Action Semantics Research Articles

Action verbs drive motor activity in adolescents but not in children

In adults, grip force has reliably been used to investigate motor simulation evoked by linguistic action, suggesting that motor phenomena are linked to semantic action. The parietal and frontal lobes and their connexions are essential neural structures for pragmatic aspects of hand semantic action. In this perspective, the aim of the study was to determine the extent to which two groups of children and adolescents, classically characterized by degree of axonal myelination in fronto-parietal circuits, monitored the occurrence of nouns and manual action verbs presented auditorily while holding a grip force sensor. Differential effects of grip force were seen only in the adolescents when monitoring action verbs. Interestingly, weaker effects of grip force were modulated by noun targets only in the younger children, revealing that the ability to profit from a full semantic representation of verbs is not clearly established in the younger children. Grip force modulation was observed as early as 300 ms post target onset and peaked at the 500–750 ms window of observation for both groups. These group differences are in line with the motor simulation difficulties seen in younger children. The results may also indicate that degree of grip force in response to specific linguistic categories parallels the maturation of the parietal-frontal circuits, including the anterior intra-parietal area which plays a determining role in semantic aspects of hand action.

How Do Living Systems Create Meaning?

Meaning has traditionally been regarded as a problem for philosophers and psychologists. Advances in cognitive science since the early 1960s, however, broadened discussions of meaning, or more technically, the semantics of perceptions, representations, and/or actions, into biology and computer science. Here, we review the notion of “meaning” as it applies to living systems, and argue that the question of how living systems create meaning unifies the biological and cognitive sciences across both organizational and temporal scales.

On the Boundaries between Decision and Action: Effector-selective Lateralization of Beta-frequency Power Is Modulated by the Lexical Frequency of Printed Words.

Current computational and neuroscientific models of decision-making posit a discrete, serial processing distinction between upstream decisional stages and downstream processes of motor-response implementation. We investigated this framework in the context of two-alternative forced-choice tasks on linguistic stimuli, words and pseudowords. In two experiments, we assessed the impact of lexical frequency and action semantics on two effector-selective EEG indexes of motor-response activation: the lateralized readiness potential and the lateralization of beta-frequency power. This allowed us to track potentially continuous streams of processing progressively mapping the evaluation of linguistic stimuli onto corresponding response channels. Whereas action semantics showed no influence on EEG indexes of motor-response activation, lexical frequency affected the lateralization of response-locked beta-frequency power. We argue that these observations point toward a continuity between linguistic processing of word input stimuli and implementation of corresponding choice in terms of motor behavior. This interpretation challenges the commonly held assumption of a discrete processing distinction between decisional and motor-response processes in the context of decisions based on symbolic stimuli.

ELiT, Multifunctional Web-Software for Feature Extraction from 3D LiDAR Point Clouds

Our research presents a complete R&D cycle—from the urban terrain generation and feature extraction by raw LiDAR data processing, through visualizing a huge number of urban features, and till applied thematic use cases based on these features extracted and modeled. Firstly, the paper focuses on the original contribution to algorithmic solutions concerning the fully automated extraction of building models with the urban terrain generation. Topography modeling and extraction of buildings, as two key constituents of the robust algorithmic pipeline, have been examined. The architectural scheme of the multifunctional software family—EOS LIDAR Tool (ELiT) has been presented with characteristics of its key functionalities and examples of a user interface. Both desktop, and web server software, as well as a cloud-based application, ELiT Geoportal (EGP), as an entity for online geospatial services, have been elaborated on the base of the approach presented. Further emphasis on the web-visualization with Cesium 3D Tiles has demonstrated the original algorithm for efficient feature visualizing though the EGP locations. Summarizing presentation of two thematic use-cases has finalized this research, demonstrating those applied tasks, which can be efficiently resolved with the workflow presented. A necessity of a conclusive workflow elaboration for use cases, which would be based on the actual semantics, has been emphasized.

Sememe Heredity of Action Semantics: Evidence From the Priming Effect and Prospective Memory.

The sememe heredity of action semantics may be affected by the related association of a verb or noun in an action phrase and the related association between one action phrase and another. The motor encoding theory and the five-component view of the subject-performed task supported by the verb’s specificity highly promote the enactment effect. However, the episodic integration theory emphasizes the role of semantic integration between the verb and noun on the enactment effect. In this study, a subject-performed task was combined with a priming paradigm and found that verb-semantic priming quantity was more significant than that of noun-semantic priming quantity under motor encoding in Experiment 1. Besides, it was observed that the verb-semantic association might play a more significant role in the sememe heredity of action semantics. Therefore, in Experiment 2, a subject-performed task was combined with the dual task of prospective memory. Results showed that the accuracy of prospective memory targets related to the learning phrases was significantly higher compared to that of the prospective memory targets unrelated to the learning phrases. Besides, the above difference is more evident verbally compared to motor encoding conditions. Thus, the sememe heredity of action semantics may rely on the related association of action semantic contents rather than on the semantic processing form of external motor encoding.

Semantic Interpretation of Top-N Recommendations

Over the years, model-based approaches have shown their effectiveness in computing recommendation lists in different domains and settings. By relying on the computation of latent factors, they can recommend items with a very high level of accuracy. Unfortunately, when moving to the latent space, even if the model embeds content-based information, we miss references to the actual semantics of the recommended item. It makes the interpretation of the recommendation process non-trivial. In this paper, we show how to initialize latent factors in Factorization Machines by using semantic features coming from knowledge graphs to train an interpretable model, which is, in turn, able to provide recommendations with a high level of accuracy. In the presented approach, semantic features are injected into the learning process to retain the original informativeness of the items available in the dataset. By relying on the information encoded in the original knowledge graph, we also propose two metrics to evaluate the semantic accuracy and robustness of knowledge-aware interpretability. An extensive experimental evaluation on six different datasets shows the effectiveness of the interpretable model in terms of both accuracy and diversity of recommendation results and interpretability robustness.

Treating limb apraxia via action semantics: a preliminary study

ABSTRACT Limb apraxia is evident in approximately 50% of patients after left hemisphere cerebral vascular accident (LCVA) and increases disability and caregiver dependence. Individuals with apraxia exhibit abnormalities in spatio-temporal aspects of gesture production and/or in knowledge of tool-related actions (action semantics). This preliminary study of three LCVA participants aimed to (i) explore the efficacy of a novel Action Network Treatment (ANT) that focused on improving the semantic association between tool actions and other types of tool knowledge, an intervention inspired by successful semantic network treatments in aphasia (e.g., Edmonds et al., 2009), and (ii) explore whether there are individuals with apraxia who benefit from ANT relative to a version of a comparatively well-studied existing apraxia treatment (Smania et al., 2006; Smania et al., 2000) that shapes gesture via focus on practicing the spatio-temporal aspects of gesture production (Tool Use Treatment or TUT). One participant demonstrated treatment benefits from both ANT and TUT, while another only benefited from TUT. These findings indicate that our novel semantic network strengthening approach to gesture training may be efficacious in at least some individuals with apraxia, and provide a foundation for future study of the characteristics of people with apraxia who benefit from each approach.

Hip-hop Identity in the Music Video <This Is America>

힙합(Hip-hop)은 전 세계 대중음악의 주류 장르로 확산하고 있다. 특히 미디어를 통해 오락적인 요소로 표면에 비치는 도식화된 힙합의 이미지와 이름으로 청소년들 사이에 랩을 유행시켰다. 10대를 중심으로 한 팬덤은 음악가의 음악에도 영향을 미치며, 주제의 특성을 변질시켰다. 힙합은 음악 이전에 흑인으로부터 발생한 미국의 하위문화 형태다. 이를 통해 문화의 정체성을 확립하였으며, 사회 비판과 저항의 특성을 지니고 있다. 본고에서는 2018년 발표, 대중적인 성공을 거둔 <This is America>의 뮤직비디오를 분석함으로 힙합의 정체성 확인을 하고자 한다. 뮤직비디오의 의미작용은 기호학자 롤랑 바르트의 신화론과 그레마스의 사각이론을 통해 분석하였다. 시간 간격으로 구성된 뮤직비디오의 서사 구조를 3가지 분류 기준으로 나누어 각각의 장면에 내재한 의미를 밝혀내는 것으로 힙합의 정체성이 현시대 힙합 음악에도 나타나는 것을 증명한다. 이를 통해 주류의 문화로만 인식되는 힙합의 방향적인 경계를 갖출 수 있으며, 하위문화라는 특성을 통해 이를 재현할 수 있음을 확인하고자 하는 논문이다.Hip-hop is gaining popularity as a mainstream genre of popular music around the world. In particular, rap has become popular among teenagers and the image and topic of hip-hop are reflected as elements of entertainment through media. The teen-centered fandom also influences the music of musicians and alters the characteristics of the theme. Hip-hop is a form of American subculture that originated in early black folk music. Through this, the identity of culture was established and the characteristics of social criticism and resistance were developed. In this article, I will confirm the identity of hip-hop by analyzing the music video <This is America>, which was released in 2018 and became a popular success. The semantic action of the music video was analyzed through theories of semiotics, Such as that of Roland Barthes, and Greimas’ square theory. By dividing the narrative structure of the music video composed of time intervals into three classification criteria, the meaning inherent in each scene can be revealed. Through this, it is possible to obtain have a directional boundary of hip-hop, which is recognized only as mainstream culture, and reconstruct it through the characteristics of subculture.

Deriving overloaded success type schemes in Erlang

Erlang is a programming language which brings together the features of functional programming and actor-based concurrency. Although it is a dynamically-typed language, there exists a tool ( Dialyzer ) that analyses Erlang programs in order to detect type discrepancies at compile-time. This tool is based on the notion of success types , which are overapproximations to the actual semantics of expressions, so that the evaluation of an ‘ill-typed’ expression will eventually fail at runtime. Dialyzer allows programmers to provide their own type specifications. Although such specifications can be polymorphic and overloaded (i.e., reflecting different executing branches) for documentation purposes, the type analysis disregards the information provided by polymorphic type schemes and so does, in some cases, with overloaded types. In this paper we introduce: (1) a type system that allows us to obtain polymorphic overloaded success type schemes for programs, (2) a semantic definition of this kind of types, and (3) correctness results that prove that the adequacy of the obtained types w.r.t. the semantics of expressions.

MALA: Cross-Domain Dialogue Generation with Action Learning

Response generation for task-oriented dialogues involves two basic components: dialogue planning and surface realization. These two components, however, have a discrepancy in their objectives, i.e., task completion and language quality. To deal with such discrepancy, conditioned response generation has been introduced where the generation process is factorized into action decision and language generation via explicit action representations. To obtain action representations, recent studies learn latent actions in an unsupervised manner based on the utterance lexical similarity. Such an action learning approach is prone to diversities of language surfaces, which may impinge task completion and language quality. To address this issue, we propose multi-stage adaptive latent action learning (MALA) that learns semantic latent actions by distinguishing the effects of utterances on dialogue progress. We model the utterance effect using the transition of dialogue states caused by the utterance and develop a semantic similarity measurement that estimates whether utterances have similar effects. For learning semantic actions on domains without dialogue states, MALA extends the semantic similarity measurement across domains progressively, i.e., from aligning shared actions to learning domain-specific actions. Experiments using multi-domain datasets, SMD and MultiWOZ, show that our proposed model achieves consistent improvements over the baselines models in terms of both task completion and language quality.

Purely functional GLL parsing

Abstract Generalised parsing has become increasingly important in the context of software language design and several compiler generators and language workbenches have adopted generalised parsing algorithms such as GLR and GLL. The original GLL parsing algorithms are described in low-level pseudo-code as the output of a parser generator. This paper explains GLL parsing differently, defining the FUN-GLL algorithm as a collection of pure, mathematical functions and focussing on the logic of the algorithm by omitting implementation details. In particular, the data structures are modelled by abstract sets and relations rather than specialised implementations. The description is further simplified by omitting lookahead and adopting the binary subtree representation of derivations to avoid the clerical overhead of graph construction. Conventional parser combinators inherit the drawbacks from the recursive descent algorithms they implement. Based on FUN-GLL, this paper defines generalised parser combinators that overcome these problems. The algorithm is described in the same notation and style as FUN-GLL and uses the same data structures. Both algorithms are explained as a generalisation of basic recursive descent algorithms. The generalised parser combinators of this paper have several advantages over combinator libraries that generate internal grammars. For example, with the generalised parser combinators it is possible to parse larger permutation phrases and to write parsers for languages that are not context-free. The ‘BNF combinator library’ is built around the generalised parser combinators. With the library, embedded and executable syntax specifications are written. The specifications contain semantic actions for interpreting programs and constructing syntax trees. The library takes advantage of Haskell’s type-system to type-check semantic actions and Haskell’s abstraction mechanism enables ‘reuse through abstraction’. The practicality of the library is demonstrated by running parsers obtained from the syntax descriptions of several software languages.

Action processing in the motor system: Transcranial Magnetic Stimulation (TMS) evidence of shared mechanisms in the visual and linguistic modalities

In two experiments, we compared the dynamics of corticospinal excitability when processing visually or linguistically presented tool-oriented hand actions in native speakers and sequential bilinguals. In a third experiment we used the same procedure to test non-motor, low-level stimuli, i.e. scrambled images and pseudo-words.Stimuli were presented in sequence: pictures (tool + tool-oriented hand action or their scrambled counterpart) and words (tool noun + tool-action verb or pseudo-words). Experiment 1 presented German linguistic stimuli to native speakers, while Experiment 2 presented English stimuli to non-natives. Experiment 3 tested Italian native speakers. Single-pulse trascranial magnetic stimulation (spTMS) was applied to the left motor cortex at five different timings: baseline, 200 ms after tool/noun onset, 150, 350 and 500 ms after hand/verb onset with motor-evoked potentials (MEPs) recorded from the first dorsal interosseous (FDI) and abductor digiti minimi (ADM) muscles.We report strong similarities in the dynamics of corticospinal excitability across the visual and linguistic modalities. MEPs’ suppression started as early as 150 ms and lasted for the duration of stimulus presentation (500 ms). Moreover, we show that this modulation is absent for stimuli with no motor content. Overall, our study supports the notion of a core, overarching system of action semantics shared by different modalities.

Usage-Based Learning in Human Interaction With an Adaptive Virtual Assistant

Today users can interact with popular virtual assistants such as Siri to accomplish their tasks on a digital environment. In these systems, links between natural language requests and their concrete realizations are specified at the conception phase. A more adaptive approach would be to allow the user to provide natural language instructions or demonstrations when a task is unknown by the assistant. An adaptive solution should allow the virtual assistant to operate a much larger digital environment composed of multiple application domains and providers and better match user needs. We have previously developed robotic systems, inspired by human language developmental studies, that provide such a usage-based adaptive capacity. Here, we extend this approach to human interaction with a virtual assistant that can first learn the mapping between verbal commands and basic action semantics of a specific domain. Then, it can learn higher level mapping by combining previously learned procedural knowledge in interaction with the user. The flexibility of the system is demonstrated as the virtual assistant can learn actions in new domains (e-mail, Wikipedia, etc.), and then can learn how e-mail and Wikipedia basic procedures can be combined to form hybrid procedural knowledge.

VisAct: a visualization design system based on semantic actions

Current visualization design toolkits help users to explore data and create visualizations. However, most of these systems do not record the executed actions during the visualization construction process with semantic context. In this paper, we present VisAct, a visualization design system based on semantic actions, that helps average users to construct visualizations step by step. Our system contributes a set of action-based visualization components and a high-level grammar for semantic actions. VisAct also guides the visualization construction process and provides an action tracker for history management and data collection. We demonstrate the usability of VisAct by visualizations and a plugin application. Finally, we conduct a user study to evaluate the efficiency and effectiveness of our system.

From action to abstraction: The sensorimotor grounding of metaphor in Parkinson's disease

Embodied cognition theories propose that the semantic representations engaged in during language comprehension are partly supported by perceptual and motor systems, via simulation. Activation in modality-specific regions of cortex is associated with the comprehension of literal language that describes the analogous modalities, but studies addressing the grounding of non-literal or figurative language, such as metaphors, have yielded mixed results. Differences in the psycholinguistic characteristics of sentence stimuli across studies have likely contributed to this lack of consensus. Furthermore, previous studies have been largely correlational, whilst patient studies are a critical way of determining if intact sensorimotor function is necessary to understand language drawing on sensorimotor information. We designed a battery of metaphorical and literal sentence stimuli using action and sound words, with an unprecedented level of control over critical psycholinguistic variables, to test hypotheses about the grounding of metaphorical language. In this Registered Report, we assessed the comprehension of these sentences in 41 patients with Parkinson's disease, who were predicted to be disproportionately affected by the action sentences relative to the sound sentences, and compared their performance to that of 39 healthy age-matched controls who were predicted to show no difference in performance due to sensory modality. Using preregistered Bayesian model comparison methods, we found that PD patients' comprehension of literal action sentences was not impaired, while there was some evidence for a slowing of responses to action metaphors. Follow up exploratory analyses suggest that this response time modality effect was driven by one type of metaphor (predicate) and was absent in another (nominal), despite the fact that the action semantics were similar in both syntactic forms. These results suggest that the conditions under which PD patients demonstrate hypothesized embodiment effects are limited. We offer a critical assessment of the PD action language literature and discuss implications for the embodiment debate. In addition, we suggest how future studies could leverage Bayesian statistical methods to provide more convincing evidence for or against embodied cognition effects.

Dissociating action and abstract verb comprehension post-stroke

The neural bases of action and abstract concept representations remain a topic of debate. While several lines of research provide evidence for grounding of action-related conceptual content into sensory-motor systems, results of traditional lesion-deficit studies have been somewhat inconsistent. Further, few studies have directly compared the neural substrates of action and relatively abstract verb comprehension post-stroke. Here, we investigated the impact of the disruption of two neural networks on comprehension of action and relatively abstract verbs in 48 unilateral left-hemisphere stroke patients using two methodologies: 1) lesion-deficit association and 2) resting-state functional connectivity (RSFC) analyses. Disruption of RSFC between the left inferior frontal gyrus and right hemisphere primary and secondary sensory-motor areas predicted greater relative impairment of action semantics. Voxel-based lesion-symptom mapping revealed that damage to frontal white matter, extending towards the inferior frontal gyrus, also predicted greater relative impairment of action semantics. On the other hand, damage to the left anterior middle temporal gyrus significantly impaired the more abstract category relative to action. These findings support the view that action and non-action/abstract semantic processing rely on partially dissociable brain networks, with action concepts relying more heavily on sensory-motor areas. The results also have wider implications for lesion-deficit association studies and show how the contralateral hemisphere can play a compensatory role following unilateral stroke.

Library of actions: Implementing a generic robot execution framework by using manipulation action semantics

When a robot has to imitate an observed action sequence, it must first understand the inherent characteristic features of the individual actions. Such features need to reflect the semantics of the action with a high degree of invariance between different demonstrations of the same action. At the same time the machine needs to be able to execute the action sequence in any appropriate situation. In this study, we introduce a new library of actions, which is a generic framework for executing manipulation actions on robotic systems by combining features that capture action semantics with a framework for execution. We focus on manipulation actions and first create a generic representation consisting of symbolic and sub-symbolic components. To link these two domains we introduce a finite state machine allowing for sequential execution with error handling. The framework is developed from observing humans which provides us with a high degree of grounding. To quantitatively evaluate the scalability of the proposed approach, we conducted a large set of experiments involving different actions performed either individually or sequentially with various types of objects in different scene contexts.

NOMINALIZED NOUNS IN UKRAINIAN AND ENGLISH LANGUAGES: SEMANTICS AND FUNCTIONING

The article is dedicated to the analysis of semantics of verbal nouns as a consequence of nominalization processes in language. The notion of nominalization leads to the reduction of traditional verbal semantics – verbal features, functions or elements related to implementation of process and points to development and flow of phenomena, facts and events localized in time and space. Nominalized unit acquires signs of static semantics, the main function of which is naming and fixing of certain phenomena, facts and events, as well as identification of these facts, phenomena and events without taking into account their possible development in time and space.
 The difference between the phenomena of substantivation and nominalization is established. Nominalized verbal nouns replace the predicate and take the position of the subject of the subordinate clause, acting as a means of secondary nomination, performing an anaphorical function.
 Verbal nouns in the sentence structure act as a means of condensation and are semantic equivalents of sentences. They form a specific stratum of vocabulary both in Ukrainian and English languages which is characterized, on the one hand, by a close connection with the original verbs and, as a consequence, the preservation of lexical and even grammatical semantics of original verbs – semantics of action, process or dynamic state, and, on the other hand, belongs to the lexical-grammatical class of nouns, which causes not only significant semantic modifications of the original semantics, but also the emergence of various semantic components and connotations.
 Both English and Ukrainian verbal nouns can transmit the meaning of state, process, the result of action, although their interpretation is modified by the context. The peculiarities of the transformation of sentences or similar units with predicative nature in English and Ukrainian languages are revealed.

Concrete processing of action metaphors: Evidence from ERP

The timing of sensory-motor activation during the comprehension of action verbs used in a metaphorical sense is not well understood. In the present Event Related Potential (ERP) study, participants read verbs in metaphoric (The church bent the rules), literal-concrete (The bodyguard bent the rod), and literal-abstract (The church altered the rules) conditions. The literal concreteness effect, obtained by subtracting the abstract from the concrete, was revealed as an N400, frontally distributed. A metaphoric effect, obtained in the metaphor-abstract contrast, was a widespread N400, and included the frontal response seen in the literal concreteness effect. Another metaphoric effect, obtained in the metaphor-concrete contrast, was a posterior N400. Further time window analyses showed that the literal concreteness effect primarily came from 200 to 300 ms, the metaphoric-concrete effect primarily came from 200 to 400 ms, and the metaphoric-abstract effect was significant throughout 200–500 ms. These results suggest that a concrete but underspecified meaning consistent with metaphoric and literal readings, was activated early and was sustained throughout the 200–500 ms window. We concluded that the metaphoric sense is based in concrete action semantics, even if these senses are underspecified.

The Particle ??: A Key Aspect to Fathom in Translation with Special Reference to the Holy Quran

<p><em>Cognizant of the deep chasm between the actual semantics of the particle </em>??<em> and the superficial way it is understood and thus translated, I have decided to devote this study to minutely examine the two types of </em>??<em>. The paper explores the interrogative and declarative </em>??<em> along with their characteristic features, meaning and their realizations in translation into English, with a special reference to the Quran. Fresh tides of thought have conquered me to clarify the semantics of </em>??<em> which often fetters trainee translators and students of translation when they- in vain- try to capture the exact meaning of each of them. It is hoped that this paper helps its readers gain practical mastery of some salient features of the Arabic language. It aims at developing readers’ level of linguistic competence. </em></p>