Research: Cognitive Brain Research 16.1:58-65.

Motoichiro Kato, Masaru Mimura, Fumihiro Yoshino, Taro Muramatu, Haruo Kashima and Kenji Ishii. 1998. Discrete Left Temporal Regions