Task Description Research Articles

Role of Rubric in Assessment of Language Learning in Higher Education

The mixed-method study aims at exploring the effectiveness of rubrics towards authentic assessment of language learning in higher education by understanding the role of the rubric. In the recent trend and style of language teaching-learning, the rubric is one of the important tools that design the set standards and criteria which teachers use to assess students’ performance, assignment or behavior. In educational technology, a rubric is the performance standard for a student population typically having four essential components: task description, the scale of achievement, dimensions and description of dimensions. The type of rubric is chosen from the range of holistic rubrics, analytic rubrics, generic rubrics or specific rubrics for assessment depending on the task being evaluated and the needs of the assessor. Research says that the rubric helps to define “quality performance” and promote awareness on critical components in a performance not only acting as an evaluation tool for instructors but also acting as a feedback proforma for students. According to the rubric experts, rubrics are vital tools that can be utilized to solve the problem of subjectivity in the evaluation as they provide consistency in evaluation, reduce subjectivity and enhance objectivity. To explore the real scenario, the researcher collected data through the analysis of students’ performance in two classes where the teachers used the rubric and in two classes where the teachers did not use it. Besides, a semi open-ended questionnaire was administered among four teachers, two of each category to see their perspectives about the impact of rubrics in reducing subjectivity. Alongside, four FGD were conducted to put light on students’ perceptions. The result shows a significant impact of rubrics on students’ performance along with a strong positive attitude of both students and teachers towards the use of rubrics. A wide scope of further research is open in this area. Teacher’s World: Journal of Education and Research, Vol. 48(2), Dec 2022, p.112-129

'Eye-Tracking' with Words for Alzheimer's Disease Detection: Time Alignment of Words Enunciation with Image Regions During Image Description Tasks.

Alzheimer's disease (AD) is a progressive neurodegenerative disease that results in cognitive decline, dementia, and eventually death. Diagnosing early signs of AD can help clinicians to improve the quality of life. We developed a non-invasive approach to help neurologists and clinicians to distinguish probable AD patients and healthy controls (HC). The patients' gaze points were followed based on the words they used to describe the Cookie Theft (CT) picture description task. We hypothesized that the timing of words enunciation aligns with the participant's eye movements. The moments that each word was spoken were then aligned with specific regions of the image. We then applied machine learning algorithms to classify probable AD and HC. We randomly selected 60 participants (30 AD and 30 HC) from the Dementia Bank (Pitt Corpus). Five main classifiers were applied to different features extracted from the recorded audio and participants' transcripts (AD and HC). Support vector machine and logistic regression had the highest accuracy (up to 80% and 78.33%, respectively) in three different experiments. In conclusion, point-of-gaze can be applied as a non-invasive and less expensive approach compared to other available methods (e.g., eye tracker devices) for early-stage AD diagnosis.

ФОРМАЛИЗОВАННОЕ ПРЕДСТАВЛЕНИЕ ШКОЛЬНЫХ МАТЕМАТИЧЕСКИХ ЗАДАЧ

The article deals with the problem of using formalization as a theoretical method of scientific cognition in the scientific description of school mathematical tasks. The essence of the formalization method and one of the results of its application in the methodology of teaching mathematics are demonstrated.

Limited connectedness of spontaneous speech may be a marker of dementia due to Alzheimer's disease.

The study evaluated the connectedness of spontaneous speech production in individuals with dementia as a potential predictor of dementia severity. Data were derived from the baseline sample of 143 individuals with dementia in the English Pitt corpus. Dementia severity was assessed via the Mini Mental Status Exam, the Mattis Dementia Rating Scale, and the Blessed Dementia Scale. Language abilities were evaluated using verbal fluency and picture description tasks. Graph analysis was carried out for the picture description task using the computational tool SpeechGraphs to calculate connectedness. Results demonstrated that higher educational attainment, higher verbal fluency and strongly-connected spontaneous speech were associated with better cognitive function. Results suggest that automated language processing approaches, such as graph structure analysis, may provide a faster and ecologically valid method of detecting dementia symptoms.

The Role of Animacy in Turkish Relative Clause Production and Distribution.

Relative clause (RC) production has been a major tool used for understanding language production mechanism in experimental linguistics. The present study analyzes language production mechanisms in Turkish by utilizing animacy effect on RC production. A picture description task was applied to two participant groups. The data were combined and analyzed to see how animacy influenced RC formation. The outcomes were also compared to the distributions of RC use in corpus data. Both participant and corpus data demonstrated significant level of passivization for RCs with animate heads, strongly affirming the grammatical function assignment proposal by Bock and Warren (Cognition 21(1):47-67, 1985) as well as the premise of Production-Distribution-Comprehension account (MacDonald in Front Psychol 4:226, 2013), emphasizing the relationship between language production mechanisms and typology. However, the corpus data were observed to have higher numbers of passivization with animate condition. Accordingly, a coarse comparison of the participant RC production proportions in the current study with some other crosslinguistic research suggests that some language-specific or discourse-related interventions can also compete with the animacy accessibility during the sentence planning procedure, which needs an extra inquiry especially in Turkish language.

Defining the Characteristics of Story Production of Autistic Children: A Multilevel Analysis

Several studies suggest that a valuable tool to examine linguistic skills in communication disorders is offered by procedures of narrative discourse assessment. Following this line of research, we present an exploratory study aimed to investigate storytelling abilities of autistic children to better define the characteristics of their story production. Participants included 41 autistic children and 41 children with typical development aged between 7.02 and 11.03 years matched on age, gender, level of formal education, intelligence quotient, working memory, attention skills, theory of mind, and phonological short-term memory. Narrative production was assessed by analysing the language samples obtained through the “Nest Story” description task. A multilevel analysis including micro- and macro-linguistic variables was adopted for narrative assessment. Group differences emerged on both micro- and macro-linguistic dimensions: autistic children produced narratives with more phonological errors and semantic paraphasias (microlinguistic variables) as well as more errors of global coherence and a fewer number of visible events and inferred events (macrolinguistic variables) than the control group.This study shows that even autistic children with adequate cognitive skills display several limitations in their narrative competence and that such weaknesses affect both micro- and macrolinguistic aspects of story production.

Extracting Pseudocode from Digital Block Diagram in Technical Documents

The technical document is an entity that consists of several essential and interconnected parts, often referred to as modalities. Despite the extensive attention that certain modalities have already received, say per the textual information, there are several aspects that get disproportional attention throughout the bibliography. An instance of such a modality is the block diagrams found in the technical document, as they can provide crucial information about the functionality and the technical details of both their related parts and the technical document as a whole. This paper deals with the automatic understanding of block diagrams and the extraction of pseudocode associated with the functionality of a given diagram. In particular we present a complete methodology for the formal modelling of digital block diagrams and their elements, then we develop a generative framework and three novel annotated datasets on diagrams classification and captioning. After mapping the initial problem to a block diagram description task, we present several original predictive setups derived from image segmentation, analyze the inference capabilities, and we offer illustrative examples justifying our approach.

Referential noun phrases distribute differently in Turkish speakers with schizophrenia.

In all human languages, noun phrases (NPs) (e.g., 'a field', 'the woman with a book') are used to identify entities in discourse. Previous evidence has shown that the spontaneous speech of patients with schizophrenia (Sz) shows differences in the distribution of grammatically different types of NPs, which are in part specific to patients with formal thought disorder (FTD). Here we sought to provide the first evidence of related grammatical effects in a non-Indo-European language. Results from a picture description task in a sample of 16 Turkish speakers with FTD (+FTD), 15 without FTD (-FTD), and 27 controls revealed that relative to controls, people with Sz over-produced NPs that are 'bare' (in the sense of lacking any grammatical items such as the or a in English). The +FTD group generally showed stronger effects than -FTD, and used more pronouns and less NPs co-referring with previously mentioned NPs. In addition, the dynamic distribution of NP types over narrative time showed an effect of increased mean distance between definite NPs in -FTD relative to controls. In +FTD but no other group there was an unexpected random distribution of indefinite DPs. Incidence rates of referential anomalies increased from controls to the -FTD and +FTD groups. These findings further confirm that Sz is manifest through specific linguistic effects in the referential structure of meaning as mediated by grammar. They provide a linguistic baseline for neurocognitive models of FTD and help to define appropriate targets for the automatic extraction of linguistic features to classify psychotic speech.

Generative pretrained autoregressive transformer graph neural network applied to the analysis and discovery of novel proteins

We report a flexible language-model-based deep learning strategy, applied here to solve complex forward and inverse problems in protein modeling, based on an attention neural network that integrates transformer and graph convolutional architectures in a causal multi-headed graph mechanism, to realize a generative pretrained model. The model is applied to predict the secondary structure content (per-residue level and overall content), protein solubility, and sequencing tasks. Further trained on inverse tasks, the model is rendered capable of designing proteins with these properties as target features. The model is formulated as a general framework, completely prompt-based, and can be adapted for a variety of downstream tasks. We find that adding additional tasks yields emergent synergies that the model exploits in improving overall performance, beyond what would be possible by training a model on each dataset alone. Case studies are presented to validate the method, yielding protein designs specifically focused on structural materials, but also exploring the applicability in the design of soluble, antimicrobial biomaterials. While our model is trained to ultimately perform eight distinct tasks, with available datasets, it can be extended to solve additional problems. In a broader sense, this study illustrates a form of multiscale modeling that relates a set of ultimate building blocks (here, byte-level utf8 characters that define the nature of the physical system at hand) to complex output. This materiomic scheme captures complex emergent relationships between universal building block and resulting properties, via a synergizing learning capacity, to express a set of potentialities embedded in the knowledge used in training via the interplay of universality and diversity. Significance statement: Predicting the properties of materials based on a flexible description of their structure, environment, or process, is a long-standing challenge in multiscale modeling. Our MaterioFormer language model, trained to solve forward and inverse tasks, incorporates a deep learning capacity through attention and graph strategies to yield a multimodal approach to model and design materials. Since our model is prompt-based and information is encoded consistently via byte-level utf8 tokenization, it can process diverse modalities of information, such as sequence data, description of tasks, and numbers, and offers a flexible workflow that integrates human intelligence and artificial intelligence. Autoregressive training, using pre-training against a large unlabeled dataset, allows for straightforward adjustment of specific objectives.

Attracting solvers' participation in crowdsourcing contests: The role of linguistic signals in task descriptions

AbstractMany companies gain external expertise, lower their costs and generate publicity by using crowdsourcing platforms to complete tasks by leveraging the power of the crowd. However, the number of solvers attracted by crowdsourcing tasks varies widely. Although some well‐known crowdsourcing contests have attracted large numbers of participants, many tasks still suffer from low participation rates. Prior research aimed at solving this problem has focused on factors such as task rewards and durations while overlooking whether a well‐written description might motivate solvers to choose a task. Based on signalling theory, this study investigates the effect of task descriptions on solvers' participation by focusing on informational and affective linguistic signals. Our model is validated by analysing 13 929 descriptions posted in single‐winner tasks on epwk.com, a Chinese competitive crowdsourcing platform. For informational linguistic signals, the results reveal that there are inverted U‐shaped relationships between both concreteness and specificity and solver participation, whereas linguistic accuracy has a positive effect on solver participation. For affective linguistic signals, positive emotional words have a positive relationship with solver participation, whereas negative emotional words have the opposite effect. Theoretical and practical implications are discussed.

The correlates of flow in the L2 classroom: Linking basic L2 task features to learner flow experiences

AbstractFlow is an intrinsic motivational state associated with full task engagement, positive affect, and enhanced performance. While research has examined how different language tasks interact with flow experiences, no study has examined learner flow experiences in a wide range of tasks using an experience sampling method to determine how universal basic task features (e.g., modality, participant structure, information distribution, and targeted skills) interact with flow. The present study aims to respond to this gap in the research. Participants were 13 teachers and 327 students from 18 intact French L2 classes in a Canadian postsecondary school. Teachers selected and implemented an average of six tasks from their personal repertoires at random moments throughout the semester. Immediately following each task, learners anonymously completed a flow experience questionnaire (N = 1408; α = 0.91), and teachers a task description questionnaire containing 17 basic task features (N = 81). Statistical analyses show that 10 of the 17 variables significantly interacted with learners’ flow experiences. The results not only validate a frequently used flow measurement and establish norms for future research but also outline a framework language teachers can use to evaluate and modify practices to improve learners’ subjective classroom experience.

Defying syntactic preservation in Alzheimer's disease: what type of impairment predicts syntactic change in dementia (if it does) and why?

IntroductionMany studies on syntax in dementia suggest that, despite syntactic simplification, speakers with Alzheimer's disease (AD) retain their basic grammatical abilities, being mainly affected in their comprehension and production of complex syntax. Moreover, there is no single position on the origin of syntactic decline in AD, which, according to some authors, can be linked to a lexical-semantic deficit or, according to others, to either cognitive or autonomous dysfunction.MethodsIn this study, we apply the model of syntactic maturity to the analysis of oral speech production elicited by the Cookie-Theft description task. We assess a sample of 60 older adults (21 HC, 19 MCI, and 20 AD) through three indexes of syntactic maturity, measuring the proportion of sentences and clauses in discourse, their mean length, and the rate of their complexity.ResultsOur results show two important tendencies in AD: the preservation of general syntactic ability, as measured by the basic syntactic organization of speech, and the disturbance of the indexes of syntactic complexity, as measured by the overall length of utterances and their indexes of complexity.DiscussionAlthough speakers with AD maintain the ability to construct grammatically acceptable sentences and produce a similar number of utterances to healthy aging speakers and speakers with MCI, the syntactic complexity of their discourse significantly changes. Importantly, such significant changes are already present at the MCI stage and are not conditioned by the lexical-semantic deficit itself. Our results may be particularly relevant to improving the detection of cognitive impairment and to theoretically discussing the relationships between language levels in aging speakers.

Narrative discourse production in Parkinson’s disease: Decoupling the role of cognitive-linguistic and motor speech changes

Introductionthe interplay between neuropsychological and communicative abilities in Parkinson’s disease (PD) has been relatively overlooked, and it is not entirely understood which difficulties are consequent to impaired motor control, and which have a linguistic/cognitive basis. Here, we examined narrative discourse in PD using a multi-level analysis procedure considering sentence-level (productivity, lexical-grammatical processing) and discourse-level processes (narrative organization, informativeness), and partialling out patients’ motor speech impairments. The interaction between cognitive (i.e. linguistic and executive) and communication abilities was also investigated. MethodsTwenty-nine PD subjects in the mild stage of the disease were compared to 29 matched healthy comparators (HC) on quantitative measures of narrative discourse derived from two picture description tasks. Multivariate (considering articulation rate and educational attainment as covariates) and univariate (with group membership as independent variable) analyses of variance were conducted on separate linguistic domains. The contribution of executive/linguistic abilities to PD’s narrative performance was explored by multiple regression analyses on narrative measures significantly differentiating patients from HC. Resultssignificant reductions in patients were observed on measures of productivity (less well-formed words, shorter sentences) and informativeness (fewer conceptual units, less informative elements, lower number of details) and these alterations were explained by variations in linguistic abilities (action and object naming) rather than executive abilities. Articulation rate and educational attainment did not impact the observed reduced productivity and under-informativeness. Conclusionreferential narrative discourse is altered in PD, regardless of motor impairments in speech production. The observed reductions in productivity/informativeness aspects of narratives were related to naming abilities and in particular to verbs processing, consistently with the neurocognitive model of motor language coupling. Since narratives are amenable to recurrent and automated analysis for the identification of linguistic patterns potentially anticipating the development of PD and the onset of cognitive deterioration, discourse abilities should be quantitatively and repeatedly profiled in the disorder.

Delphi Survey of Items for the Test of Stuttering Screening in Children (TSSC)

Purpose: Stuttering is a fluency disorder that mostly begins in childhood and affects many people in our societies. No standardized screening tools are available to check for stuttering in the Indian school-going population. Thus, the study aimed at developing a screening tool to identify children who stutter among the school-going population using a Delphi-based approach. Methods: This study was carried out in four phases. During the first phase, five Speech Language Pathologists (SLPs) were asked about the need for screening and the nature & attributes of a stuttering screening test for school-going children. The second phase involved constructing appropriate stimuli for the screening tool based on expert opinion, relevant literature and students’ academic textbooks. The third phase involved content validation of the speech elicitation stimuli by four teachers, five SLPs and an English Lecturer teaching in a university. The fourth phase encompassed the development of differential diagnosis criteria for stuttering identification in children using a rank analysis of the expert opinions. Results: A stuttering screening stimuli comprising age, language and culture-specific reading, picture description and narration tasks for 1st to 10th-standard students was developed. The contents of the tool obtained satisfactory consensus of agreement among the panel of experts. Further, the tool outlined five critical diagnostic criteria which could differentially diagnose school-going children with stuttering from typically speaking counterparts using the developed material. Conclusion: The developed screening tool could help practicing clinicians quickly identify stuttering in school-going populations. This would enable early identification and build up the statistical data to estimate the prevalence of stuttering among the school-going population. Further studies examining the psychometric properties of the developed test are in progress.

Delphi Survey of Items for the Test of Stuttering Screening in Children (TSSC).

Stuttering is a fluency disorder that mostly begins in childhood and affects many people in our societies. No standardized screening tools are available to check for stuttering in the Indian school-going population. Thus, the study aimed at developing a screening tool to identify children who stutter among the school-going population using a Delphi-based approach. This study was carried out in four phases. During the first phase, five Speech Language Pathologists (SLPs) were asked about the need for screening and the nature & attributes of a stuttering screening test for school-going children. The second phase involved constructing appropriate stimuli for the screening tool based on expert opinion, relevant literature and students' academic textbooks. The third phase involved content validation of the speech elicitation stimuli by four teachers, five SLPs and an English Lecturer teaching in a university. The fourth phase encompassed the development of differential diagnosis criteria for stuttering identification in children using a rank analysis of the expert opinions. A stuttering screening stimuli comprising age, language and culture-specific reading, picture description and narration tasks for 1st to 10th-standard students was developed. The contents of the tool obtained satisfactory consensus of agreement among the panel of experts. Further, the tool outlined five critical diagnostic criteria which could differentially diagnose school-going children with stuttering from typically speaking counterparts using the developed material. The developed screening tool could help practicing clinicians quickly identify stuttering in school-going populations. This would enable early identification and build up the statistical data to estimate the prevalence of stuttering among the school-going population. Further studies examining the psychometric properties of the developed test are in progress.

ADscreen: A speech processing-based screening system for automatic identification of patients with Alzheimer's disease and related dementia

Alzheimer's disease and related dementias (ADRD) present a looming public health crisis, affecting roughly 5 million people and 11 % of older adults in the United States. Despite nationwide efforts for timely diagnosis of patients with ADRD, >50 % of them are not diagnosed and unaware of their disease. To address this challenge, we developed ADscreen, an innovative speech-processing based ADRD screening algorithm for the protective identification of patients with ADRD. ADscreen consists of five major components: (i) noise reduction for reducing background noises from the audio-recorded patient speech, (ii) modeling the patient's ability in phonetic motor planning using acoustic parameters of the patient's voice, (iii) modeling the patient's ability in semantic and syntactic levels of language organization using linguistic parameters of the patient speech, (iv) extracting vocal and semantic psycholinguistic cues from the patient speech, and (v) building and evaluating the screening algorithm. To identify important speech parameters (features) associated with ADRD, we used the Joint Mutual Information Maximization (JMIM), an effective feature selection method for high dimensional, small sample size datasets. Modeling the relationship between speech parameters and the outcome variable (presence/absence of ADRD) was conducted using three different machine learning (ML) architectures with the capability of joining informative acoustic and linguistic with contextual word embedding vectors obtained from the DistilBERT (Bidirectional Encoder Representations from Transformers). We evaluated the performance of the ADscreen on an audio-recorded patients' speech (verbal description) for the Cookie-Theft picture description task, which is publicly available in the dementia databank. The joint fusion of acoustic and linguistic parameters with contextual word embedding vectors of DistilBERT achieved F1-score = 84.64 (standard deviation [std] = ±3.58) and AUC-ROC = 92.53 (std = ±3.34) for training dataset, and F1-score = 89.55 and AUC-ROC = 93.89 for the test dataset. In summary, ADscreen has a strong potential to be integrated with clinical workflow to address the need for an ADRD screening tool so that patients with cognitive impairment can receive appropriate and timely care.

The acquisition of Hindi split-ergativity and differential object marking by Dutch L1 speakers: systematicity and variation

ABSTRACT We investigated the acquisition of Hindi split ergativity (zero or -ne marking) and Hindi differential object marking (DOM; zero or -ko marking) by L1 speakers of Dutch. Both grammatical phenomena are conditioned by multiple syntactic and semantic features. On a descriptive level, the study aims to examine when and how Dutch-speaking learners acquire and apply the conditional features associated with -ne and -ko marking in Hindi as a foreign language (HFL). A specific learner corpus was created based on a picture description task that elicited semi-spontaneous oral production data from 15 Dutch-speaking learners of Hindi, from four cross-sectional stages of the Hindi course trajectory. We annotated the corpus data for multiple features associated with -ne and -ko marking. Using a mixed-effects logistic regression analysis, we found an increase in the use and accuracy of each case marker over the different years of study, but individual learner profile analyses revealed considerable intersubject differences in learner behavior. We argue that the developmental stages for the emergence of -ne and -ko marking are in line with predictions based on Processability Theory. We additionally include mastery level analysis to account for a combined perspective on language development. Our findings reveal that HFL learners reach higher mastery levels for split ergativity than for DOM, even though DOM (-ko marking) emerges before split ergativity (-ne marking). We conclude that developmental stages and between-learner variation are not mutually exclusive.

Interactively learning behavior trees from imperfect human demonstrations.

Introduction: In Interactive Task Learning (ITL), an agent learns a new task through natural interaction with a human instructor. Behavior Trees (BTs) offer a reactive, modular, and interpretable way of encoding task descriptions but have not yet been applied a lot in robotic ITL settings. Most existing approaches that learn a BT from human demonstrations require the user to specify each action step-by-step or do not allow for adapting a learned BT without the need to repeat the entire teaching process from scratch. Method: We propose a new framework to directly learn a BT from only a few human task demonstrations recorded as RGB-D video streams. We automatically extract continuous pre- and post-conditions for BT action nodes from visual features and use a Backchaining approach to build a reactive BT. In a user study on how non-experts provide and vary demonstrations, we identify three common failure cases of an BT learned from potentially imperfect initial human demonstrations. We offer a way to interactively resolve these failure cases by refining the existing BT through interaction with a user over a web-interface. Specifically, failure cases or unknown states are detected automatically during the execution of a learned BT and the initial BT is adjusted or extended according to the provided user input. Evaluation and results: We evaluate our approach on a robotic trash disposal task with 20 human participants and demonstrate that our method is capable of learning reactive BTs from only a few human demonstrations and interactively resolving possible failure cases at runtime.

Voice Acoustic Instability During Spontaneous Speech in Parkinson’s Disease

In people with Parkinson's disease (PwPD), both motor and cognitive deficits influence voice and other aspects of communication. PwPD demonstrate vocal instability, but acoustic declines over the course of speaking are not well characterized and the role of cognition on these declines is unknown. We examined voice acoustics related to speech motor instability by comparing the first and the last utterances within a speech task. Our objective was to determine if mild cognitive impairment (MCI) status was associated with different patterns of acoustic change during these tasks. Participants with PD (n=44) were enrolled at University of Massachusetts Chan Medical School and classified by gold-standard criteria as normal cognition (PD-NC) or mild cognitive impairment (PD-MCI). The speech was recorded during the Rainbow Passage and a picture description task (Cookie Theft). We calculated the difference between first and last utterances in fo mean and standardized semitones (STSD), cepstral peak prominence-smoothed (CPPS), and low to high ratio (LH). We used t-tests to compare the declines in acoustic parameters between the task types and between participants with PD-NC versus PD-MCI. Mean fo, fo variability (STSD) and CPPS declined from the first to the last utterance in both tasks, but there was no significant difference in these declines between the PD-NC and PD-MCI groups. Those with PD-MCI demonstrated lower fo variability on the whole in both tasks and lower CPPS in the picture description task, compared to those with PD-NC. Mean and STSD fo as well as CPPS may be sensitive to PD-MCI status in reading and spontaneous speech tasks. Speech motor instability can be observed in these voice acoustic parameters over brief speech tasks, but the degree of decline does not depend on cognitive status. These findings will inform the ongoing development of algorithms to monitor speech and cognitive function in PD.

Type Analysis of Speaking Performance Assessment Task in English for Nusantara Textbook

One of the key factors in the success of learning English is the use of textbooks. The English textbook entitled "English for Nusantara SMP/MTs Class VIII" is one of the textbooks used in many schools as a guide for teachers and students in teaching and learning English. Although English for Nusantara textbooks have been widely used, there is no research has focused on the analysis of speaking assessment tasks in this textbook. Therefore this study aims to analyze the task of assessing speaking skills in Nusantara English textbooks. This study use a qualitative approach. The subject of this research is Nusantara English textbooks used in the context of learning English in Indonesia. Data collected through document analysis will include task descriptions, types of assessment, and instructions given for each task. The collected data will be analyzed descriptively to identify the types of assessment included in the speaking tasks. The results of this study show that intensive assessment tasks are the most common in this book. However, this study also found a limited variety of activities in speaking assessment task. The implications of the results of this research are important in the development of textbooks and the improvement of learning speaking skills in the context of English in Indonesia.