MMManuel Jose Marte

Manuel Jose Marte

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Manuel Jose Marte has a total of 50 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Subtle aphasia above the WAB-R cutoff one year after left-hemisphere stroke

    by Manuel Jose Marte, Mathew Chaves, Lindsey Kelly, Isidora Diaz-Carr, Voss Neal, Andreia V. Faria, Melissa D. Stockbridge, Argye E. Hillis

  • Open Access

    Assessing Spoken Discourse in Aphasia Using Multimodal Artificial Intelligence: Development and Validation Study (Preprint)

    by Manuel Jose Marte, Seung Hee Lee, Sharon Wang, Kari Goldin, Maria Varkanitsa, Jeffrey Girard, Rahul Brito, Xuehu Wei, Justin Baker, Yanmei Tie, et al.

    BACKGROUND Discourse analysis can reliably predict real-world communicative success, yet is rarely implemented clinically, due to challenges in manually generating stimulus-specific main concept inventories and scoring patient narratives. OBJECTIVE We evaluated (1) the quality of main concept inventories generated by a vision-language model relative to main concepts produced by healthy controls and (2) the value of automatically derived macrolinguistic and microlinguistic features extracted from the narrations for classification of persons with aphasia versus healthy controls and prediction of aphasia severity. METHODS Main concept inventories were generated for each movie clip with Gemini 2.5 Pro, a vision-language model, from the clip’s video frames, audio, and subtitles. Five sets of concepts were generated independently per clip and assembled into a single inventory. Features were extracted from narrative transcripts of all participant-clip combinations (51 healthy controls, 54 persons with aphasia, 8 clips), resulting in 688 narrations. Group comparisons used analysis of covariance with age as a covariate. Least absolute shrinkage and selection operator logistic regression with leave-two-out cross-validation classified the groups, and models restricted to the 15 microlinguistic or to the 4 macrolinguistic features were compared with the full model. Within the subsample of persons with aphasia, associations between each feature and the Western Aphasia Battery-Revised Aphasia Quotient (AQ) were estimated using partial correlations adjusting for age. RESULTS The main concept inventories generated directly from each clip’s video, audio, and subtitles using the vision-language model approximated the main concepts healthy controls produced. Persons with aphasia produced fewer unique main concepts than healthy controls (19% vs 46% of the inventory; d=−1.84, 95% CI −2.30 to −1.38). Group classification was excellent, with an area under the receiver operating characteristic curve (AUC) of 0.948 (95% CI 0.897-0.986) and an accuracy of 0.895 (95% CI 0.838-0.952). The AUC of neither restricted model differed significantly from the full model (microlinguistic only: AUC 0.914, P=.07; macrolinguistic only: AUC 0.908). More severe aphasia (lower AQ) was associated with greater distance of produced concepts from the main concept inventory (r=−0.79, 95% CI −0.87 to −0.66), lower main concept completeness (r=0.66), and more negative sentiment (r=0.65). In a hierarchical regression, distance to the centroid explained AQ variance beyond age, completeness, and the microlinguistic features associated with AQ (ΔR²=0.069, 95% CI 0.001-0.115; F1,41=16.62; P<.001). CONCLUSIONS Main concept inventories generated directly from multimodal movie input using a vision-language model approximated the main concepts healthy speakers produced, making them adequate to serve as reference inventories for computerized main concept analysis. The automatically derived macrolinguistic features improved prediction of aphasia severity beyond the traditional microlinguistic measures alone. The new methodology opens a path to libraries of discourse assessment stimuli that are more realistic, diverse, and adaptable to varying cultural and clinical needs.

  • Open Access

    Assessing spoken discourse in aphasia using multimodal artificial intelligence

    by M. J. Marte, S. Lee, S. Wang, K. Goldin, M. Varkanitsa, J.M. Girard, R. Brito, X. Wei, J.T. Baker, Y. Tie, et al.

    Discourse analysis can reliably predict real-world communicative success, yet is rarely implemented clinically, due to challenges in manually generating stimulus-specific main concept inventories (MCIs) and scoring patient narratives. We evaluated whether automatically derived macrolinguistic (main concept, sentiment) and microlinguistic (fluency, lexical, syntactic) features extracted from movie clip narrations can augment the clinical assessment of discourse in aphasia. The MCIs generated directly from each clip’s video, audio and subtitles using a vision-language model approximated the main concepts healthy controls produced. Group classification (control: n=51; aphasia: n=54) based on the discourse features was excellent (AUC=0.95, 95% CI [0.90, 0.99]), with macrolinguistic features showing the largest group differences (main concept completeness: d=-1.84, 95% CI [−2.30, −1.38]) and strongest associations with aphasia severity (semantic distance to inventory: r=-0.79, 95% CI [−0.87, −0.66]). Findings support the feasibility and scalability of automated criterion-based discourse assessment using ecological stimuli, and highlight its potential for clinical implementation.

  • Open Access

    Subtle language deficits in WAB-recovered patients at 12 months after left-hemisphere stroke

    by Manuel Jose Marte, Mathew Chaves, Lindsey Kelly, Isidora Diaz-Carr, Voss Neal, Andreia V. Faria, Melissa D. Stockbridge, Argye E. Hillis

    Background: The Western Aphasia Battery-Revised (WAB-R) Aphasia Quotient is the most widely used standardized post-stroke aphasia measure; its conventional 93.8 cutoff has limited sensitivity to mild residual impairment. Beyond the cutoff, it offers limited objective discourse assessment, no action-naming assessment, and naming tests limited to very common objects. Aims: We sought short tests capturing subtle aphasia in patients recovered above the WAB-R threshold, then its demographic and lesion correlates. Methods & Procedures: Sixty-seven patients with acute left-hemisphere ischemic stroke completed acute structural MRI and a 12-month language battery comprising the WAB-R, Boston Naming Test (BNT), Hopkins Action Naming Assessment (HANA), and Modern Cookie Theft (MCT) picture description. Hierarchical logistic (binary deficit) and linear (control-referenced composite z-score) regressions evaluated acute aphasia history, sex, education, age, acute depression (PHQ-9), and residualized regional lesion load. Outcomes & Results: Of 67 participants, 45 (67%) recovered above the WAB-R threshold. Of these, 18 (40%) had residual deficits on at least one supplemental test ("subtle aphasia"). BNT plus MCT content-unit count captured all 18 (100%); HANA added none beyond these two. The binary model discriminated deficit from no-deficit at AUC = 0.80 (95% CI [0.70, 1.00]); higher education significantly lowered deficit odds (OR = 0.80/year, 95% CI [0.64, 1.00], p = .049). On the continuous composite, acute PHQ-9 independently predicted 12-month outcome (β = −0.13 per point, 95% CI [−0.22, −0.04], p = .006, cumulative R-squared = 0.38). Applying the Senthilkumar et al. (2026) stricter cutoff (WAB-AQ ≥ 96.7) reclassified 12 of 45 (27%) out of recovery, capturing 8 of 18 (44%) subtle-aphasia patients. Composite residualized lesion load did not differentiate the groups when adjusted. Conclusions: Above the WAB-R recovery threshold, subtle aphasia is present on the BNT or MCT in ~40%, with higher education associated with lower odds at 12 months; acute depression emerged as a candidate correlate but did not survive removal of a single high-influence observation, warranting replication in larger samples. Regional lesion variables informative at greater stroke severity contribute little as large lesions cluster in the persistently aphasic group, reducing lesion variance within the recovered subgroup and its discrimination of subtle deficits. This adds to evidence that clinicians should not infer complete language recovery from the WAB-AQ alone, and that identifying residual deficits may require greater investment in behavioral assessment and consideration of alternative WAB-AQ cutoffs. Structural anatomical information, by contrast, appears to add little discriminative value at the upper performance range.

  • Open Access

    Evidence for impaired emotional reactivity in aphasia during naturalistic movie-viewing

    by Manuel Marte, Bryce Gillis, Xuehu Wei, Colin P Galvin, Laura Rigolo, Yanmei Tie, Swathi Kiran, Einat Liebenthal

    Language is thought to play a fundamental role in how humans experience emotion. Chronic aphasia, a persistent language impairment, provides a unique window for examining predictions about the role of language in shaping emotions. Here, we tested the hypothesis that aphasia is associated with altered patterns of emotional reactivity during naturalistic movie viewing. We analyzed real-time and holistic emotion ratings, and language comprehension, of movie clips, in persons with aphasia (PWA, n=57) and healthy controls (HC, n=43). PWA versus HC showed reduced typicality and temporal complexity of real-time emotion judgements (p<.006), reduced typicality of holistic emotion judgements (p<0.001), and reduced language task accuracy (p<.001). Principal component analysis yielded a primary component that distinguished the groups (p<.001), and in PWA correlated with aphasia (r=0.644) and depression (r=-0.305) scores. Machine learning achieved strong group classification (AUC=0.964, accuracy=86%) and moderate aphasia severity prediction (r=0.589), with both language and emotion contributing features. The findings converge in demonstrating that language and emotion impairments are coupled, during both lexical and discourse level processing. The work highlights that scientific accounts and clinical evaluations of language would benefit from incorporating real-time measures of emotion processing, particularly for improving our understanding of real-world functional communication and impairments.


  • Open Access

    Evidence for an Integrated Bilingual Language System from Discourse Tasks in Aphasia

    by Xuanyi Jessica Chen, Manuel J. Marte, Swathi Kiran, Esti Blanco-Elorrieta

    Background: The extent to which bilingual individuals represent and process their two languages within a shared or partially distinct neural architecture remains a topic of ongoing debate. While both parallel and divergent patterns of impairment have been reported in bilingual aphasia, such findings likely reflect a spectrum of representational overlap influenced by dominance, proficiency, and task demands. Critically, few studies have examined how breakdown manifests across multiple levels of linguistic structure using ecologically valid, discourse-based tasks. Aims: This study investigates whether Spanish-English bilinguals with aphasia exhibit parallel or dissociable patterns of impairment across their two languages, focusing on naturalistic narrative production and fine-grained analysis of error types and code-switching. Methods & Procedures: Thirteen bilingual individuals with aphasia following acquired brain injury produced story retellings in both languages. Speech samples were transcribed and coded for phonological, morphological, syntactic, and semantic errors, and for the word type at which they occurred. Code-switches were also identified and categorized. Analyses included generalized linear modeling and unsupervised clustering. Outcomes & Results: While participants made more errors in their non-dominant language, the structure and distribution of errors were highly similar across languages. Clustering algorithms revealed that impairments were parallel across languages and did not group by language dominance. Code-switching occurred more frequently from the non-dominant to the dominant language, consistent with activation-based lexical selection. Conclusions: Findings support an integrated bilingual language system that spans multiple levels of linguistic representation, modulated by language dominance. Naturalistic discourse tasks allow for richer characterization of bilingual language breakdown and may better inform both theoretical models and clinical management of bilingual aphasia.


  • Open Access

    Navigating the Complexity of Bilingual Aphasia: Current Insights and Future Directions

    by Marissa Russell-Meill, Manuel J. Marte, Erin Carpenter, Swathi Kiran

    Bilingual aphasia is shaped by a dynamic interplay of neural, cognitive, linguistic, and experiential factors that influence both impairment and recovery. This review synthesizes current evidence on bilingual language organization, assessment, and treatment, emphasizing how individual language histories and cognitive systems contribute to variability in outcomes. We highlight the challenges of estimating pre-stroke proficiency, evaluating impairment across languages, and interpreting recovery patterns. Finally, we explore emerging technological directions while emphasizing that advances in machine learning, automated assessment, and neurotechnology must be developed with explicit attention to cultural responsiveness and equity to ensure benefits reach diverse multilingual populations.


  • Open Access

    The evolution of word retrieval errors during semantic feature-based therapy in bilingual aphasia

    by Michael Scimeca, Claudia Peñaloza, Erin Ann Carpenter, Manuel Jose Marte, Marissa Russell-Meill, Swathi Kiran

    Bilinguals with aphasia routinely experience anomia in one or both of their languages that may be ameliorated by language treatment. Traditionally, treatment response has been captured by binary scoring systems that measure the presence or absence of improvement without examining how word retrieval attempts may change over time as a function of treatment. This study analyzed word retrieval errors and subsequent treatment outcomes for a group of 48 Spanish-English bilinguals with aphasia to determine if longitudinal error patterns could capture language recovery. Results revealed naming improvement for trained words in the treated language and translations of trained words in the untreated language. Specific types of word errors at baseline were associated with overall improvement in both languages; furthermore, patterns of responses changed over time as a function of lexical-semantic treatment. These results demonstrate that error analyses may characterize bilingual treatment outcomes and provide new evidence for mechanisms of impaired word retrieval.


  • Measurement of cross-language and cross-domain generalization following semantic feature-based anomia treatment in bilingual aphasia

    by Marissa Russell-Meill, Erin Carpenter, Manuel J. Marte, Michael Scimeca, Claudia Peñaloza, Swathi Kiran

    Semantic feature-based treatment (SFT), which engages the semantic network by repeatedly targeting retrieval of conceptual features to improve lexical-semantic access, has shown promise for facilitating generalization in aphasia rehabilitation. However, its capacity to drive broad improvement across cognitive-linguistic domains in bilinguals with aphasia (BWA) remains unclear. This study examined generalization effects (i.e., direct transfer, near transfer, and far transfer) following SFT in 48 Spanish-English BWA who took part in a randomized controlled trial. Participants received 40 h of SFT targeting word retrieval, with generalization assessed across three domains: naming of untrained items (direct transfer), semantic processing (near transfer), as well as global language ability and nonverbal abstract reasoning (far transfer). Results showed (i) robust improvements for trained and untrained naming targets, demonstrating direct transfer, (ii) near transfer effects for select semantic processing tasks, and (iii) far transfer limited to overall language function, with no gains in domain-general cognition. Notably, treatment benefits extended across languages, demonstrating cross-language generalization to multiple domains of language processing. Findings highlight SFT's capacity to drive comprehensive language recovery in BWA, revealing broad generalization effects across languages and linguistic domains. Such effects underscore the importance of systematically examining generalization patterns to optimize rehabilitation outcomes.Trial registration: ClinicalTrials.gov identifier: NCT02916524.