The neural correlates of verb and noun processing. A PET study. - - PowerPoint PPT Presentation

The neural correlates of verb and noun processing. A PET study. Perani et al. Brain 1999 Dec;122. The hypothesis that categorical information, distinguishing among grammatical word classes, such as nouns and verbs in the lexical knowledge in the brain, is supported by the observation of aphasic subjects who are selectively impaired in the processing of nouns and verbs. The study of lesion locations has suggested that the left temporal lobe plays a crucial role in processing nouns, while the left frontal lobe is necessary for verbs. We measure regional cerebral activity with PET during tasks requiring reading of concrete and abstract nouns and verbs for lexical decision.

Activation of an extensive network of brain areas, mostly in the left frontal and temporal cortex, which represents the neural correlate of single word processing (lexical-semantic processing of words). Some left hemispheric areas, including the dorsolateral frontal and lateral temporal cortex, were activated only by verbs, while there were no brain areas more active in response to nouns. Additional brain activations for verbs appear to be related to specific semantic content, or associated with the automatic access of syntactic information.

Neuroanatomical models

3 main views that bear some correspondence to the cognitive architectures:

• 1. On the basis of neuropsychological data, nouns and verbs are represented in

at least partially separable neural networks, with noun processing engaging left temporal areas and verb processing engaging left inferior frontal areas

• 2. Morpho-syntactic processes that apply to the different grammatical classes

(verbs and nouns) would be computed by neural networks in close physical proximity to the networks engaged by the semantic content of words

• 3. No neural separability is assumed for words of different grammatical classes;

rather neural separability is assumed for words referring to actions vs. words referring to objects (regardless of their grammatical class). The same shared neural network would be engaged in integration processes for both nouns and verbs.

Vigliocco et al. Neuroscience and Biobehavioral Reviews 35, 2011

Vigliocco et al. Neuroscience and Biobehavioral Reviews 35, 2011

NEUROPSYCHOLOGY

Evidenze neuropsicologiche: la doppia dissociazione tra nomi e verbi in pazienti con lesioni focali

10 20 30 40 50 60 70 80 90 PR AL Nouns Verbs

Cappa, 1998

Tranel et al., 2001 Verbs and nouns in lesion studies

Neuropsychology Verb processing is more demanding than noun processing In the review we argued for greater difficulty in processing verbs than nouns

Neuropsychology

Arevalo et al. 2007

The processing of words and pictures: picture-naming (PN), single word reading (WR) and word repetition (WRP). Analysis 1 targeted task and lexical category (noun-verb), revealing worse performance on Picture Naming of verb items for patients. Analysis 2 (hand manipulation and actions), patients displayed relative difficulty with the ‘manipulable’ items These results carry implications for theories

• f

embodiment, lexico-semantic dissociations, and the

• rganization of meaning in the brain.

Conclusions

Object and action naming in early PD

10 20 30 40 50 60 70 80 90 100 PD controls

• bjects

actions

Cotelli et al. 2006

According to embodiment, the recruitment of the motor system is necessary to process language material expressing a motor content. Coherently, an impairment

• f

the motor system should affect the capacity to process language items with a motor content Deficts in processing graspable objects and their nouns.

• 1. These findings provide evidence in favour of

embodiment, by supporting a causal link between lesions affecting the motor system and the capacity to process language with a motor content.

• 2. All suggests that language in the brain is not segregated in

terms of grammar items (verbs, nouns) but rather in terms of semantic fields, as embodiment would suggest (Buccino et al., 2016; Vigliocco, Vinson, Druks, Barber, & Cappa, 2011).

Object vs. action naming in ND dementia

5 10 15 20 25 30 35 40 AD FTD controls Objects Actions

Cappa et al., 1998

FTD spectrum

• Fronto-temporal degeneration
• Frontal variant
• Progressive nonfluent aphasia
• Semantic dementia
• Progressive supranuclear palsy
• Corticobasal degeneration

10 20 30 40 50 60 70 80 90 100

controls CBD PSP NFPA Frontal SD AD

Noun Verb

Action-object naming

Cotelli et al. 2006

Manipulation effect in object/action items

10 20 30 40 50 60 70 80 CBD PSP NFPA Fv FTD SD AD Manip Non-manip

5 10 15 20 25 30 35 CBD PSP Non manip Manip

Cotelli et al. 2006

Different mechanisms of impairment, reflecting the anatomical selectivity of the pathological ND process, may be responsible for defective verb and noun processing (i.e. NfPPA agrammatism, morphosyntactic disorder) As for focal lesion studies, a primary factor responsible for dissociations is semantics: defective action knowledge appears to be a crucial factor in particular in the case of movement disorders

Vigliocco et al. Neuroscience and Biobehavioral Reviews 35, 2011

Perani personal data

Nf PPA FDG PET hypometabolism

Perani D personal data

NEUROPHYSIOLOGY

Barber et al., 2009 Comparable topographical distribution for the grammatical class (nouns vs. verbs) and semantic manipulations (sensory vs. motor) in processing of single words Neurophysiology

Neurophysiology

Picture Naming

VERB GENERATION

Serial progression of activations is seen at a larger timescale, showing distinct stages of perception, semantic association/selection, and speech production.

a “dual ventral stream” model

MRI, fMRI

These findings suggest that processing seen graspable

• bjects and written nouns referring to graspable
• bjects similarly modulates the motor system.

Behaviour

These data provide the first direct evidence that listening to sentences that describe actions engages the visuomotor circuits which subserve action execution and observation

Subjects: 16 right-handed healthy participant fMRI: 1.5 T whole body scanner (GE, Med Sys) Experimental conditions 120 sentences (40 hand, foot, mouth) I am biting an apple, I am grasping the hammer, I am kicking the balloon) 120 baseline “abstract”sentences I appreciate sincerity, I understand the truth

Auditory processing of sentences

mouth hand foot mouth

. . . . . . . . . . . . . . .

ADDENTO UNA MELA LECCO IL GELATO

. . .

MASTICO IL PANE SOFFIO SUL FUOCO

. . .

AFFERRO UN COLTELLO SPEZZO IL PANE

. . .

CALCIO IL PALLONE INFILO GLI ZOCCOLI RICORDO IL FATTO VALUTO GLI ERRORI TEMO LA PROVA IMMAGINO LA COLLERA CONFONDO I RICORDI DESIDERO LA GIOIA ADORO L’ OZIO APPREZZO LE DOTI

t 20 s 4 s

mouth hand foot mouth

. . . . . . . . . . . . . . .

ADDENTO UNA MELA LECCO IL GELATO

. . .

MASTICO IL PANE SOFFIO SUL FUOCO

. . .

AFFERRO UN COLTELLO SPEZZO IL PANE

. . .

CALCIO IL PALLONE INFILO GLI ZOCCOLI RICORDO IL FATTO VALUTO GLI ERRORI TEMO LA PROVA IMMAGINO LA COLLERA CONFONDO I RICORDI DESIDERO LA GIOIA ADORO L’ OZIO APPREZZO LE DOTI

t 20 s 4 s

mouth hand baseline foot mouth

. . . . . . . . . . . . . . .

ADDENTO UNA MELA LECCO IL GELATO

. . .

MASTICO IL PANE SOFFIO SUL FUOCO

. . .

AFFERRO UN COLTELLO SPEZZO IL PANE

. . .

CALCIO IL PALLONE INFILO GLI ZOCCOLI RICORDO IL FATTO VALUTO GLI ERRORI TEMO LA PROVA IMMAGINO LA COLLERA CONFONDO I RICORDI DESIDERO LA GIOIA ADORO L’ OZIO APPREZZO LE DOTI

t 20 s 4 s

Beyond the basic difference that verbs typically refer to actions and nouns typically refer to

• bjects,
• ther

semantic distinctions might play a role as organizing principles within and across word classes. One possible candidate is the notion of manipulation and manipulability, which may modulate the word class dissociation.

METHODS

• Task: overt naming
• Stimuli: 60 drawings from the UCSD corpus

(matched for frequency, visual complexity, age of acquisition)

Objects-NOUNS Actions-VERBS

manipulable non-manipulable non-manipulation manipulation

Stimuli

Objects-NOUNS Actions-VERBS

manipulable non-manipulable non-manipulation manipulation mixer ship brush to tie fence to jump to skate to polish

Stimuli characteristics

Age of acquisition

1 2 3 4 5 VnM VM NnM NM

Stimulus type years

VnM VM NnM NM

Response Time

800 900 1000 1100 1200 1300 1400 1500

VnM VM NnM NM

Stimulus type msecs

VnM VM NnM NM

Frequency

1 2 3 4 VnM VM NnM NM

Stimulus type

log nat freq

VnM VM NnM NM

p< 0.001

n.s. n.s

R L p > .001

10 10

z

Hand >noHand noHand >Hand

z

Manipulation effect

we were looking for common activation of nouns and objects, and actions and verbs, representing the canonical and mirror neuron system,respectively “Kugel”, “Ring”, Zylinder” “fangen”, “nehmen”, “rollen” the model weights underlying action-versus-object distinction

fMRI

Nouns vs. Verbs Verbs vs. Nouns

fMRI

Noun and verb production recruit the same neural networks for L1 and L2, independently of AOA. These results support the neural convergence hypothesis

Source: Vigliocco, Vinson, Lewis & Garrett (2004) Source: Vigliocco, Vinson, Lewis & Garrett (2004)

Action Properties

Source: Vinson, Vigliocco, Cappa & Siri (2003) Source: Vinson, Vigliocco, Cappa & Siri (2003)

Object Properties

Vigliocco et al. Neuroscience and Biobehavioral Reviews 35, 2011

SIGNIFICATO PAROLE D’AZIONE SIGNIFICATO PAROLE PERCETTIVE

INTEGRAZIONE

Differenze tra nomi e verbi

• Referenza semantica
• Ruolo sintattico
• Morfologia
• Struttura fonologica
• Frequenza
• Età di acquisizione
• Concretezza

….

Siri S, Tettamanti M, Cappa SF, Della Rosa P, Saccuman C, Scifo P, Vigliocco G. The neural substrate of naming events: effects of processing demands but not of grammatical class. Cereb Cortex. 2008 In order to tease apart the role of semantic and grammatical factors: 1) Infinitive Verb (e.g., mangiare [to eat]); 2) Inflected Verb (e.g., mangia [she/he eats]); 3) Action Noun (e.g., mangiata [the eating])

• We did not find any verb-specific activation
• Left IFG activations were found when contrasting the Action

Noun with the Infinitive Verb

• Activation in left IFG was greatest for Action Nouns,

intermediate for Inflected Verbs, and least for Infinitive Verbs.

We conclude that, when all

• ther

factors are controlled, nouns and verbs are processed by a common neural system. In the present case, differences in left IFG activation emerge as a consequence of increasing linguistic and/or general processing demands.

Hierarchical functional connectivity between the core language system and the working memory system. Makuuchi M1, Friederici AD. Cortex. 2013 Oct;49(9) An anterior-posterior gradient of cognitive control within the dorsomedial striatum. Mestres-Missé A, Turner R, Friederici AD.

• Neuroimage. 2012 Aug 1;62(1):41-7.

Revisiting the role of Broca's area in sentence processing: syntactic integration versus syntactic working memory.Fiebach CJ, Schlesewsky M, Lohmann G, von Cramon DY, Friederici AD. Hum Brain Mapp. 2005 Feb;24(2):79-91

Once we take into account the confounding in most studies between semantic distinctions (objects vs. actions) and grammatical distinction (nouns vs. verbs), and the conflation between mechanisms of single word processing and sentence integration, the emerging picture is relatively clear-cut:

• 1. clear neural separability is observed between the processing of object

words (nouns) and action words (typically verbs),

• 2. grammatical class effects emerge or become stronger for tasks and

languages imposing greater processing demands. These findings indicate that grammatical class per se is not an

• rganisational principle of knowledge in the brain; rather, all the findings

are compatible with two general principles described by typological linguistics as underlying grammatical class membership across languages: semantic/pragmatic, and distributional cues in language that distinguish nouns from verbs.

CONCLUSIONS

Vigliocco et al. Neuroscience and Biobehavioral Reviews 35, 2011

THANK YOU FOR YOUR ATTENTION