Lisa Pyun, 2009 How do bilinguals represent and distinguish two languages in their brain when the languages have overlapping words? Brain-imaging findings, using primarily fMRI and PET, reveal that languages of bilinguals overlap considerably in their neuroanatomical distribution. However, the manner in which bilinguals distinguish between two languages is unclear. Neuroimaging using fMRI and PET relies on hemodynamic differences, which are insensitive to the rapid time course of neurophysiological activity. This may be crucial in detecting subtle differences in neural activity, such as that between two languages of a bilingual. Hence a novel approach, using MEG, was used to examine the organization of language in bilinguals. In this study, (i) cognates (e.g. director (English) and director (Spanish)) were compared to non-cognates (e.g. hat (English) and sombrero (Spanish)) and (ii) within cognates, word forms of varying orthography and phonology were examined.

To our knowledge, no other study has investigated the role of cognates in bilingual language access using MEG. Fourteen fluent Spanish-English bilinguals listened to cognates and non-cognates and determined if words were Spanish or English, as neural activity was measured with MEG. Behavioral (RTs) and neurophysiological (M350) responses were analyzed. Cognates with phonological overlap had faster RTs than the non-cognates. At the neurophysiological level, there were no significant distinctions in M350 peaks among the different word categories. Results indicate that early bilinguals neurophysiologically process words in their two languages with similar temporal patterns and supplement prior findings of neuroanatomical overlap. These results provide further insight into the organization of the bilingual brain. Click here to view poster

Monica Sampson, 2009
Investigation of self-monitoring in fluent aphasia with jargon
This study investigated the extent to which fluent aphasic individuals with jargon (FWJ) monitor their speech, and the possible locus of breakdown resulting in their self monitoring deficits. Self-monitoring was tested in three production tasks (picture naming, non-word repetition and word repetition) under two listening conditions (normal listening and masking noise). Auditory processing was assessed in speech discrimination, lexical decision and single word identification tasks. Results from five participants with fluent aphasia and jargon indicated that severity of self-monitoring deficits is: 1) strongly correlated with severity of jargon production, 2) independent of auditory processing abilities, 3) is worse under masking noise, and 4) is worse for naming compared to repetition. The possibility of a post-semantic deficit in accessing phonological codes resulting in self-monitoring deficits in fluent aphasic individuals with jargon production is discussed.
Read more here

Tomato to Baritood : The source of non-words in jargon aphasia
Some aphasic individuals produce excessive non-words, resulting in speech that sounds like jargon. The source of jargon is attributed to impairments in psycholinguistic processes such as lexical-semantic retrieval, phonological encoding, and self-monitoring (Moses, Nickels & Sheard, 2004). Few studies have systematically examined self-generated speech to identify a locus of deficit resulting in jargon production. In this study, the nature of jargon errors and their relationship to target was compared across three tasks employing lexical processes to different extents: word repetition, non-word repetition and picture naming. They all entail phonological encoding, but differ in whether they place heavy (picture naming), light (word repetition) or no (non-word repetition) demands on semantic processes. Additionally, stimulus length (monosyllabic/multisyllabic) places different demands on phonological encoding. The results indicated that the primary breakdown occurs during retrieval of phonological information following lexical-semantic access. Hence phonological encoding proceeds with underspecified information about the target word form leading to jargon. When the surface form of the target word activated phonological planning processes directly (as in word repetition), these individuals produce less jargon and closer approximation to target even among produced errors. However, some individuals may experience additional breakdown during phonological encoding (as in PH) which can compound the severity of jargon production. Click here to view poster.

Lauren Graham, 2009
Verb retrieval difficulties are common in aphasia. Treatments for verb retrieval deficits are fewer and have met with less success than noun retrieval treatments (e.g., Conroy et al., 2006). The present study investigated the efficacy of a novel verb retrieval treatment that focused on training semantic features unique to two specific verb classes, cut and contact verbs, in two aphasic individuals. Acquisition of trained verbs and generalization to untrained verbs within and across verb classes was investigated. As per theories of action representation, cut verbs possess a larger set of semantic features while contact verbs possess only a subset of these semantic features. It was predicted that semantically-based training of cut verbs would generalize to retrieval of contact verbs. The findings were mixed: One participant (P1) improved in trained verbs, while the other (P2) did not. Neither participant displayed within or across-class generalization to untrained verbs. However, both participants’ improved significantly in action naming scores on the Object and Action Naming Battery and in the Aphasia Quotient of the Western Aphasia Battery. The findings illustrate that while specific verbs can be trained, spread of treatment effects to untrained verbs is elusive.
Click here to view findings

Therese Kling, 2007
Encoding for language production is assumed to consist of at least three post-conceptual, pre-articulatory processes: lexical (word retrieval), syntactic (sentence structure), and phonological (word form). These processes are impaired to varying extents in individuals with post-stroke aphasia. The neuroanatomical lesions associated with each of these impairments are incompletely understood. In this study, voxel-wise lesion symptom mapping (VLSM) was used to investigate if specific cortical regions can be identified for deficits in lexical, syntactic, and phonological production in 31 post-stroke aphasic individuals with left hemisphere lesions. Accuracy scores on object naming, sentence production, and nonword repetition were used as lexical, syntactic, and phonological measures respectively and correlated with T1-weighted structural MRI scans. VLSM revealed mostly unique, but a few shared lesions of five crucial perisylvian regions that patterned differently for the three measures. Overall, syntactic and phonological deficits were associated with left anterior perisylvian lesions, including the pars opercularis and triangularis of the left inferior frontal lobe, anterior superior temporal gyrus, anterior portions of the supramarginal gyrus, the putamen, and anterior portions of the insula. In contrast, lexical deficits were associated with posterior perisylvian lesions including major portions of the inferior parietal lobe and middle temporal gyrus. Interestingly, the distribution of lesions in the insula was consistent with this antero-posterior perisylvian gradient. The posterior plannum temporale was associated with a combination of all three deficits. The neuroanatomical distribution of lesions is largely consistent with regions identified in functional neuroimaging studies of language production and processing. These findings emphasize the participation of each perisylvian region in multiple linguistic functions, suggesting a many(functions)-to-many(networks) framework while also identifying functional subdivisions within each region. Our findings attest recent neuroantomical models of language (processing) and extend these to encoding for language production.
Click here to see the presentation

Christine Virion, 2008
Aphasia therapy that involves a high weekly intensity, short overall duration, restriction of nonverbal communication, coupled with constraints on verbal complexity, has recently gained momentum (constraint-induced language therapy, or CILT). The gains have been documented primarily for formal language tests, especially in lexical retrieval, repetition, and comprehension measures. Measures of grammatical well-formedness, which have greater ecological validity, have not been commonly reported in prior studies. Further, it is as yet unclear if the nature of verbal constraints has any impact on expressive language outcomes, particularly when the primary deficit in verbal production is grammatical inaccuracy (as in the case of agrammatic aphasia). This study aimed to examine whether constraint-induced therapy is applicable for individuals with agrammatic aphasia and if the addition of a morphosyntactic constraint would influence expressive language outcomes. In this phase I study a single participant design was used with four chronic agrammatic aphasic individuals who received 24 hours of constraint induced therapy over 10 days, as per prior published protocols. Two of these individuals received additional morphosyntactic constraints regarding tense morphology. Formal aphasia tests, Cinderella story narration, and conversational samples were analysed at three time points: pre-treatment, post-treatment, and 3-month follow-up. The study found that while all participants improved on at least some language measures, the overall changes were minimal and not maintained at 3 months. Participants who received morphosyntactic constraints dramatically improved on an elicited morphosyntactic test, but did not respond differently in other severity and discourse measures. Participants with lower initial language severity scores showed quantitatively larger gains after treatment. We conclude that while constraint-induced therapy was minimally effective for the agrammatic participants in this study, and addition of a grammaticality constraint did not significantly enhance the functional outcomes, the findings do indicate that initial severity and aphasic deficit patterns may be useful in determining candidacy for constraint-induced therapy.
Click here for poster
Click here for article