Word Images Research Articles

Transparency and translucency indices for 1,525 pictograms from the Aragonese Portal of Augmentative and Alternative Communication.

This study investigated the transparency and translucency of 1,525 pictograms from the Aragonese Portal of Augmentative and Alternative Communication (ARASAAC). A total of 521 participants took part in tasks that involved providing the word that best described the meaning of a pictogram or rating the relationship between a pictogram and a verbal label. This process allowed us to obtain indices of transparency (the quality of pictograms that makes their meaning easily "guessable" in the absence of their referent) and translucency (the degree of perceived relationship between the pictogram and its referent when the latter is present) which were further analyzed to assess their reliability and comparability with similar studies. Additionally, the relationship of those indices with various visual and psycholinguistic characteristics was explored, particularly focusing on the match between the original ARASAAC pictogram names and the most frequently provided names by the participants (modal names) for the pictograms. Results showed relatively low levels of transparency, as well as high levels of translucency, with nouns displaying the highest values in both metrics. For transparency and translucency, word imageability and concreteness were the most correlated factors, and, together with age of acquisition, they were the most important features related to the name matching with ARASAAC. The norms derived from this study enhance our understanding of pictogram perception, empowering stakeholders to leverage data-driven insights for the development and implementation of pictographic systems, thus improving cognitive accessibility.

The Roles of Rule Type and Word Term in the Deductive Reasoning of Adults with and without Dyslexia.

Despite its importance to everyday functioning, reasoning is underexplored in developmental dyslexia. The current study investigated verbal deductive reasoning on the Wason selection task, not previously used in dyslexia research despite its well-established pedigree. Reasoning rule was manipulated, with the conditional rules varying in the logical values presented. The word frequency and imageability of the word terms was also manipulated. Twenty-six adults with dyslexia and 31 adults without dyslexia completed Wason selection task problems. No group difference in reasoning accuracy or completion time was found. However, the participants were most accurate when reasoning with the rule type "If p, then not q" and least accurate with the rule type "If p then q". More trials were also answered correctly when the word terms were highly imageable but of average word frequency. These findings are in line with the general reasoning literature. Dyslexia status did not interact with either rule type or word term type. The study expands upon previous research by testing verbal deductive reasoning in dyslexia, highlighting the role of imageability in facilitating reasoning performance for all, regardless of the presence or absence of dyslexia. Implications for the design of educational materials are considered.

Body-object interaction ratings for 3600 French nouns.

Body-object interaction (BOI) measures the ease with which the human body can interact with the concept represented by a word. This research focuses on two main objectives: first, to establish French norms for the psycholinguistic variable BOI, and second, to investigate the contribution of BOI to language processing in French. We collected BOI ratings for 3600 French nouns from participants through an online platform. The inter- and intrastudy reliability of these new ratings indicate that the ratings are robust. We then aimed to determine the role of BOI in word recognition. A hierarchical regression analysis was conducted using lexical decision reaction times (RTs) as the dependent variable. BOI was found to be a significant predictor of lexical decision latencies, beyond the contribution of word length, frequency, orthographic distinctiveness, and imageability. Contrary to previous findings in English, higher BOI values were associated with longer RTs in French, indicating an inhibitory effect of BOI on French word processing. Methodological differences may account for this divergent result. Taken together, the results of this study show the independent contribution of BOI to word recognition in French. This supports the notion that sensorimotor information is a crucial component of language processing. By providing a reliable and sizable BOI database for French nouns, we offer a valuable resource for psycholinguistic and language processing research. This research underscores the complex relationship between language, cognition, and sensorimotor experiences, advancing our comprehension of language processing mechanisms.

Optimizing Convolutional Neural Networks for Image Classification on Resource-Constrained Microcontroller Units

Running machine learning algorithms for image classification locally on small, cheap, and low-power microcontroller units (MCUs) has advantages in terms of bandwidth, inference time, energy, reliability, and privacy for different applications. Therefore, TinyML focuses on deploying neural networks on MCUs with random access memory sizes between 2 KB and 512 KB and read-only memory storage capacities between 32 KB and 2 MB. Models designed for high-end devices are usually ported to MCUs using model scaling factors provided by the model architecture’s designers. However, our analysis shows that this naive approach of substantially scaling down convolutional neural networks (CNNs) for image classification using such default scaling factors results in suboptimal performance. Consequently, in this paper we present a systematic strategy for efficiently scaling down CNN model architectures to run on MCUs. Moreover, we present our CNN Analyzer, a dashboard-based tool for determining optimal CNN model architecture scaling factors for the downscaling strategy by gaining layer-wise insights into the model architecture scaling factors that drive model size, peak memory, and inference time. Using our strategy, we were able to introduce additional new model architecture scaling factors for MobileNet v1, MobileNet v2, MobileNet v3, and ShuffleNet v2 and to optimize these model architectures. Our best model variation outperforms the MobileNet v1 version provided in the MLPerf Tiny Benchmark on the Visual Wake Words image classification task, reducing the model size by 20.5% while increasing the accuracy by 4.0%.

Optical Character Recognition of Balochi Script

Optical Character Recognition is considered one of the fastest methods of data entry. OCR converts the text image representation of x and y coordinates representing pixel information to be converted into text data in a particular language. OCR as a field of pattern recognition and document image understanding, OCR requires a challenging job once a different language text is available on the image. Difference in language script will pose different challenges for OCR which requires entirely different approaches and algorithms. Latin scripts require a different approach whereas the Balochi adopted language scripts require a different approach. In this regard, various solutions have been proposed for different languages. Segmentation is considered one of the important tasks in the process of OCR. A good segmentation will definitely increase the accuracy of an OCR. Segmentation includes the segmentation of text lines from text images which are further divided into words. These segmented words are further divided into characters which are to be recognized. A single segmentation algorithm to segment various scripts of the languages is proposed in this study which checks the script and then segments the text image for the further processing in OCR. The proposed generalized algorithm will check the style, direction and other properties of the script and then adopts the segmentation process to segment text lines, words and characters of the language. The proposed algorithm segments more than ten languages of three scripts and segments for their OCRs. These images can be further processed for feature extraction and classification further. The process of OCR for selected languages will be made easier to recognize. Multiple scripts, languages and images were experimented, and the proposed algorithm successfully segmented 42,833 images of text line, words and character image. The algorithm provides 97% accuracy while segmenting these images and can be extended to further languages as well as scripts .

AENet: Image Retrieval of Kazakh Handwritten Documents Based on Attention Mechanism and Feature Aggregation

Kazakh is one of the multilingual languages of China and is widely spoken in some areas of Xinjiang, China. However, due to the fact that Kazakh is a language in which several characters are glued together to form a continuous word with a unique shape and complex structural combinations of relationships. This paper explores a solution for offline image retrieval of handwritten Kazakh words, which is a challenging task because, due to the lack of relevant datasets and the special writing morphology of the Kazakh language, traditional text image retrieval algorithms often struggle to achieve satisfactory results when dealing with writing styles that are varied and adherent to the language. Therefore, a dataset of offline Kazakh handwritten document images was created in this paper. The dataset contains 300 pages of document images with 20 500 words. Then, a new model called the ‘AENet’ is proposed. The model utilizes an attention mechanism to focus more finely on focal regions such as centers, inflection points, and contours of handwritten word images and to capture important local features from different scales. Fusion space pyramid pooling, feature aggregation, encoding operations, and feature downscaling and reconstruction are used to extract and reconstruct more representative feature representations from local to global to capture the overall information in the word images. Through experimental evaluation on Kazak‐80, Zilla‐64, and HWDB1.1‐375 datasets, it is verified that the method significantly improves the mAP for image retrieval of handwritten words, which is especially applicable to adhesive languages like Kazakh. © 2024 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.

Kemampuan Komunikasi Matematis Siswa SMP ditinjau dari Gaya Belajar

Taking into account auditory, visual, and kinesthetic learning modalities, this research animed to evaluate students mathematical communication abilities on the subject of Two Variable Linear Equation Systems. This qualitative study surveyed six junior high school students from the city of Sukabumi. The tools used for this study were a written exam to evaluate mathematical communication abilities and a learning style questionnaire. Mathematical communication skills are the basis of the learning style questionnaire. These skills include the ability to draw connections between mathematical ideas and real word objects, images, and diagrams to explain mathematical concepts, situations, and relationships in writing and to express events or ideas using mathematical symbols or language. Communication abilities are excellent among students with low visual learning styles, as are those with high visual learning styles, according to the study’s findings. The communication abilities of pupils whose learning style is mostly auditory are good, while those whose learning style is predominantly visual are poor. The same holds true for students communication abilities, those who learn best via movement tend to have excellent ones, while those who learn best by sitting still tend to struggle. Knowing one’s own learning style is crucial for pupils to enhance their mathematics communication abilities

Neural models for semantic analysis of handwritten document images

Semantic analysis of handwritten document images offers a wide range of practical application scenarios. A sequential combination of handwritten text recognition (HTR) and a task-specific natural language processing system offers an intuitive solution in this domain. However, this HTR-based approach suffers from the problem of error propagation. An HTR-free model, which avoids explicit text recognition and solves the task end-to-end, tackles this problem, but often produces poor results. A possible reason for this is that it does not incorporate largely pre-trained semantic word embeddings, which turn out to be one of the most powerful advantages in the textual domain. In this work, we propose an HTR-based and an HTR-free model and compare them on a variety of segmentation-based handwritten document image benchmarks including semantic word spotting, named entity recognition, and question answering. Furthermore, we propose a cross-modal knowledge distillation approach to integrate semantic knowledge from textually pre-trained word embeddings into HTR-free models. In a series of experiments, we investigate optimization strategies for robust semantic word image representation. We show that the incorporation of semantic knowledge is beneficial for HTR-free approaches in achieving state-of-the-art results on a variety of benchmarks.

Algorithms of interpretation of poly-timber vocal and instrumental ensembles in the concertmaster class

Statement of the problem. A significant part of the repertoire in the concertmaster class is chamber vocal works that engage the synthesis of Word and Music, that is, the inseparable trinity of the poetic text, its vocal embodiment, and instrumental accompaniment, which calls for the need for accurate correlation of the efforts of the singer and concertmaster, the performers’ synergy. And although chamber vocal works mostly have purely piano accompaniment, in some cases a part of a string or wind instrument is added to it, and then the mentioned trinity of musical and poetic synthesis is complicated by the appearance of a new, “performance” trinity unit of vocal and two instrumental parts, which requires especially high-quality ensemble, and therefore a special study of algorithms for its achievement, because the teacher from the concertmaster class has to be a kind of director in coordinating all the layers of the musical texture of such a poly-timbre ensemble. Objectives, methods, and novelty of the research. The purpose of this article is to designate a certain algorithm for the interpretation of a chamber vocal-instrumental ensemble in the concertmaster class as a poly-timbre structure represented by the performance trinity of vocal and instrumental parts. Until now, the interpretative issues of a musical and poetic work have not yet been thoroughly considered in scientific publications under such angle. Imagery and dramatic, intonation, timbre and textural, structural and functional, and performance analysis of the musical and poetic text of a number of chamber vocal compositions of different times and styles (by J. Brahms, V. Barvynsky, V. Sylvestrov), carried out in the study, reveals the specifics of the formimg the ensemble vocal and instrumental texture of works as an artistic integrity. Research results and conclusion. Given the specifics of vocal and instrumental compositions with the involvement of an additional string or wind instrument in the piano accompaniment, we state the advisability of using a certain algorithm for learning them in the concertmaster class. First, it is an in-depth analysis of the poetic original source of the work, if necessary, its complete translation from a foreign language, in order to reveal its leading idea and circle of images. Secondly, it is necessary to determine the main factors influencing the style and quality of the performance of the piece, namely: – the type of relationship between Words and Music, as well as the syntactic features of the verbal text, which determine the character of vocal and instrumental intonation; – genre orientation of the composition; – the functions of the instruments in the ensemble, according to the figurative dramaturgy and structure of the composition, both purely textural (solo, accompaniment, subvoice), and figurative-semantic (“Author’s voice”, “adding”, “finishing” of the vocal part, “leit-timbre”, etc.). Therefore, the poly-timbre quality of the accompaniment of the vocal part becomes an important “starting point” of the interpretive work on the vocal piece in the class of concertmaster skill. The performance “trinity unit” obliges to a particularly coordinated ensemble, in particular, between the instrumental parts, when they together realize the artistic idea of the vocal work as the embodiment of the “correlation of the author’s systems (Poet – Composer)” (Buchok, 2022: 44). Regarding the latter, according to our observation, the considered type of chamber vocal instrumental ensemble is connected with the involvement of poetic sources with special semantics – those that appeal to images and ideas of a “higher plane”, are aimed at spiritual contemplation, or convey personal “movements of the soul”, the adequate perception of which is impossible outside the context (in the examined compositions by J. Brahms, V. Barvinsky and V. Sylvestrov – spiritual and religious ones), which implies the presence of poly-semantic figurative structures in such works. So, the use by composers of the timbre of an additional solo instrument in chamber vocal compositions can be connected with the artistic embodiment of a “higher principle”: the inclusion of a segment of “pure” (instrumental) music best conveys the sublimity and spiritual pathos of the images of the poetic Word, reveals its semantic connotations, “personifying” the higher ideological level of the work or the actual “voice” of the Author (Poet – Composer).

Co-teaching literacy strategies for the inclusion of second-language learners: possibilities for professional development

This narrative study, set in lower-secondary school (grades 7–9) in Finland, explores a co-teaching initiative aimed at developing inclusive education for second-language students by integrating literacy strategies in content-area teaching. The co-teaching was realized by three experienced content-area teachers working in dyads with a resource teacher with competence in literacy strategies. Co-teaching among experienced teachers at higher grade-levels and with the purpose of including second-language students is a relatively under-researched field. Data was collected through seven teacher–researcher conversations over two school-years. Word images were constructed from the field texts to represent four narrative themes, depicting the teachers’ narration of their knowledge construction and role formation: daring to share, extended expertise, fine-tuning roles and increased well-being. Results suggest benefits such as well-being through shared responsibilities along with increased expertise to support all students’ learning of both language and content in the co-taught classroom and beyond. Trust, being open to change and daring to share one’s knowledge landscapes and classroom space were seen as crucial in the negotiation of educational priorities and teacher roles. Co-teaching is suggested to be a viable option for the professional development of inclusive practices in the form of integration of literacy strategies across the curriculum.

Memory enhancement for emotional words is attributed to both valence and arousal

We do not memorize items in our surroundings with equal priority. Previous literature has widely shown that emotional stimuli are better remembered than neutral stimuli. However, given emotional stimuli and neutral stimuli often differ in both valence and arousal dimensions, it remains unclear whether the enhancement effects can be attributed to valence, or just to arousal. Importantly, most prior studies relied on a relatively small number of stimuli and non-emotional factors such as word length, imageability and other confounds were hard to control. To address these challenges, we analyzed multiple large databases of recognition memory and free recall tasks from previous research by items with many lexical and semantic covariates included, examining the effects of valence or arousal when controlling for each other. Our results showed a U-shaped relationship between valence and memory performance for both recognition and free recall, and a linear relationship between arousal and memory performance for both tasks. These findings showed that the memory enhancement effects can be attributed to both valence and arousal. We demonstrated these effects with generalizability across many stimuli and controlled for non-emotional factors. Together, these findings disentangle the contribution of valence and arousal in emotional memory enhancement effects and provide insights for current major theories of emotional memory.

Efforts to Improve Critical Thinking Skills Through a Whole Language Approach in Elementary Schools

This research was conducted with the aim of describing: (1) learning in material conveying fairy tale content developed using a whole language approach for elementary school (SD) students (2) changes in elementary school students' attitudes after receiving material conveying fairy tale content using a whole language approach, and (3) elementary school students' learning achievement after receiving material to convey the contents of fairy tales using a whole language approach. This research includes case study research. The subjects of this research were second grade elementary school students. The data collection techniques used were observation, interviews and tests. Test the validity of the data in this research by triangulating data sources, triangulating methods, and informant reviews. The results of the research show (1) the application of the whole language approach in conveying fairy tale content is carried out through the enjoyment stage, appreciation stage, and understanding stage, (2) the attitude changes that occur include (a) increasing critical thinking skills, (b) understanding the correlation between images in words, c) activeness when watching videos and discussing, (d) embedding cooperation and respect between students, and (e) increasing speaking skills, (3) all students experienced an increase in learning achievement with a score of 77-95 through three domains, namely cognitive domain, affective domain, and psychomotor domain.

Weber's Law-based Regularization for Blind Image Deblurring

Blind image deblurring aims to recover an output latent image and a blur kernel from a given blurred image. Kernel estimation is a significant step in blind image deblurring and requires a regularization technique to minimize the cost function and the edges of objects to generate a sharp image in a better way. This study proposes a new image regularization technique called Weber's Law Regularization (WLR) based on the Weber law phenomenon. The Weber ratio was used to preserve the edges of small salient objects and to minimize the cost function to obtain a sharp image while minimizing the ringing effect. To validate the WLR, experiments were conducted on benchmark synthetic and real word images and compared with existing state-of-the-art methods. The experimental results showed that WLR can effectively and efficiently deblur images even in the absence of prior knowledge.

Noun imageability and the processing of sensory-based information.

The aim of this study was to test whether the availability of internal imagery elicited by words is related to ratings of word imageability. Participants are presented with target words and, after a delay allowing for processing of the word, answer questions regarding the size or weight of the word referents. Target words differ with respect to imageability. Results show faster responses to questions for high imageability words than for low imageability words. The type of question (size/weight) modulates reaction times suggesting a dominance of the visual domain over the physical-experience domain in concept representation. Results hold across two different languages (Czech/German). These findings provide further insights into the representations underlying word meaning and the role of word imageability in language acquisition and processing.

Siamese-based offline word level writer identification in a reduced subspace

In this paper, we explore the notion of similarity learning by utilizing the Siamese neural network employing the residual framework for the purpose of writer identification based on offline handwritten input word images. Apart from being text-independent, our method does not impose any limitations on the number of characters of the input word image being employed also it can be used in real-world applications where input image patches with a few letters exist. The novelty in our proposal is in the exploration of a sparse-based model for representing the output feature vector of the Siamese network in a reduced dimensional space. We also formulate a divergence-based approach for assigning a saliency score to each component in the sparse representation based on their discriminatory power. The system efficacy has been demonstrated on well-known word-level databases and the results obtained are promising when compared with previous works.

Toward Innovative Recognition of Handwritten Arabic Characters: A Hybrid Approach with SIFT, BoVW, and SVM classification

The goal of handwriting recognition has been a top priority for those who want to enter data into computer systems for more than thirty years. In several fields, the advent of handwriting recognition technology is highly anticipated. OCR technology has made it possible for computers to recognize characters as visual objects and collect data about their unique characteristics in recent years. In particular, several studies in this field have focused on Arabic writing. The use of machines to examine handwritten papers is the first step in the character identification process. The identification of specific Arabic characters is the main goal of this particular investigation. In computer vision, Arabic character recognition is very important since it's necessary to correctly recognize and classify Arabic letters and characters in manuscripts. In this research, an innovative approach based on identifying Arabic character characteristics using BoVW (bag of visual words) and SIFT (Scale Invariant Feature Transform) features is proposed. These features are clustered using k-means clustering to produce a dictionary. Following that, SVM (Support Vector Machine) is utilized to classify the word images in a visual codebook created using these terms. The proposed approach is an innovative method to deal with the difficulties associated with Arabic hand-writing recognition. The utilization of BoVW and SIFT features is expected to enhance the system's robustness in recognizing and classifying Arabic characters. The proposed approach will be experimentally evaluated using a dataset that includes a variety of Arabic characters written in various styles. The results of this study will offer important new perspectives on the effectiveness and practicality of the approach suggested

Generalized Segmentation Algorithm for Dissimilar Script Languages

Optical Character Recognition is considered one of the fastest methods of data entry. OCR converts the text image representation of x and y coordinates representing pixel information to be converted into text data in a particular language. OCR as a field of pattern recognition and document image understanding, OCR requires a challenging job once a different language text is available on the image. Difference in language script will pose different challenges for OCR which requires entirely different approaches and algorithms. Latin scripts require a different approach whereas the Arabic adopted language scripts require a different approach. In this regard, various solutions have been proposed for different languages. Segmentation is considered one of the important tasks in the process of OCR. A good segmentation will definitely increase the accuracy of an OCR. Segmentation includes the segmentation of text lines from text images which are further divided into words. These segmented words are further divided into characters which are to be recognized. A single segmentation algorithm to segment various scripts of the languages is proposed in this study which checks the script and then segments the text image for the further processing in OCR. The proposed generalized algorithm will check the style, direction and other properties of the script and then adopts the segmentation process to segment text lines, words and characters of the language. The proposed algorithm segments more than ten languages of three scripts and segments for their OCRs. These images can be further processed for feature extraction and classification further. The process of OCR for selected languages will be made easier to recognize. Multiple scripts, languages and images were experimented, and the proposed algorithm successfully segmented 32,833 images of text line, words and character image. The algorithm provides 97% accuracy while segmenting these images and can be extended to further languages as well as scripts.

Automatic Topic-Based Web Page Classification Using Deep Learning

The internet is frequently surfed by people by using smartphones, laptops, or computers in order to search information online in the web. The increase of information in the web has made the web pages grow day by day. The automatic topic-based web page classification is used to manage the excessive amount of web pages by classifying them to different categories based on the web page content. Different machine learning algorithms have been employed as web page classifiers to categorise the web pages. However, there is lack of study that review classification of web pages using deep learning. In this study, the automatic topic-based classification of web pages utilising deep learning that has been proposed by many key researchers are reviewed. The relevant research papers are selected from reputable research databases. The review process looked at the dataset, features, algorithm, pre-processing used in classification of web pages, document representation technique and performance of the web page classification model. The document representation technique used to represent the web page features is an important aspect in the classification of web pages as it affects the performance of the web page classification model. The integral web page feature is the textual content. Based on the review, it was found that the image based web page classification showed higher performance compared to the text based web page classification. Due to lack of matrix representation that can effectively handle long web page text content, a new document representation technique which is word cloud image can be used to visualize the words that have been extracted from the text content web page.

The image of Russia in the Minds of Modern Youth Through the Perception of the Stimulus Words "Globalization", "Culture" (Based on the Materials of an Associative Experiment)

The article is a comprehensive analysis of the image of Russia, based on theoretical research and the results of an associative experiment conducted with two groups of subjects: the main (17-25 years old) and control (26-50years old). The aim is to study the perception of the image of Russia in the context of modern globalization and the change of the socio-cultural paradigm. The object of study is representatives of the modern generation of Russianyouth. The subject of the study is the image of Russia formed in the minds of modern youth under the influence of external and internal factors. The relevance of the research is due to the need to study the relationship in the triad of man-culture-globalization, in which an active youth stratum occupies an important place, acting as an indicator of public mood and a determinant of cultural values. The novelty of the research lies in considering the image of Russia through the prism of perception of the words-stimuli "globalization", "culture" according to the results of an associative experiment carried out using two methods – a free associative experiment and a directed associative experiment. The analysis of the obtained data is carried out within the framework of an interdisciplinary approach: psycholinguistic, linguophilosophical and sociological. The method of hermeneutical and comparative analysis of the words image and εἷδoς is used. The question is considered: what dominates in the linguistic consciousness of representatives of modern youth regarding the image of Russia: "image" as an artificial construct imposed from the outside, or constructed by the efforts of the mind and spirit of "eidos"? According to the results of the study, a number of associative reactions to stimulus words were obtained, showing that the image of Russia in the representations of the main age group is formed mainly under the influence of external factors caused by the modern media space (Internet, social networks), and not as a result of internal factors (the work of will, spirit and mind).

Multivariate word properties in fluency tasks reveal markers of Alzheimer's dementia.

Verbal fluency tasks are common in Alzheimer's disease (AD) assessments. Yet, standard valid response counts fail to reveal disease-specific semantic memory patterns. Here, we leveraged automated word-property analysis to capture neurocognitive markers of AD vis-à-vis behavioral variant frontotemporal dementia (bvFTD). Patients and healthy controls completed two fluency tasks. We counted valid responses and computed each word's frequency, granularity, neighborhood, length, familiarity, and imageability. These features were used for group-level discrimination, patient-level identification, and correlations with executive and neural (magnetic resonanance imaging [MRI], functional MRI [fMRI], electroencephalography [EEG]) patterns. Valid responses revealed deficits in both disorders. Conversely, frequency, granularity, and neighborhood yielded robust group- and subject-level discrimination only in AD, also predicting executive outcomes. Disease-specific cortical thickness patterns were predicted by frequency in both disorders. Default-mode and salience network hypoconnectivity, and EEG beta hypoconnectivity, were predicted by frequency and granularity only in AD. Word-property analysis of fluency can boost AD characterization and diagnosis. We report novel word-property analyses of verbal fluency in AD and bvFTD. Standard valid response counts captured deficits and brain patterns in both groups. Specific word properties (e.g., frequency, granularity) were altered only in AD. Such properties predicted cognitive and neural (MRI, fMRI, EEG) patterns in AD. Word-property analysis of fluency can boost AD characterization and diagnosis.