Language Comprehension Research Articles

Neuropsychological tests associated with symptomatic HIV-associated neurocognitive disorder (HAND) in a cohort of older adults in Tanzania.

Human immunodeficiency virus (HIV)-associated neurocognitive disorder (HAND) prevalence is expected to increase in East Africa as treatment coverage increases, survival improves, and this population ages. This study aimed to better understand the current cognitive phenotype of this newly emergent population of older combination antiretroviral therapy (cART)-treated people living with HIV (PLWH), in which current screening measures lack accuracy. This will facilitate the refinement of HAND cognitive screening tools for this setting. This is a secondary analysis of 253 PLWH aged ≥50 years receiving standard government HIV clinic follow-up in Kilimanjaro, Tanzania. They were evaluated with a detailed locally normed low-literacy neuropsychological battery annually on three occasions and a consensus panel diagnosis of HAND by Frascati criteria based on clinical evaluation and collateral history. Tests of verbal learning and memory, categorical verbal fluency, visual memory, and visuoconstruction had an area under the receiver operating characteristic curve >0.7 for symptomatic HAND (s-HAND) (0.70-0.72; p < 0.001 for all tests). Tests of visual memory, verbal learning with delayed recall and recognition memory, psychomotor speed, language comprehension, and categorical verbal fluency were independently associated with s-HAND in a logistic mixed effects model (p < 0.01 for all). Neuropsychological impairments varied by educational background. A broad range of cognitive domains are affected in older, well-controlled, East African PLWH, including those not captured in widely used screening measures. It is possible that educational background affects the observed cognitive impairments in this setting. Future screening measures for similar populations should consider assessment of visual memory, verbal learning, language comprehension, and executive and motor function.

Childhood Listening and Associated Cognitive Difficulties Persist Into Adolescence.

Listening difficulty (LiD) refers to the challenges individuals face when trying to hear and comprehend speech and other sounds. LiD can arise from various sources, such as hearing sensitivity, language comprehension, cognitive function, or auditory processing. Although some children with LiD have hearing loss, many have clinically normal audiometric thresholds. To determine the impact of hearing and cognitive factors on LiD in children with a clinically normal audiogram, we conducted a longitudinal study. The Evaluation of Children's Listening & Processing Skills (ECLiPS), a validated and standardized caregiver evaluation tool, was used to group participants as either LiD or typically developing (TD). Our previous study aimed to characterize LiD in 6- to 13-year-old children during the project's baseline, cross-sectional phase. We found that children with LiD needed a higher signal-to-noise ratio during speech-in-speech tests and scored lower on all assessed components of the NIH Cognition Toolbox than TD children. The primary goal of this study was to examine if the differences between LiD and TD groups are temporary or enduring throughout childhood. This longitudinal study had three data collection waves for children with LiD and TD aged 6 to 13 years at Wave 1, followed by assessments at 2-year (Wave 2) and 4-year (Wave 3) intervals. Primary analysis focused on data from Waves 1 and 2. Secondary analysis encompassed all three waves despite high attrition at Wave 3. Caregivers completed the ECLiPS, while participants completed the Listening in Spatialized Noise-Sentences (LiSN-S) test and the NIH-Toolbox Cognition Battery during each wave. The analysis consisted of (1) examining longitudinal differences between TD and LiD groups in demographics, listening, auditory, and cognitive function; (2) identifying functional domains contributing to LiD; and (3) test-retest reliability of measures across waves. Mixed-effect models were employed to analyze longitudinal data. The study enrolled 169 participants, with 147, 100, and 31 children completing the required testing during Waves 1, 2, and 3, respectively. The mean ages at these waves were 9.5, 12.0, and 14.0 years. On average, children with LiD consistently underperformed TD children in auditory and cognitive tasks across all waves. Maternal education, auditory, and cognitive abilities independently predicted caregiver-reported listening skills. Significant correlations between Waves 1 and 2 confirmed high, long-term reliability. Secondary analysis of Wave 3 was consistent with the primary analyses of Waves 1 and 2, reinforcing the enduring nature of listening difficulties. Children with LiD and clinically normal audiograms experience persistent auditory, listening, and cognitive challenges through at least adolescence. The degree of LiD can be independently predicted by maternal education, cognitive processing, and spatial listening skills. This study underscores the importance of early detection and intervention for childhood LiD and highlights the role of socioeconomic factors as contributors to these challenges.

What do you learn from a single cue? Dimensional reweighting and cue reassociation from experience with a newly unreliable phonetic cue

In language comprehension, we use perceptual cues to infer meanings. Some of these cues reside on perceptual dimensions. For example, the difference between bear and pear is cued by a difference in voice onset time (VOT), which is a continuous perceptual dimension. The present paper asks whether, and when, experience with a single value on a dimension behaving unexpectedly is used by the learner to reweight the whole dimension. We show that learners reweight the whole VOT dimension when exposed to a single VOT value (e.g., 45 ms) and provided with feedback indicating that the speaker intended to produce a /b/ 50% of the time and a /p/ the other 50% of the time. Importantly, dimensional reweighting occurs only if 1) the 50/50 feedback is unexpected for the VOT value, and 2) there is another dimension that is predictive of feedback. When no predictive dimension is available, listeners reassociate the experienced VOT value with the more surprising outcome but do not downweight the entire VOT dimension. These results provide support for perceptual representations of speech sounds that combine cues and dimensions, for viewing perceptual learning in speech as a combination of error-driven cue reassociation and dimensional reweighting, and for considering dimensional reweighting to be reallocation of attention that occurs only when there is evidence that reallocating attention would improve prediction accuracy (Harmon, Z., Idemaru, K., & Kapatsinski, V. 2019. Learning mechanisms in cue reweighting. Cognition, 189, 76–88.).

Students’ reading comprehension: an educational case about a methodological proposal in education

Introduction: The intention of this article is to show the experience of an eclectic proposal implementation in first grade students. Objective: To deeply know how in elementary school students are beginning to work on an ability that will become very important during their learning process, which will be the reading skill in a second language. Method: As they are developing this skill through the use of reproductive skills, in this study it is proposed to use an alternative methodology to this language development, which will be an eclectic methodological proposal with a qualitative approach, in order to take advantage of this skill acquisition process to begin growing reading comprehension and second language learning as well. Results: It could be demonstrated that an eclectic methodological proposal could help students to improve their reading comprehension. Conclusion: Teachers can feel confident about using eclectic methodologies in the classroom as it can become a success along the learning sessions.

Syntactic constructions drive cortical tracking in the absence of lexical content: an electrophysiological investigation of sentence processing during reading

ABSTRACT Previous work has established that neural oscillations can temporally synchronise with the occurrence of multi-unit linguistic representations but that neural synchrony to phrase boundaries is reduced in the absence of lexical content . In this study, we extend these findings to test whether reducing parsing ambiguity eliminates the reduction in phrase tracking associated with a lack of lexical content. Building on previous work, we reduced word-level parsing ambiguity by visually presenting sentences, and we reduced phrase-level parsing ambiguities by using the caused-motion construction for all sentences. Results indicated no difference in word or phrase-level tracking when comparing English and jabberwocky stimuli. We interpret this finding as indicating that tracking phrase structure can take place in the absence of content words if parsing ambiguities are sufficiently reduced.

Comprehension of complex sentences containing temporal connectives: How children are led down the event-semantic kindergarten-path.

Children up to school age are known to have difficulty comprehending complex sentences with temporal connectives, but the reasons remain controversial. We tested six- to twelve-year-old children to assess how the iconicity of event-language mapping, type of connective, and clause order mediate the comprehension of temporal sentences. Sixty monolingual Greek-speaking children and 15 adult controls completed a picture-sequence selection task in which they judged after- and before-sentences in iconic and non-iconic order. Up to age twelve, children did not reach full adult-like comprehension of the connectives; performance in non-iconic after-sentences was significantly lower than in the other three conditions across all ages. We conclude that neither iconicity, connective, nor clause order can fully explain these findings and propose an account based on the interaction of iconicity and clause order: non-iconic, sentence-medial after requires revision of the initial event representation, resulting in an event-semantic kindergarten-path that children find difficult to overcome.

Content Summarization and Question Answering System Using LLM

This paper presents a novel approach to content summarization, a pivotal task in information processing and knowledge extraction. Our methodology integrates cutting-edge technologies, including large language models (LLMs) and ad-vanced retrieval techniques, to efficiently distill extensive textual data into concise and informative summaries. By leveraging Anyscale LLMs for language comprehension and employing prompt engineering for model guidance, we propose a frame-work that prioritizes efficient document indexing and rapid retrieval using DuckDB vector stores. Additionally, we introduce the RetrievalQAWithSourcesChain framework, which combines LLMs with retrievers for proficient question answering. Em-pirical validation demonstrates the efficacy of our approach in generating high-quality content summaries with minimal manual intervention, thereby enhancing information retrieval and knowledge extraction processes. Index Terms—LLMs, Anyscale, prompt engineering, document indexing, DuckDB, vector stores, RetrievalQAWithSourcesChain framework, question answering, empirical validation

Examining Adolescent Language Performance in Discourse Production Across Four Elicitation Tasks.

Comprehensive spoken language assessment should include the evaluation of language use in naturalistic contexts. Discourse elicitation and analysis provides the opportunity for such an evaluation to occur. In this article, our overall aim was to describe adolescents' language performance on four elicitation tasks and determine if there are task-related differences across the elicitation tasks. Forty-four typically developing adolescents with ages ranging from 12;2 to 17;11 (years;months; M = 15;2; 21 boys and 23 girls) participated in the study. They completed four spoken discourse tasks: (a) story generation using a wordless picture book, (b) fable retell, (c) six personal narratives in response to emotion-based prompts, and (d) monologic response to two stories that contained a moral dilemma. Responses were transcribed and analyzed for four language performance measures tapping into language productivity, syntactic complexity, lexical diversity, and verbal facility. Despite individual variability in performance, mean scores were close to median scores for most measures, suggesting a symmetrical distribution. As expected, all four language performance measures were significantly different across the four elicitation tasks. The personal narrative task elicited the longest samples, with the highest verbal fluency. In contrast, both lexical diversity and syntactic complexity were the strongest in response to the fable retell and the moral dilemma tasks. This investigation provides speech-language pathologists with an overview of how task-related factors may impact adolescent language performance. These findings may be used to support their clinical decision-making processes in choosing a suitable discourse task when conducting a comprehensive spoken language assessment. Three hypothetical case examples are used to illustrate the decision-making process. https://doi.org/10.23641/asha.25761768.

The Concise Language Paradigm (CLaP), a framework for studying the intersection of comprehension and production: electrophysiological properties.

Studies investigating language commonly isolate one modality or process, focusing on comprehension or production. Here, we present a framework for a paradigm that combines both: the Concise Language Paradigm (CLaP), tapping into comprehension and production within one trial. The trial structure is identical across conditions, presenting a sentence followed by a picture to be named. We tested 21 healthy speakers with EEG to examine three time periods during a trial (sentence, pre-picture interval, picture onset), yielding contrasts of sentence comprehension, contextually and visually guided word retrieval, object recognition, and naming. In the CLaP, sentences are presented auditorily (constrained, unconstrained, reversed), and pictures appear as normal (constrained, unconstrained, bare) or scrambled objects. Imaging results revealed different evoked responses after sentence onset for normal and time-reversed speech. Further, we replicated the context effect of alpha-beta power decreases before picture onset for constrained relative to unconstrained sentences, and could clarify that this effect arises from power decreases following constrained sentences. Brain responses locked to picture-onset differed as a function of sentence context and picture type (normal vs. scrambled), and naming times were fastest for pictures in constrained sentences, followed by scrambled picture naming, and equally fast for bare and unconstrained picture naming. Finally, we also discuss the potential of the CLaP to be adapted to different focuses, using different versions of the linguistic content and tasks, in combination with electrophysiology or other imaging methods. These first results of the CLaP indicate that this paradigm offers a promising framework to investigate the language system.

Language impairment in sporadic and familial (type 8) amyotrophic lateral sclerosis: A comparative study.

Language is frequently affected in patients with sporadic amyotrophic lateral sclerosis (sALS), with reduced performance in naming, syntactic comprehension, grammatical expression, and orthographic processing. However, the language profile of patients with familial type 8 ALS (ALS8), linked to p.P56S VAPB mutation, remains unclear. We investigated language in patients with ALS8 by examining their auditory comprehension and verbal production. We included three groups of participants: (1) patients with sALS (n = 20), (2) patients with familial ALS8 (n = 22), and (3) healthy controls (n = 21). The groups were matched for age, sex, and education level. All participants underwent a comprehensive language battery, including the Boston Diagnostic Aphasia Examination, the reduced Token test, letter fluency, categorical fluency (animals), word definition from the Cambridge Semantic Memory Research Battery, and a narrative discourse analysis. Participants also were evaluated using Addenbrooke's Cognitive Exam-Revised Version, the Hospital Anxiety and Depression Scale, and the ALS Functional Rating Scale-Revised. Compared to controls, sALS and ALS8 patients had impaired performance on oral (syntactic and phonological processing) comprehension and inappropriate discourse cohesion. sALS and ALS8 did not differ in any language measure. There was no correlation between language scores and functional and psychiatric scales. ALS8 patients exhibit language deficits that are independent of motor features. These findings are consistent with the current evidence suggesting that ALS8 has prominent non-motor features.

Fast Recurrent Neural Network with Bi-LSTM for Handwritten Tamil Text Segmentation in NLP

Tamil text segmentation is a long-standing test in language comprehension that entails separating a record into adjacent pieces based on its semantic design. Each segment is important in its own way. The segments are organised according to the purpose of the content examination as text groups, sentences, phrases, words, characters or any other data unit. That process has been portioned using rapid tangled neural organisation in this research, which presents content segmentation methods based on deep learning in natural language processing (NLP). This study proposes a bidirectional long short-term memory (Bi-LSTM) neural network prototype in which fast recurrent neural networks (FRNNs) are used to learn Tamil text group embedding and phrases are fragmented using text-oriented data. As a result, this prototype is capable of handling variable measured setting data and gives a vast new dataset for naturally segmenting text in Tamil. In addition, we develop a segmentation prototype and show how well it sums up to unnoticeable regular content using this dataset as a base. With Bi-LSTM, the segmentation precision of FRNN is superior to that of other segmentation approaches; however, it is still inferior to that of certain other techniques. Every content is scaled to the required size in the proposed framework, which is immediately accessible for the preparation. This means, each word in a scaled Tamil text is employed to prepare neural organisation as fragmented content. The results reveal that the proposed framework produces high rates of segmentation for manually authored material that are nearly equivalent to segmentation-based plans.

A Systematic Review of Using Deep Learning in Aphasia: Challenges and Future Directions

In this systematic literature review, the intersection of deep learning applications within the aphasia domain is meticulously explored, acknowledging the condition’s complex nature and the nuanced challenges it presents for language comprehension and expression. By harnessing data from primary databases and employing advanced query methodologies, this study synthesizes findings from 28 relevant documents, unveiling a landscape marked by significant advancements and persistent challenges. Through a methodological lens grounded in the PRISMA framework (Version 2020) and Machine Learning-driven tools like VosViewer (Version 1.6.20) and Litmaps (Free Version), the research delineates the high variability in speech patterns, the intricacies of speech recognition, and the hurdles posed by limited and diverse datasets as core obstacles. Innovative solutions such as specialized deep learning models, data augmentation strategies, and the pivotal role of interdisciplinary collaboration in dataset annotation emerge as vital contributions to this field. The analysis culminates in identifying theoretical and practical pathways for surmounting these barriers, highlighting the potential of deep learning technologies to revolutionize aphasia assessment and treatment. This review not only consolidates current knowledge but also charts a course for future research, emphasizing the need for comprehensive datasets, model optimization, and integration into clinical workflows to enhance patient care. Ultimately, this work underscores the transformative power of deep learning in advancing aphasia diagnosis, treatment, and support, heralding a new era of innovation and interdisciplinary collaboration in addressing this challenging disorder.

The Operational Laws of Symmetric Triangular Z-Numbers

To model fuzzy numbers with the confidence degree and better account for information uncertainty, Zadeh came up with the notion of Z-numbers, which can effectively combine the objective information of things with subjective human interpretation of perceptive information, thereby improving the human comprehension of natural language. Although many numbers are in fact Z-numbers, their higher computational complexity often prevents their recognition as such. In order to reduce computational complexity, this paper reviews the development and research direction of Z-numbers and deduces the operational rules for symmetric triangular Z-numbers. We first transform them into classical fuzzy numbers. Using linear programming, the extension principle of Zadeh, the convolution formula, and fuzzy number algorithms, we determine the operational rules for the basic operations of symmetric triangular Z-numbers, which are number-multiplication, addition, subtraction, multiplication, power, and division. Our operational rules reduce the complexity of calculation, improve computational efficiency, and effectively reduce the information difference while being applicable to other complex operations. This paper innovatively combines Z-numbers with classical fuzzy numbers in Z-number operations, and as such represents a continuation and innovation of the research on the operational laws of Z-numbers.

A Paradigm for the missing middle

ABSTRACT This commentary highlights what I see as the most important of Walter Kintsch’s many contributions to cognitive psychology and especially text comprehension. His work filled the great void between the top and the bottom in research on comprehension. The state of the field at the time was (a) strong research foundation on word reading, (b) detail-focused research on syntactic processes by a vibrant sentence processing research community, and (c) demonstration research on the importance of higher-level top down influences. The void between (b) and (c) meant there was no possibility of a comprehension science that was both rigorous, like (a) and (b), and reflective of the role of outside-the-text knowledge, like (c). Beyond the specific empirical contributions of Kintsch’s work is that he showed that it was possible to create a paradigm for the missing middle, a characterization of how word-driven and knowledge-driven processes yield a coherent meaning-based understanding of a text a few pieces at a time. This allowed the flourishing of work on the incremental processes that result in a reader’s (often) coherent representation of text meaning.

Negation and social avoidance in language recruits the right inferior frontal gyrus: a tDCS study.

In the process of comprehension, linguistic negation induces inhibition of negated scenarios. Numerous studies have highlighted the role of the right Inferior Frontal Gyrus (rIFG) - a key component of the inhibitory network - in negation processing. Social avoidance can be linguistically portrayed using attitudinal verbs such as "exclude" vs. "include", which inherently carry negative connotations. Consequently, we hypothesize that the interplay between explicit negation and the implicit negativity of avoidance verbs can be modulated via transcranial direct current stimulation (tDCS) targeting the rIFG. In our study, sixty-four participants read approach/avoidance sentences, which were either affirmative or negative, such as "Anne included (did not include) meat in her diet" vs. "Anne excluded (did not exclude) meat in her diet". This reading task followed a 20-minute tDCS session. The sentences were sequentially displayed, and at 1500 ms post-sentence, a verb was shown - either the one previously mentioned or its semantic alternative counterpart (e.g., included vs. excluded). Findings revealed that anodal stimulation intensifies the inhibitory impact of negation during sentence comprehension. Under anodal conditions, negative sentences led to extended reading times for the mentioned verbs compared to their affirmative counterparts, suggesting an increased inhibitory effect on the verb. Furthermore, in avoidance sentences, anodal stimulation resulted in reduced reading times for alternative verbs (e.g. "included") in negative sentences compared to alternative verbs (e.g. "excluded") in negated approach sentences. As "avoidance" is semantically equivalent to "non-approach", the inhibitory effect of negation is primarily applied to the implicit negation: NOT EXCLUDED = NOT→NOT (INCLUDED), which consequently activates the representation of the alternative verb making it more available. We further discuss these findings in light of the rIFG's pivotal role in processing attitudinal verbs and linguistic negation. This discussion is framed within the overarching context of the two-step model of negation processing, highlighting its significance in the realm of social communication.

Overcoming Language Barriers in Contracts – A South African Perspective

Language forms the basis of written contracts, but the use of language may also function as a barrier to contracting parties, particularly when the parties are not proficient in the commercial language of choice, such as English in the case of South Africa. The plain language movement and specifically legislative interventions have placed significant pressure on lawmakers to reform the use of traditional legal English, and consequently, traditional drafting styles. However, many contracts continue to exhibit convoluted language, legalese and poor visual appeal, which ultimately functions as a language barrier for many people. This article examines the various language barriers that exist in contracts from a South African perspective. It does so by assessing the barriers that exist under first, the common law (or the so-called default rules), second, the use of standard contracts, and third, the resistance to the use of plain language as well as the limited application of existing plain language legislation in certain types of contractual engagements. The author proposes dismantling the language barriers by adopting both textual and visual communication devices in order to achieve clear and comprehensible language in all types of contracts. Furthermore, it is suggested that such communication devices should not necessarily be limited to consumer contracts that are legislatively required to comply with plain language requirements. It is only by dismantling existing language barriers that social justice within contracts can be achieved in a society with diverse language proficiencies, such as South Africa. Therein, this article offers insights that are relevant not only within the South African context but also for countries that may face similar language barriers characterised by multiple languages and varying language proficiencies within their populations.

The role of prediction error in the development of language learning and memory

AbstractPrediction error plays a pivotal role in theories of learning, including theories of language acquisition and use. Researchers have investigated whether and under which conditions children, like adults, use prediction to facilitate language comprehension at different levels of linguistic representation. However, many aspects of the reciprocal relation between prediction error and the development of language learning remain unclear. In this article, we review studies in language development that can inform us about the role of prediction error in updating, learning, and retrieving linguistic information. We argue that the study of individual differences in linguistic and cognitive skills will help the field understand more thoroughly whether, when, and why prediction aids language learning, and whether prediction error necessarily results in language learning and retrieval from memory. We close with a discussion of the needs and challenges for researchers to answer these questions.

Consumer Profiles and Drink Menus Made from Medicinal Plants at the Jamu Cafe in Depok Sleman District, Yogyakarta

The use of medicinal plants as ingredients in making drinks in cafes is increasing along with the number of cafes. Herbal café makes a unique menu using medicinal plants in drink recipes to produce a distinctive taste, color, and aroma. This research aims to identify visitor profiles and drink menus made from medicinal plants and explore the understanding of visitors and cafe owners regarding the benefits of medicinal plants added to drinks. The research data collection technique uses a questionnaire equipped with an interview sheet. The questionnaires and interview sheets used have been tested for language comprehension and validity. The research was conducted at 272 cafes in Depok District, Sleman Regency, Special Region of Yogyakarta. The sample was determined using purposive sampling, data was collected after obtaining ethical clearance approval from the Health Research Ethics Committee Team. Data was obtained through filling out surveys by visitors and cafe owners, complemented by interviews with cafe owners. Data was collected three times: August, September, and October 2023. Data analysis was carried out descriptively and qualitatively. Most herbal cafe visitors (70%) are aged between 21–30 years with the majority of professions (44%) being students. The source of information about herbal cafe menus was mostly obtained (56%) from friends or family, and the frequency of visits to cafes was mostly (82%) between 1–5 times per month. More than 98% of visitors stated the benefits of medicinal plants in the drink menu, namely as a refresher or body warmer. There are 14 types of medicinal plants used as ingredients in the herbal cafe's drink menu, with a total monthly requirement for all medicinal plants of 545.1 kilograms; ginger ranks among the highest requirements. Recipe information for the menu was obtained from the family, the Ministry of Health Republic Indonesia website, and the herbal cafe owner made the recipe independently.

Bi-syntax guided transformer network for aspect sentiment triplet extraction

Aspect Sentiment Triplet Extraction is an emerging and challenging task that attempts to present a complete picture of aspect-based sentiment analysis. Prior research efforts mostly leverage various tagging schemes to extract the three elements in a triplet. However, these methods fail to explicitly model the complicated relations between aspects and opinions and the boundaries of multi-word aspects and opinions. In this paper, we propose a bi-syntax guided transformer network in an end-to-end manner to address these challenges. Firstly, we devise three types of representations, including sequence distance representation, constituency distance representation, and dependency distance representation, to learn the comprehensive language representation. Specifically, sequence distance representation utilizes sequence distance between words to enhance the contextual representation. Constituency distance representation adopts constituency distance between words in a constituency tree to capture the intra-span relation between words. Dependency distance representation employs dependency distance between words in a dependency tree to capture the long-distance relation between aspects and opinions. Extensive experiments are conducted on four benchmark datasets to validate the effectiveness of our method. The results demonstrate that the proposed approach achieves better performance than baseline methods. We conduct further detailed analysis to demonstrate that our method effectively handles multi-word terms and overlapping triplets.

Exploring the Interplay Between Language Comprehension and Cortical Tracking: The Bilingual Test Case

Abstract Cortical tracking, the synchronization of brain activity to linguistic rhythms is a well-established phenomenon. However, its nature has been heavily contested: Is it purely epiphenomenal or does it play a fundamental role in speech comprehension? Previous research has used intelligibility manipulations to examine this topic. Here, we instead varied listeners’ language comprehension skills while keeping the auditory stimulus constant. To do so, we tested 22 native English speakers and 22 Spanish/Catalan bilinguals learning English as a second language (SL) in an EEG cortical entrainment experiment and correlated the responses with the magnitude of the N400 component of a semantic comprehension task. As expected, native listeners effectively tracked sentential, phrasal, and syllabic linguistic structures. In contrast, SL listeners exhibited limitations in tracking sentential structures but successfully tracked phrasal and syllabic rhythms. Importantly, the amplitude of the neural entrainment correlated with the amplitude of the detection of semantic incongruities in SLs, showing a direct connection between tracking and the ability to understand speech. Together, these findings shed light on the interplay between language comprehension and cortical tracking, to identify neural entrainment as a fundamental principle for speech comprehension.