In search of the language switch: an fMRI study of picture naming in spanish-english bilinguals hernandez, martinez & kohnert (2000)Functional Magnetic Resonance Imaging (fMRI)Slide 3fMRI techniquefMRI technique (cont’d)Slide 6Slide 7Slide 8Slide 9Hernandez et al. (2000): Motivation for the studyBackgroundKim, Pelkin, Lee & Hirsch (1997)Kim, Pelkin, Lee & Hirsch (1997) (cont’d)Background (cont’d)Research Questions and PredictionsMethods: ParticipantsMethods: Design and MaterialsMethods: ProcedureMethods: Procedure (cont’d)Methods: fMRI data collectionMethods: fMRI data analysisResults: Behavioral experimentResults: fMRI experimentResults: fMRI experiment (cont’d)Results (cont’d)DiscussionDiscussion (cont’d)Slide 28Slide 29IN SEARCH OF THE LANGUAGE SWITCH: AN fMRI STUDY OF PICTURE NAMING IN SPANISH-ENGLISH BILINGUALSHERNANDEZ, MARTINEZ & KOHNERT (2000)ISRAEL DE LA FUENTEMARCH 2, 2010Functional Magnetic Resonance Imaging (fMRI)Functional Magnetic Resonance Imaging (fMRI)fMRI techniqueIncrease in metabolic rateDepletion of hemoglobin (iron) in a tissue sampleTissue over-supplied with oxygenated blood (after 5-8 sec)fMRI technique (cont’d)Relative concentration of hemoglobin constitutes the object to be depicted in fMR imagesHemoglobin is not associated to electromagnetic signals that could reveal its presenceIt does affect the behavior of hydrogen atoms that emit recordable electromagnetic signalsElectromagnetic signals from hydrogen atomsElectromagnetic signals from hydrogen atomsDistribution estimation & image representation of hemoglobinDistribution estimation & image representation of hemoglobinAffected by concentration of hemoglobinAffected by concentration of hemoglobinfMRI technique (cont’d)Hydrogen protons spin around axes of different orientationWith the help of a magnetic field, the axes of spin of protons tend to align with the magnetic fieldRadio Frequency (RF) pulse is additionally usedThe RF pulse brings all protons in phase and results in maximal transverse magnetizationThis component results in a resonant magnetic signal captured by the same RF coilfMRI technique (cont’d)Intensity of the recorded signal varies depending on how much hemoglobin is present in the tissue sampleFrequencies of the resonant signal will be different depending on the pre-arranged values of the magnetic field (voxel)Relative intensity of the signal > information about the degree of activation of a given spot in the brainFrequency of the signal > location of the activated spotsfMRI technique (cont’d)Superimposition (or co-registration) of the functional and structural images completes the process of developing the fMR imageTo visualize this activation, we must partition the 3D set into 2D slices (“back-projection”)Depending on the threshold chosen, the area of activation imaged may very in sizefMRI technique (cont’d)Pattern of activation seen in the image may surge in the metabolic rate of neural tissue or may also reflect activity of neighboring blood vesselsTemporal resolution of fMRI is in the order of seconds (due to the nature of the reality imaged)cf. Marianna’s question #1Fidelity of fMR imagesSpatial resolution of fMRI is superior to that of MEG (no “inverse problem”)The actual degree of activation (spatial extent and intensity) is unclearHernandez et al. (2000):Motivation for the studyHow are a bilingual’s two languages represented in the brain?What brain areas are active when bilinguals switch from one language to the other?BackgroundDifferent recovery patterns in bilingual and polyglot aphasicsSelective ImpairmentEach language is functionally different and a lesion can functionally dissociate themEach language is represented in different brain areascf. Pawel’s question #1Recent studies using neuroimaging techniques find that:Languages are represented in non-overlapping areas in late bilingualsLanguages are represented in overlapping areas in early bilingualscf. Lucy’s question #1Kim, Pelkin, Lee & Hirsch (1997)Kim, Pelkin, Lee & Hirsch (1997) (cont’d)Background (cont’d)If the two language systems of bilingual speakers utilize the same or adjacent pieces of neural tissueHow is it that bilinguals keep information from one language constantly interfering with processing of the other language?“Language Switch” at the neurophysiological level (Penfields & Roberts, 1959)There appears to be no specific area of the brain that is dedicated exclusively to language switchingThere is evidence that switching between two tasks can result in activation in the dorsolateral prefrontal cortex and in the supramarginal gyrusResearch Questions and PredictionsIn a picture-naming task, will different brain areas be activated when processing each language?There will be very small differences in the activation pattern associated with naming pictures in each languageHow will the switching of languages affect this activation pattern? And will this “switch” be associated with a particular brain area?The conditions of language switching will involve additional activation both in terms of intensity and extent of activationIt remains to be seen if this increase is in the dorsolateral prefrontal cortex, the supramarginal gyrus, or bothMethods: ParticipantsSix participants (4 female and 2 male)English-Spanish bilinguals (AoA before 5)English dominant speakersYears of formal study: 16 in English, 3 in SpanishBoston Naming Test: English 54, Spanish 40All of them strongly right-handed; no left-handed members in their immediate familyMethods: Design and MaterialsSet of 180 pictures from Snodgrass & Vanderwert (1980) and the Pictures Please Catalog (Abbate, 1984)Two versions of the experiment:Behavioral (administered first)Experimental (fMRI part; administered second)Methods: ProcedureBehavioral version:Participant presses the space keyAuditory cue (“say” or “diga”) and visual picture presented simultaneouslyPictures disappear after 3000 ms or after responseExperimental (fMRI) version:Stimuli presentation controlled by computerParticipants were shown the cue visually for 200 msFollowed by the picture for 400 ms1200 ms delay prior to the next stimulus presentationMethods: Procedure (cont’d)Design was run in blocks of 40 sec of experimental task (Spanish, English or