Morphological Processing in Lexical Access

Six lexical decision experiments were conducted to examine the influence of complex structure on the processing speed of English compounds. All experiments revealed that semantically transparent compounds (eg, rosebud) were processed more quickly than matched monomorphemic words (eg, giraffe). Opaque compounds (eg, hogwash) were also processed more quickly than monomorphemic words. However, when the experimental materials and/or procedure encouraged decomposition/integration, this advantage ...

A large body of morphological processing research refers to the role of semantic transparency in late visual word recognition. In the absence of semantic transparency effects, it has been claimed that all morphologically complex words are decomposed, irrespective of meaning (Smolka et al., 2009). Conversely, semantic transparency effects indicate late decomposition for semantically transparent but not for semantically opaque complex words (Rastle & Merkx, 2011). Furthermore, graded effects of semantic transparency have been reported (Feldman et al., 2004; Xu & Taft, 2015). Interestingly, semantic transparency effects in late morphological processing seem to correlate with cross-linguistic differences (Smolka et al., 2009), since they are typically detected in morphologically impoverished, concatenative languages (e.g., English) but not in morphologically rich, non-concatenative languages (e.g., Hebrew).
The present study examines the parameter of semantic transparency in derivational morphology of Modern Greek (MG), an Indo-European (IE) language that is described as both morphologically rich and concatenative (Ralli, 2005). Specifically, we deal with denominal suffixed verbs and adjectives, aiming to investigate whether semantic transparency affects late morphological processing and –if this is the case– whether semantic transparency effects are “all-or-none” or graded. To test these hypotheses, we conducted a visual lexical decision experiment, employing the overt priming paradigm with 250 ms prime-target stimulus of asynchrony (SOA), which is considered to tap into late intra-modal processing.
Forty eight university students participated in the experiment. Ninety morphologically related prime-target pairs, each comprising a (pseudo-)derived prime and a (putative) noun-base target, were equally distributed in three semantic transparency conditions (transparent, e.g., ψηφίζω /psi'fizo/ ‘to vote’ – ψήφος /'psifos/ ‘vote’; semitransparent, e.g., στολίζω /sto'lizo/ ‘to decorate’ – στολή /sto'li/ ‘costume’; opaque, e.g., θυμίζω /θi'mizo/ ‘to remind’ – θυμός /θi'mos/ ‘anger’), according to mean rating scores from three semantic relatedness pretests. Critical primes and targets were matched across conditions for several extraneous variables (e.g., length, lemma and word frequency, syllable and bigram frequency, prime-target formal similarity, orthographic neighborhood size, uniqueness point, morphological family size, grammatical category, inflectional and derivational suffixes). Critical stimuli along with 90 control primes (matched to critical primes but morphologically, orthographically and semantically unrelated to targets) were divided into two counterbalancing lists in a Latin square design, so that each participant was exposed to all priming conditions but saw each target only once. Moreover, each list included 180 word-nonword pairs and 90 filler word-word pairs, such that the overall prime-target relatedness proportion was reduced to 25% and target responses could not be predicted from prime-target formal similarity, derivational suffixes of primes, or total word-nonword proportion (50%).
Statistical analyses revealed no priming for the opaque condition. In contrast, significant priming was found for the transparent and semitransparent conditions (Figure 1), with the amount of priming being equal between them (Figure 2). Results suggest that late morphological processing in MG is semantically constrained, as is the case with other morphologically rich, concatenative IE languages (e.g., Polish, Serbian; but not German [Smolka et al., 2009]). Importantly, the observed priming effects posit whole-word processing for opaque (pseudo-)derivatives on the one hand and morpho-semantic decomposition for transparent and semitransparent derivatives on the other hand, without further gradation between transparent and semitransparent derivatives. Namely, morpho-semantic decomposition seems to be driven by a predictable part of meaning (Corbin, 1987/1991; Lieber, 2004), which brings together transparent and semitransparent derivatives. Therefore, the present findings favor a dichotomous rather than graded account of semantic transparency in late visual morphological processing.

The present study investigates semantic transparency effects in late visual word recognition, with regard to Modern Greek derivation. We conducted an overt priming lexical decision experiment, comparing morphologically related prime–target pairs in three semantic transparency conditions (i.e., opaque, semitransparent, transparent). Statistical analyses revealed significant priming of the same magnitude in the transparent and semitransparent condition, but no priming in the opaque condition. Results suggest that late morphological processing in Modern Greek derivation is semantically constrained, implying whole-word processing for opaque derivatives and pseudoderivatives, but morphosemantic decomposition for transparent and semitransparent derivatives. Moreover, the absence of any further gradation between the transparent and semitransparent condition indicates that morphosemantic decomposition is driven by a predictable part of meaning that brings together transparent and semitransparent derivatives. Therefore, we support a dual route account of late semantic transparency effects.

Dans le cadre de la problématique de l’apprentissage de la lecture par des enfants sourds, nous proposons d’étudier la nature des relations morphologiques lexicales entre la Langue des Signes Français (LSF) et le français écrit au travers d’une étude linguistique comparative de 62 familles morphologiques portant sur les différences et les similarités structurelles et distributionnelles des mots de ces deux langues.
La construction d’un tel corpus nécessite d’une part la sélection des familles morphologiques les plus représentatives du français (notamment en termes de fréquences d’usage et de familiarité pour les enfants de CP-CE1), l’analyse de leurs propriétés distributionnelles et structurelles (i. e. taille de la famille, fréquences cumulées, nature morphologique des membres et nature de leurs relations paradigmatiques) et d’autre part, l’analyse de la structure morphologique de leurs correspondants en LSF parmi lesquels on peut distinguer une part importante de mots composés mais également de mots que l’on peut considérer comme dérivés (e. g. Aronoff et al., 2005). L’hypothèse que nous formulons à l’égard de la comparaison des différentes caractéristiques de ces deux langues en contact chez l’apprenti lecteur sourd, est liée à la notion de conscience morphologique (selon Carlisle, 1995). Le corpus présenté se donne pour objectif de constituer une base de données pour l’étude de la conscience morphologique chez ces apprentis lecteurs particuliers qui ne disposent pas – ou dans une très faible proportion – d’accès à la conscience phonologique (e. g. Izzo, 2002).


Several studies showed a role for phonology in reading acquisition in deaf children (Transler, 2005) but few focused on the role of morphology and even less so when talking about sign language’s speakers, taught in sign language from the early age of school (Daigle & al., 2006). The structural characteristics of the sign language suggest however that phonology is highly related to morphology. More precisely, some signs systematically share together a common element and this can be attributed to morphological relationships (as suggested by Aronoff & al., 2005). The role of morphology in reading acquisition in hearing children is nowadays clearly established (see Colé et al., 2004), especially in an early age of acquisition (Carlisle, 2003) but still not in signed language. A thorough examination of French Sign Language (LSF) suggests however that some signs may correspond to morphological composition (Bauer, 2004).
Morphological awarness refers to ‘conscious awareness of the morphemic structure of words and their ability to reflect on and manipulate that structure’ (Carlisle, 1995, p. 194) and accordingly, we postulate that abstract morphemic units could stand as bases for making correspondences between signing and reading, facilitating reading acquisition in deaf children. In the present paper we aimed to study the nature of the morphological relationships between written French and French Sign Language (LSF) through a comparative linguistic analysis of morphological families. More precisely, we examined the distributional and structural similarities and differences of the words shared by these two languages. The construction of such a lexical database required on one hand the selection of the most representative morphological families (in terms of the frequency of use and their familiarity for second grade children), the analyses of their structural and distributional properties (i.e. family size, cumulated frequencies, morphological nature of the family members and of their paradigmatic relationships) and on the other hand, the morphological analysis of the morphological structure of the words of their corresponding words in LSF among which one can distinguish a majority of composed words (i.e. the combination of two lexically independent signs) and some words that can be considered as derived (i.e., a sign + a modifier, as Aronoff (2005) suggested).
We first selected the 100 most representative families of French in Grade 2 (CE1 in French) from EDUSCOL lexical database, proposed by the French Educational Ministry (http://eduscol.education.fr). Then we kept nouns, verbs and adjectives and rejected other grammatical forms. Then we used a second database (MANULEX: database created with scholar books, Lété et al., 2004) to extend our corpus to 62 morphological families, representing 1140 words (whose mean frequency= 330 occ./million). We spread the data according to their morphological structure that is derivative or compound. We observed that 79% of the 1140 words were derived (896 words) and 21% were compound (244 words). Among the 1140 words extracted in written, only 504 were related to LSF. Moreover, the ratio between derivatives and compounds showed that this distribution was more homogeneous in LSF than in written French: 75 % of the analyzed corpus we obtained a distribution composed with 47 % of derived signs and 53 % of compound signs. The distribution of Nfam was nevertheless different with a range of {4-22} and a mean of 11 members. Finally on a total of 1594 studied words (FR= 1140, LSF= 504), 47% shared morphological relationships.
In LSF, correspondences can therefore be established for the majority of the words belonging to the morphological families we selected. Accordingly, common linguistic units can be found between LSF and written French, offering the possibility of use of this overlaps during the process of reading acquisition in deaf children. The present constitutes a first step for studying the role of morphological awareness in these particular reading learners who do not, or for a very few part, have access to phonological awareness (see Izzo, 2002). The second step gives rise to the question: if deaf children do have morphological awareness of their mother tong and if so, are they able to perceive the relationships between signs and written words? As a matter of fact, morphological awareness has been studied in the general context of reading acquisition but need to be explored in various reading learners, from hearing children to deaf children in order to establish if morphological awareness has to be explicitly taught or not to those children.

Galls are plant tissue which is controlled by the insect. Galls are outgrowths on the surface of leaf caused by invasion of various organisms like Bacteria, Fungi, Insects, Parasites and Mites. Galls may be found on leaves, stems, twigs, branches, trunks and roots. There are hundreds of unique galls caused by insects and mites. They are formed on a variety of plants and in a broad range of sizes, shapes, colors and textures. Some galls are common and abundant and easily noticed. Others are rare or less conspicuous. This paper reports physicochemical study in leaf galls of Syzygium cumini. The parameters assayed were Total ash, acid insolubility and water solubility values and extractive values (Alcohol, water and petroleum ether).

Theories that give support to the existence of a morphological decomposition step on the path between the visual exposition to the verbal stimulus and the access to its entry in the lexicon can adopt two different positions. On the one hand, supra-lexical theories claim that such a step would follow the access to the meaning of the word. Pre-lexical theories, on the other hand, state that it would be previous to lexical access so that the decomposition would not take any semantic information into account. Another option is the one chosen by Schreuder and Baayen (1995) and Longtin and Meunier (2007), as they propose a double influence of morphological units at the pre- lexical stage as well as at the supra-lexical one. However, these two convergent theories show a central divergence, as the former posits a model with three processing stages in morphological processing, whereas the latter claims for a model with only two steps. The present paper aims at solving this opposition by proposing a lexical decision task where three types of critical stimuli were presented: Spanish words (aceptación), possible pseudo-words (aceptamiento) and impossible pseudo-words (aceptaz). Our results point to a model in two steps for words that are stored in the mental lexicon and in three steps for new, unknown, stimuli.

Our vocabulary is, at least in principle, infinite. We can create new words combining existing ones in meaningful ways to form new linguistic expressions. The present study investigated the morphological processing of novel compound words in overt speech production. Native speakers of Dutch learned a series of new compounds (e.g. appelgezicht, 'apple-face') that were later used as primes in a morphological priming task. In this protocol, primes were compound words morphologically related to a target's picture name (e.g. appelgezicht was used for a picture of an apple, Dutch appel). The novel primes were compared with corresponding familiar compounds sharing a free morpheme (e.g. appelmoes, 'applesauce') and with unrelated compounds. Participants were required to read aloud words and to name pictures in a long-lag design. Behavioral and event-related potentials (ERPs) data were collected in two sessions, separated by 48 h. Clear facilitation of picture naming latencies was obtained when pictures were paired with morphological related words. Notably, our results show that novel compounds have a stronger priming effect than familiar compounds in both sessions, which is expressed in a marked reduction in target naming latencies and a decrease in the N400 amplitude. These results suggest that participants focused more on the separate constituents when reading novel primes than in the case of existing compounds.

The authors compared sublexical and supralexical approaches to morphological processing with unam-biguous and ambiguous inflected words and words with ambiguous stems in 3 masked and unmasked priming experiments in Finnish. Experiment 1 showed equal facilitation for all prime types with a short 60-ms stimulus onset asynchrony (SOA) but significant facilitation for unambiguous words only with a long 300-ms SOA. Experiment 2 showed that all potential readings of ambiguous inflections were activated under a short SOA. Whereas the prime–target form overlap did not affect the results under a short SOA, it significantly modulated the results with a long SOA. Experiment 3 confirmed that the results from masked priming were modulated by the morphological structure of the words but not by the prime–target form overlap alone. The results support approaches in which early prelexical morphological processing is driven by morph-based segmentation and form is used to cue selection between 2 candidates only during later processing.

The paper explores the extent to which surface gender marking impacts on online gender processing and lexical access, aiming to determine the way grammatical gender is represented in the lexicon and accessed in real-time. Two online experiments were conducted: an online gender monitoring task and a lexical decision task. The effects of explicit gender marking and suffix type were examined by comparing adjusted and unadjusted loans as well as derived and inflected pseudo-nouns. The results revealed a significant gender marking effect in the gender monitoring task (Exp.1) but not in the lexical decision task (Exp.2), while a suffix type effect was found on pseudo-nouns only, with derived items yielding faster gender decision times (Exp.1) but slower lexical decisions (Exp.2). The findings are consistent with models supporting a dual-route access to grammatical gender, while the lexical nature of both inflectional and derivational suffixes is also attested.

This paper deals with the design of an online morphological priming task, aiming to investigate the parameter of semantic transparency with regard to denominal derivation in Standard Modern Greek. First, we outline the methodological framework and specify our research scope. Then, we examine semantic transparency, providing a theoretical account, describing three offline semantic relatedness pretests and defining semantic transparency as independent variable from functional point of view. Next, we set out certain criteria for selecting experimental items and refer to the extraneous variables matched among the priming conditions (transparent, semitransparent, opaque). Finally, we present the experimental design and discuss some appropriate techniques for its application, as well as the research questions we expect to address.

Deverbal word formation is subject to two distinct types of constraints , those concerning the syntactic category of the base (categorial constraints) and those relating to the thematic properties of the verb (thematic constraints). For instance,-able suffixation involves a transitive verb with argument structure , as in to train > trainable. Violation of these constraints results in the creation of pseudo-words with categorial (e.g. riverable) or thematic violations (e.g. arrivable). The study discusses psycholinguistic experiments involving these types of deverbal pseudo-words, in Greek and English, two languages with morphologically distinct properties. Greek has a rich derivational system with a variety of deverbal formations, which follow strong constraints, in the sense that most suffixes that participate in deverbal word formation lack the polysemy that allows them to attach to other-than-verbal bases. English, on the other hand, demonstrates an equally rich derivational s...

Two lexical decision tasks, one with overt and one with masked priming, were carried out in order to explore lexical access and mental representation of coordinative compounds in Modern Greek. Results of both experiments showed strong priming effects in both first and second constituent recognition when primed by the whole compound, pointing to lexical access through robust decomposition and, consequently, a structured mental representation where both constituents contribute equally. The study provides some initial insights about coordination at the lexical level and also discusses the role of different relational information between constituents during lexical access of compounds.

In relevance-theoretic pragmatics, lexical adjustment, or ad hoc-concept construction, is considered a case of free pragmatic enrichment. It is hence included within the group of non-linguistically mandated processes that are automatically carried out during comprehension. Its output is an inferentially-specified representation of the conceptual content encoded, or activated, by open-class words like nouns, adjectives, verbs or adverbs. Ad hoc-concept construction is a necessary step to obtain a fully propositional, truth-evaluable form that captures in a more precise manner the portion of meaning that the speaker communicates explicitly. However, this paper argues that this lexical pragmatic process needs not always be described as completely exempt from linguistic mandate: it may sometimes be enacted and steered by a series of linguistic elements. These fall within two categories: (i) the evaluative morphemes that some lexical items may receive in some languages, and (ii) lexical and phrasal items adjacent to content words. These elements would finetune the representations arising from content words in terms of more specific, context-dependent features or shades. Additionally, they could include in such representations information about the speaker’s attitude towards, or stance about, what the speakers refer to. Therefore, that information needs not always be derived as implicatures.

Strict compositionality in morphological theory is problematic for explaining how language-users comprehend phenomena like the partial yet non-decomposable forms in phonaesthemes and in blends like EDUTAINMENT. An alternative account, based on discriminative learning, proposes that language-users associate linguistic cues (e.g., short segment or letter strings) with multiple simultaneous possible lexical and grammatical meanings. We evaluate this account on off-line human identifications of partial word-forms, using English blend words as our test case. We hypothesize that readers' ability to parse out source meanings from written blend forms should be correlated with how strongly a na\"ive discriminative reading  model associates the cues in each form with the correct source meanings. We provide evidence for this claim in two experiments, in which the discriminative learning model reliably predicted participants' success rate in guessing the sources of both attested and novel blends. This finding supports discriminative learning as a realistic model of how readers parse wordforms and map them to meanings. Further, the result points towards a novel, precise account of blend processing.

In this article we present two lexical decision experiments that examine the role of base frequency and of derivative suffix productivity in visual recognition of Spanish words. In the first experiment we find that complex words with productive derivative suffixes result in lower response times than those with unproductive derivative suffixes. There is no significant effect for base frequency, however. In experiment two, the same procedure was undertaken with pseudowords, showing that when they are composed by productive derivative suffixes they take longer to be rejected than when they are composed by unproductive derivative suffixes. Again, the role of base frequency fails to reach significance. These results endorse the view that derivative suffixes have a relevant role in visual recognition of complex words. According to our results, derivative suffixes create the conditions for taking a lexical candidate as a legal lexical entry and therefore they contribute decisively to the l...

English makes use of a wide-spread pattern of word class alternation known as ‘zero-derivation.’ This involves pairs of homophonous forms which are semantically related, yet differ in part-of-speech (e.g. a knot vs. to knot). Many theories have been proposed to describe the relationship between these forms, with some proposing that to knot is covertly derived from a knot in the same way as government is from govern; an alternative view is that these pairs are instead two forms of a single lexeme with no inherent word class. We explore these claims in the context of morphological processing, using three delayed priming lexical decision tasks. The results suggest that some pairs are in a covert derivational relationship, in which the derived form is morphologically more complex than the base. However, not all such pairs are related this way, as some instead behave like inflectional relatives belonging to a single, underspecified lexical entry. Together, the experiments offer support for a mixed model, in which the grammar distinguishes between different kinds of zero-related pairs based on their underlying morphological relationships – a covert distinction to which morphological processing may be sensitive.

This work presents the results of a masked lexical decision experiment in which we explore the morphological parsing of Spanish suffixed or pseudosuffixed words through the suffix priming effect. Priming the bases or pseudo-bases with their suffixed or pseudosuffixed forms is the standard process in experiments aimed at understanding the processes underlying morphological parsing in visual word recognition with masked priming lexical decision (e.g. darkness - DARK; corner - CORN). We, however, compare the effect of suffix priming on the lexical decision of suffixed (ero - JORNALERO) and pseudosuffixed words (ero - CORDERO), as well as the effect of orthographic priming on non-suffixed words (eba - PRUEBA). The results show that in the case of suffixed and pseudosuffixed words, related primes (ero - JORNALERO; ero - CORDERO) significantly accelerated response latencies in comparison to unrelated primes (ista - JORNALERO; ura - CORDERO), while for simple words there was no facilitation from the orthographically related prime in comparison to the unrelated prime (eba - PRUEBA; afo - PRUEBA). These results are consistent with the so-called morpho-orthographic segmentation process in the course of visual word recognition, which might also be independent of orthographic and purely semantic factors. Our results also support the view that morphological parsing takes place regardless of whether a stem is present in a word. These results complement findings from studies dealing with CORNER- and BROTHEL-like stimuli.

Deverbal word formation is subject to two distinct types of constraints, those concerning the syntactic category of the base (categorial constraints) and those relating to the thematic properties of the verb (thematic constraints). For instance, -able suffixation involves a transitive verb with argument structure <Agent<Theme>>, as in to train > trainable. Violation of these constraints results in the creation of pseudo-words with categorial (e.g. riverable) or thematic violations (e.g. arrivable). The study discusses psycholinguistic experiments involving these types of deverbal pseudo-words, in Greek and English, two languages with morphologically distinct properties. Greek has a rich derivational system with a variety of deverbal formations, which follow strong constraints, in the sense that most suffixes that participate in deverbal word formation lack the polysemy that allows them to attach to other-than-verbal bases. English, on the other hand, demonstrates an equally rich derivational system, but it differs in two crucial ways: (a) there is significant affix homophony (e.g. -er is a nominalizer if attached to verbal stems, or forms the comparative if attached to adjectives), (b) it is extremely permissive in allowing zero-derived verbs (to fax). In an off-line and two on-line lexical decision tasks we investigated whether categorial and thematic constraints are treated in the same way by speakers of both languages. Results showed that speakers of both languages differentiated between pseudo-words that violate these two types of constraints both when it comes to acceptance rates and processing time. Taking together results from both languages, we make claims about the structured mental representation of deverbal derivatives and the fact that their various properties can be accessed via distinct operations and at distinct points of time. Implications for the psycholinguistic theory of lexical access and the morphological theory of word formation are also discussed.