Morphological Decomposition Research Articles

Influence of verb and noun bases on reading aloud derived nouns: evidence from children with good and poor reading skills

Several studies on children and adults with and without linguistic impairment have reported differences between verb and noun processing. The present study assessed whether noun and verb bases affect differently children’s reading of derived words. Thirty-six Italian good readers and 18 poor readers, all 4th or 5th graders, were asked to read aloud nouns derived from either a noun base (e.g., artista, artist) or a verb base (e.g., punizione, punishment). Word and base frequency affected latencies only for deverbal nouns, while an effect of word length emerged for denominal nouns and an inhibitory effect of suffix length was found for both types of stimuli. A high base frequency and a high whole-word frequency both led to higher levels of accuracy. Verb bases led to higher error rates than noun bases. Poor readers, although slower and less accurate than good readers, showed a pattern of results similar to that of typically developing readers. Data confirm that in 4th and 5th graders morphological decomposition may affect reading aloud of long complex words, and that the grammatical class of the base can modulate this effect.

A Feature Extraction Method of Gear Fault Based on the SVD EMD and Morphology

In order to extract the weak signal from strong background signal characteristics, a feature extraction method combined of the singular value decomposition (SVD), empirical mode decomposition (EMD) and mathematical morphology was proposed. The signal got through the singular value decomposition first. Next took the average value of the decomposed main components. And carried on the empirical mode decomposition and selected the main component to summate and refactor. Then morphological difference filter was used to extract the frequency characteristics of the fault signal. The results of numerical simulation test and gear fault simulation experiments show that the proposed method can clearly extract the frequency characteristics of weak signal from strong background signal and noise. Comparison has been done with the results of singular value decomposition (SVD) and morphological filtering method and empirical mode decomposition form of filtering method. It proves the effectiveness of the proposed method.

L1 AND L2 WORD RECOGNITION IN FINNISH

This study investigated the effect of the first language (L1) on the visual word recognition of inflected nouns in second language (L2) Finnish by native Russian and Chinese speakers. Case inflection is common in Russian and in Finnish but nonexistent in Chinese. Several models have been posited to describe L2 morphological processing. The unified competition model (UCM; MacWhinney, 2005) predicts L1-L2 transfer, whereas processability theory (Pienemann, 1998) posits a universal hierarchy in L2 acquisition regardless of the L1. The morphological decomposition deficiency hypothesis (Ullman, 2001b; VanPatten, 2004) claims that nonnatives cannot morphologically decompose words. Finally, DeKeyser (2005) proposes that morphophonological transparency affects nonnative processing. The current study explores which model best accounts for the processing of L2 Finnish by native Russian and Chinese speakers. The materials included simple nouns, transparently inflected nouns, and semitransparently inflected nouns. The results showed that Finns and Russians had longer reaction times (RTs) for morphologically complex nouns, but Chinese had longer RTs for semitransparent nouns. The RT results support the UCM by showing a L1-L2 transfer. Furthermore, transparency influenced word recognition among nonnatives; they made the most errors with semitransparent nouns.

Morphological decomposition and the transposed-letter (TL) position effect

When a nonword is created from a real word by transposing two medial letters (e.g., oeby from obey), the former is confused with the latter more than when the transposition involves the initial letter (e.g., boey). This is called the “transposed-letter (TL) Position” effect. It is shown here that the addition of a prefix eliminates the TL Position effect (i.e., disboey shows as much interference as disoeby relative to a nontransposed control). The TL Position effect also disappears if the prefix creates a nonword when added to the stem (e.g., reboey shows as much interference as reoeby), but there is no interference at all when a nonprefix is added to the stem instead (e.g., raboey or raoeby). The fact that there is strong TL interference for prefixed nonwords (e.g., reboey and reoeby) strongly points to unavoidable morphological decomposition. The disappearance of the TL Position effect when a prefix is added to the stem is ascribed to the reduction in perceptual salience for the initial letter, and this is confirmed when the TL Position was also shown to disappear when the prefix was replaced by digits. The results of the five experiments lead to a consideration of the way in which models of orthographic processing might handle both TL interference and morphological decomposition.

Morphological decomposition and lexical access in a native and second language: A nesting doll effect

This study tests the hypothesis that second language (L2) learners move developmentally from decomposition of inflected words with less complex stem allomorphy to words with more complex stem allomorphy in lexical access. Robust auditory priming effects were observed for three types of Russian high- and low-frequency inflected verbs with graded regularity treated as complexity and productivity in stem allomorphy in native speakers (NS) of Russian. Similar effects were found in American L2 learners of Russian at three proficiency levels for high-frequency verbs. Low-frequency verbs showed an interaction of the degree of regularity with proficiency level, with priming effects present for regular verbs at all three proficiency levels, semi-regular verbs at two higher levels, and irregular verbs only at the highest level. Lemma and surface frequency effects were observed in irregular verbs only in L2 learners. Results suggest that in Russian, a highly inflected language, auditory lexical access occurs in two stages: first, decomposition into stem and inflectional affix, and second, access of the stem representation at the lemma level, which can occur directly or by further decomposing the stem into root and suffix (the nesting doll pattern). The first stage takes place automatically both in NS and L2 learners for all productive inflections, while the second is gradually acquired by late learners, from productive and less complex to unproductive and more complex stem allomorphy. This developmental tendency is in contrast to the claims that late second language learners store and access regularly inflected words undecomposed and supports the view that L2 learners beyond low proficiency levels process inflectional morphology.

The role of form in morphological priming: Evidence from bilinguals

This article explores how bilinguals perform automatic morphological decomposition processes, focusing on within- and cross-language masked morphological priming effects. In Experiment 1, unbalanced Spanish (L1)–English (L2) bilingual participants completed a lexical decision task on English targets that could be preceded by morphologically related or unrelated derived masked English and Spanish prime words. The cognate status of the masked Spanish primes was manipulated, in order to explore to what extent form overlap mediates cross-language morphological priming. In Experiment 2, a group of balanced native Basque–Spanish speakers completed a lexical decision task on Spanish targets preceded by morphologically related or unrelated Basque or Spanish masked primes. In this experiment, a large number of items were tested and the cognate status was manipulated according to a continuous measure of orthographic overlap, allowing for a fine-grained analysis of the role of form overlap in cross-language morphological priming. Results demonstrated the existence of between-language masked morphological priming, which was exclusively found for cognate prime-target pairs. Furthermore, the results from balanced and unbalanced bilinguals were highly similar showing that proficiency in the two languages at test does not seem to modulate the pattern of data. These results are correctly accounted for by mechanisms of early morpho-orthographic decomposition that do not necessarily imply an automatic translation of the prime. In contrast, other competing accounts that are based on translation processes do not seem to capture the present results.

Fault Feature Extraction of High-Speed Automaton Based on Motion Morphology Decomposition

In order to obtain the characteristic parameters reflecting fault state of high-speed automaton (HSA), the fault feature extraction method based on motion morphology decomposition and wavelet packet transform (WPT) was presented. According to the movement law of the automaton, the vibration signal generated in three bursts of fire was decomposed into three separate signals, then the response signal in each shooting is a separate signal. Then using WPT to respectively extract wavelet packet energy from three separate signals as the fault characteristic parameters of HSA. By the example, the results show that the extracted fault features can well reflect the working conditions of automaton. Thus the proposed method could be used to extract the fault feature of automaton for monitoring the condition and diagnosing the fault of automaton.

Object Recognition using Disk based Morphological Shape Decomposition Features

The ability of object recognition system is to recognize a large number of objects constrained by a variety of factors such as the selection of a feature extraction method, quality of the images, and the classification models. This paper presents an approach to the recognition of complex shape objects using shape representation features. The shape representation features are the disk components which are calculated from morphological shape decomposition technique. The disk components of the shapes are generated using disk component generation Algorithm. These disk components are more primitive and easily matched with other disk components that are from another shape. These features are tested using the Quadratic classifier on different shapes. It is observed that the classifier gives good accuracy.

Iterative morphological binary decomposition frames interpolation technique

Iterative morphological binary decomposition (IMBD) is based on first calculating the binary morphological skeleton or binary morphological shape decomposition using digital squares or digital crosses (structuring elements) and then reiterating the method using digital lines. The modified and extended binary digital frames interpolation technique is based on changing some of the IMBD subsets of input frame with the same number of the IMBD subsets of output frame. Simulations results are provided.

A New Intelligent Fault Recognition Method for Gearbox

In this paper, a novel method to recognize gear fault pattern was approached based on multi-scale morphological undecimated wavelet decomposition, sample entropy and grey incidence. Firstly, multi-scale morphological undecimated wavelet decomposition was developed based on the characteristic of impulse feature extraction in difference morphological filter. And it was used to process different gear fault signals in five levels. Secondly, the sample entropy of each level was calculated. Finally, the sample entropy was served as the feature vectors and the grey incidence of different gear vibration signals was calculated to identify the fault pattern and condition. Practical example shows the efficiency of the proposed recognition method. It is suitable for on-line monitoring and fault diagnosis of gear.

MEG masked priming evidence for form-based decomposition of irregular verbs

To what extent does morphological structure play a role in early processing of visually presented English past tense verbs? Previous masked priming studies have demonstrated effects of obligatory form-based decomposition for genuinely affixed words (teacher-TEACH) and pseudo-affixed words (corner-CORN), but not for orthographic controls (brothel-BROTH). Additionally, MEG single word reading studies have demonstrated that the transition probability from stem to affix (in genuinely affixed words) modulates an early evoked response known as the M170; parallel findings have been shown for the transition probability from stem to pseudo-affix (in pseudo-affixed words). Here, utilizing the M170 as a neural index of visual form-based morphological decomposition, we ask whether the M170 demonstrates masked morphological priming effects for irregular past tense verbs (following a previous study which obtained behavioral masked priming effects for irregulars). Dual mechanism theories of the English past tense predict a rule-based decomposition for regulars but not for irregulars, while certain single mechanism theories predict rule-based decomposition even for irregulars. MEG data was recorded for 16 subjects performing a visual masked priming lexical decision task. Using a functional region of interest (fROI) defined on the basis of repetition priming and regular morphological priming effects within the left fusiform and inferior temporal regions, we found that activity in this fROI was modulated by the masked priming manipulation for irregular verbs, during the time window of the M170. We also found effects of the scores generated by the learning model of Albright and Hayes (2003) on the degree of priming for irregular verbs. The results favor a single mechanism account of the English past tense, in which even irregulars are decomposed into stems and affixes prior to lexical access, as opposed to a dual mechanism model, in which irregulars are recognized as whole forms.

ECG Signal Pre-processing by Lifting Scheme Constructing Multi-resolution Morphological Decomposition

ECG Signal Pre-processing by Lifting Scheme Constructing Multi-resolution Morphological Decomposition

ERPs and morphological processing: the N400 and semantic composition

Both behavioral and electrophysiological evidence suggests that fluent readers decompose morphologically complex words into their constituent parts. Previous event-related potential (ERP) research has been equivocal with regard to whether the N400 component indexes morphological decomposition or the integration of the products of decomposition, a process called semantic composition. In a visual lexical decision task with college students, we recorded ERPs to a well-controlled set of words and nonwords made up of bound morphemes (discern, predict; disject, percern) or free morphemes (cobweb, earring; cobline, bobweb) and monomorphemic control words and nonwords (garlic, minnow; gartus, buzlic). For each of the three morphological types, participants were faster to respond to words than to nonwords. Furthermore, for each of the three morphological types, the amplitude of the N400 was more negative to nonwords than to matched words, an effect indicating that the N400 is more sensitive to the lexicality of the whole stimulus than to the meaningfulness of the constituent parts of the stimulus. The N400 lexicality effect was not significantly different across the three morphological types. To our knowledge, this is the first ERP study to directly compare the processing of printed sets of words composed of bound and free morphemes and monomorphemic control stimuli in order to explore the relative sensitivity of the N400 to morphological decomposition (i.e., the status of the parts) and semantic composition (i.e., the status of the whole). Our findings are consistent with an interpretation of the N400 as an index of a process of semantic composition.

¿Cómo se da la descomposición morfológica en el procesamiento léxico? Una cuestión de etapas

Las teorías que proponen una etapa de descomposición en morfemas en el camino desde la visualización de la palabra hasta el acceso a su entrada léxica correspondiente en el lexicón pueden adoptar dos posturas. Por un lado, las teorías supra-léxicas defienden que dicha etapa se presenta una vez conseguido el acceso al significado de la palabra. Las teorías pre-léxicas, por su parte, muestran la postura contraria, según la cual la descomposición morfémica se daría de modo automático, sin influjo alguno de la semántica. Otra opción es aquella adoptada por Schreuder y Baayen (1995) y Longtin y Meunier (2007), quienes proponen un doble influjo de los factores morfológicos, tanto en la etapa pre-léxica como en la supra-léxica. Sin embargo, ambos modelos difieren en el número de etapas que proponen (tres y dos respectivamente). Este trabajo pretende solucionar dicha oposición mediante el diseño de un experimento de decisión léxica en el que se presentaron palabras españolas (aceptación), pseudopalabras posibles (aceptamiento) y pseudopalabras imposibles (aceptaz). Los resultados parecen indicar hacia un favorecimiento de un modelo en dos etapas, en el caso de las palabras almacenadas en el lexicón, y en tres cuando se trata de estímulos novedosos.

Preserved morphological processing in semantic dementia

Individuals with semantic dementia (SD) show progressive worsening of lexical–conceptual single word knowledge alongside preservation of nonsemantic aspects of language. The current study examines morphological processing in SD. S.H.S., a Hebrew-speaking person with SD, completed eight tasks that tested inflection, linear derivation, and nonlinear derivation at three time points over six and a half years. S.H.S. correctly produced plural nouns and judged the grammaticality of noun–verb and noun–adjective inflectional agreement. A steeper decline was seen in the ability to judge irregularly versus regularly inflected forms. S.H.S. judged nationality suffixes accurately and produced diminutive and agentive suffixes successfully. He also relied on morphological decomposition when performing lexical decision. Judgement of sentences consisting of derived verbal participles was intact with regard to morphological considerations alone, but impaired when semantic considerations determined grammaticality. S.H.S.'s performance suggests that structural aspects of word knowledge might be functional in the face of a severe semantic impairment.

Effectiveness of coupling agents in the poly (methyl methacrylate)-modified starch/styrene-butadiene rubber interfaces

In this work, the mechanical performances of poly (methyl methacrylate)-modified starch/styrene-butadiene rubber (PMMA-modified starch/SBR) biocomposites treated by several coupling agents, i.e., bis(3-triethoxysilyl propyl)tetrasulfide (TESPT), 3-aminopropyl triethoxysilane (APTES), 3-Mercaptopropyl triethoxysilane (MPTES) and 4,4-methylene bis (phenyl isocyanate) (MDI), were investigated. The measurements of resulted biocomposites demonstrated that the biocomposites treated by MDI exhibited the superior reinforcement, in which the tensile strength and 300% modulus were increased up to 11.8 and 7.5MPa, respectively. The strong chemical bonding in the PMMA-modified starch/SBR interfaces resulted from MDI intermediated linkages was responsible for this observed reinforcement. Furthermore, the enhanced interfacial interaction was confirmed via the changes in the morphology, storage modulus, glass temperature and initial decomposition temperature.

Morphological operation-based bi-dimensional empirical mode decomposition for automatic background removal of fringe patterns

A modified bi-dimensional empirical mode decomposition (BEMD) method is proposed for sparsely decomposing a fringe pattern into two components, namely, a single intrinsic mode function (IMF) and a residue. The main idea of this method is a modified sifting process which employs morphological operations to detect ridges and troughs of the fringes, and uses weighted moving average algorithm to estimate envelopes of the IMF, replacing respective local extrema detection and envelope interpolation of conventional BEMDs. The background intensity of the fringe pattern is automatically removed by extracting the single IMF, thereby relieving the mode mixing problem of the BEMDs. A fast algorithm based on 2D convolution is also presented for reducing the calculation time to several seconds only. This approach is applied to process simulated and real fringe patterns, and the results obtained are compared with Fourier transform, discrete wavelet transform, and other EMD methods. The MATLAB code is downloadable at http://gr.xjtu.edu.cn/web/zhouxiang.

Construction of 3D hierarchical SnO2 microspheres from porous nanosheets towards NO decomposition

Three-dimensional (3D) hierarchical architectures are currently attracting worldwide interest owing to their fascinating morphology-dependent properties and potential applications. Herein we constructed SnO2 microspheres with 3D hierarchical flower-like architectures self-assembled with porous SnS2 nanosheets by a facile hydrothermal method with subsequent calcination. The chemical and physical properties as well as photocatalytic application of SnO2 microspheres were investigated. The size and morphology were examined with scanning electron microscopy and transmission electron microscopy. The phase and crystalline structure were determined with powder X-ray diffraction. The UV–Vis absorption property was determined with UV–Vis diffuse reflectance. The photocatalytic activities were evaluated with nitrogen monoxide (NO) decomposition under UV–Vis light irradiation. The effects of calcination temperature on morphology and NO decomposition were also studied.

Parallel Processing of Whole Words and Morphemes in Visual Word Recognition

Models of morphological processing make different predictions about whether morphologically complex written words are initially decomposed and recognized on the basis of their morphemic subunits or whether they can directly be accessed as whole words and at what point semantics begin to influence morphological processing. In this study, we used unprimed and masked primed lexical decision to compare truly suffixed (darkest) and pseudosuffixed words (glossary) with within-boundary (drakest/golssary) to across-boundary (darekst/glosasry) letter transpositions. Significant transposed-letter similarity effects were found independently of the morphological position of the letter transposition, demonstrating that, in English, morphologically complex whole-word representations can be directly accessed at initial word processing stages. In a third masked primed lexical decision experiment, the same materials were used in the context of stem target priming, and it was found that truly suffixed primes facilitate the recognition of their stem-target (darkest-DARK) to the same extent as pseudosuffixed primes (glossary-GLOSS), which is consistent with theories of early morpho-orthographic decomposition. Taken together, our findings provide evidence for both whole-word access and morphological decomposition at initial stages of visual word recognition and are discussed in the context of a hybrid account.

Classification of hyperspectral images by tensor modeling and additive morphological decomposition

Pixel-wise classification in high-dimensional multivariate images is investigated. The proposed method deals with the joint use of spectral and spatial information provided in hyperspectral images. Additive morphological decomposition (AMD) based on morphological operators is proposed. AMD defines a scale-space decomposition for multivariate images without any loss of information. AMD is modeled as a tensor structure and tensor principal components analysis is compared as dimensional reduction algorithm versus classic approach. Experimental comparison shows that the proposed algorithm can provide better performance for the pixel classification of hyperspectral image than many other well-known techniques.