Search

Research has shown that first-time voting experiences affect subsequent voting behavior, with salient elections boosting subsequent turnout and non-salient ones suppressing it. We challenge this view. Following research on the context-dependent nature of habit formation, we argue that all elections should affect subsequent turnout in elections of the same type. Comparing individuals that differ only in how salient their first eligible election was (Presidential or Midterm), we find support for this expectation. Individuals are more likely to vote for, and be interested in, elections of the same type as their first voting experience. Leveraging voting age laws in the US, we also show that such laws affect subsequent participation by changing the type of election individuals are first eligible for.

Educational gaps are increasingly salient as skills and knowledge gain prominence in digital societies. E-privacy management, namely, the ability to control the flow of information about the self, is an important asset nowadays, since a skillful use of digital technologies enables full participation in social life and limits the exposure to unwarranted algorithmic processes. We investigate whether and why education affects e-privacy management, and whether the educational gaps vary following a country’s degree of digitalization. We empirically test two sets of mechanisms, one derived from the digital divide and diffusion of innovations theories and the other from the reflexive modernization theory. The study employs Eurobarometer 87.1 data ( N = 21,177), collected in 2017 among representative samples from 28 European countries, and uses multilevel linear regression model. Findings suggest that the years spent in education positively affect e-privacy management, and that this effect is largely mediated by digital skills and Internet use, and to a lesser extent by a reflexive mind-set. The educational gap in e-privacy management narrows in more digitalized countries.

Most existing light field saliency detection methods have achieved great success by exploiting unique light field data-focus information in focal slices. However, they process light field data in a slicewise way, leading to suboptimal results because the relative contribution of different regions in focal slices is ignored. How we can comprehensively explore and integrate focused saliency regions that would positively contribute to accurate saliency detection. Answering this question inspires us to develop a new insight. In this article, we propose a patch-aware network to explore light field data in a regionwise way. First, we excavate focused salient regions with a proposed multisource learning module (MSLM), which generates a filtering strategy for integration followed by three guidances based on saliency, boundary, and position. Second, we design a sharpness recognition module (SRM) to refine and update this strategy and perform feature integration. With our proposed MSLM and SRM, we can obtain more accurate and complete saliency maps. Comprehensive experiments on three benchmark datasets prove that our proposed method achieves competitive performance over 2-D, 3-D, and 4-D salient object detection methods. The code and results of our method are available at https://github.com/OIPLab-DUT/IEEE-TCYB-PANet.

Video-based person re-identification (Re-ID) aims at retrieving the person through the video sequences across non-overlapping cameras. Some characteristics of pedestrians are not consecutive across frames due to the variations of viewpoints, postures, and occlusions over time. However, existing methods ignore such data peculiarity and the networks tend to only learn those salient consecutive characteristics among frames in video sequences. As a result, the learned representations fail to cover all the characteristics of pedestrians, thus lacking integrity and discrimination. To tackle this problem, we present a novel deep architecture termed Hierarchical Mining Network (HMN), which mines as many pedestrians’ characteristics by referring to the temporal and intra-class knowledge. It consists of a novel Attentive Temporal Module (ATM) and a Dynamic Supervising Branch (DSB), with a Balancing Triplet Loss (BTL) assisting the training. The proposed ATM, with pedestrian perceiving capacity, is capable of evaluating each activation of features through temporal analysis, so that the temporally scattered characteristics of pedestrians can be better aggregated and the contaminated ones can be eliminated. Then, the DSB along with the BTL further enhances the integrity of representations by multiple supervision. Specifically, the DSB perceives the diversities of intra-class samples in each mini-batch and generates targeted supervising signals for them, in which process the BTL guarantees the signals with smaller intra-class variations and larger inter-class variations. Comprehensive experiments on two video-based datasets, i.e., MARS, and DukeMTMC-VideoReID, demonstrate the contribution of each component and the superiority of the proposed HMN over the state-of-the-arts. Benchmarking our model on three popular image-based datasets, i.e., Market1501, DukeMTMC-Reid, and MSMT17 additionally verifies the promising generalizability of the proposed DSB and BTL.

The Tax Cuts and Jobs Act (TCJA) significantly altered how business income is taxed in the US. This paper provides a quantitative assessment of the distributional and macroeconomic effects of the TCJA, both in the short run and in the long run, using a life-cycle model with occupational choice and accumulation of entrepreneurial human capital. When salient provisions of the TCJA are implemented, the economy experiences an increase of 0.20 and 0.5 percentage points in annual GDP and capital stock growth for the first decade 2018-2027, respectively. However, economic inequality increases across and within occupations under the TCJA. The provisions affect the incentives for individuals to save, making the occupational choice extensive margin critically important in analyzing the potential impacts of the new tax reform.

Over the past decades, video salient object detection has achieved rapid development. However, the effect of edge information on spatial and temporal features extraction has not been explored. The method STEG-Net is proposed to solve the problem, we use the extracted edge information to guide the extrction of spatial and temporal features simultaneously. We combine deep texture information with shallow edge information, which can not only retain the edge of the object but also enhance the global information, leading to the more accurate location of the object. At the same time, the shallow edge information and the deep texture information are complementary. As a result, edge information and salient features could advance each other while extracting features. Finally, we evaluate the 16 state-of-the-art VSOD methods on ViSal and FBMS datasets. At the same time we also evaluate the 13 state-of-the-art VSOD methods on UVSD and MCL datasets. The experiment results indicate that the proposed method outperforms the state-of-the-art approaches.

Categorizing aerial photographs with varied weather/lighting conditions and sophisticated geomorphic factors is a key module in autonomous navigation, environmental evaluation, and so on. Previous image recognizers cannot fulfill this task due to three challenges: 1) localizing visually/semantically salient regions within each aerial photograph in a weakly annotated context due to the unaffordable human resources required for pixel-level annotation; 2) aerial photographs are generally with multiple informative attributes (e.g., clarity and reflectivity), and we have to encode them for better aerial photograph modeling; and 3) designing a cross-domain knowledge transferal module to enhance aerial photograph perception since multiresolution aerial photographs are taken asynchronistically and are mutually complementary. To handle the above problems, we propose to optimize aerial photograph's feature learning by leveraging the low-resolution spatial composition to enhance the deep learning of perceptual features with a high resolution. More specifically, we first extract many BING-based object patches (Cheng et al., 2014) from each aerial photograph. A weakly supervised ranking algorithm selects a few semantically salient ones by seamlessly incorporating multiple aerial photograph attributes. Toward an interpretable aerial photograph recognizer indicative to human visual perception, we construct a gaze shifting path (GSP) by linking the top-ranking object patches and, subsequently, derive the deep GSP feature. Finally, a cross-domain multilabel SVM is formulated to categorize each aerial photograph. It leverages the global feature from low-resolution counterparts to optimize the deep GSP feature from a high-resolution aerial photograph. Comparative results on our compiled million-scale aerial photograph set have demonstrated the competitiveness of our approach. Besides, the eye-tracking experiment has shown that our ranking-based GSPs are over 92% consistent with the real human gaze shifting sequences.

In this paper I look at how through the use of teasing as a socially recurrent activity the members of a multilingual, multicultural and multiethnic peer group (comprised of majority Greek and minority Turkish-speaking children of Roma heritage) make particular identity ascriptions and displays salient and position themselves and others in particular ways in peer talk during break-time in an Athens primary school. Taking as a point of departure that identities are produced relationally, through systems of opposition (Barth 1969), the paper deals with how members of this school-based peer group exploit teasing as a versatile discursive device to construct one particular peer as a “poor” pupil and themselves by extension as “good” pupils in talk-in-interaction. The focus on the situated and relational construction of identity makes visible how children position themselves with regard to others in order to construct academic hierarchies. At the same time, it brings to the fore how through such positionings children may reproduce but also challenge powerful institutional discourses of academic success and failure in circulation in the classroom by negotiating identity options closer to their peer concerns. These processes of identity construction demonstrate how social selves are produced in interaction through contestation and collaboration and how identities may be simultaneously chosen and imposed through language use.

Interactants’ non-reciprocal use of Japanese speech styles, i.e., the addressee honorific masu form and the non-honorific plain form, is frequently treated as the salient feature constituting speakers’ hierarchical identities. The hierarchical identities in question in this study are senpai-koohai ‘senior-junior’relationships among Japanese college students. The paper presents analyses that demonstrate that the construction of these hierarchical relationships depends on context. The data derive from nine hours of audio recordings of dyadic and multiparty interactions among college students at the meetings of an extracurricular club. Conceptualizing on-stage and off-stage as frames of talk that function as context in this data set, the study finds that hierarchical identities are not foregrounded during on-stage talk, but can be foregrounded during off-stage talk when the participants’ club roles are not foregrounded; the use of non-reciprocal speech styles that lead to hierarchical identity construction is observed in this situation. On the other hand, hierarchical identities are backgrounded during on-stage talk when the participants’ club roles are foregrounded. The use of the addressee honorific masu form in this situation indexes that the speaker is engaged in a club role, such as discussion leader or participant.

One of the canonical uses of jakby in Polish is that of the Lakoffian hedge, which modifies the propositional content of an utterance by pointing to its fuzziness, inexactitude or approximation. In conversational speech the word is frequently put to excessive use, which appears to significantly deviate from the prescribed one, and as such deserves closer attention. The aim of the present study, which makes use of corpus linguistics tools to collect naturally-occurring data and discourse analysis framework to manually examine them, is twofold. Initially, it sets out to examine the linguistic contexts of jakby, which are assumed to furnish valuable guidelines for sifting out the prototypical uses of the word from the innovative ones. Next, the focus shifts onto indentifying context-sensitive functions of the latter in highly diversified stretches of discourse. The research findings demonstrate that the cotextual settings of the non-canonical jakby exhibit a number of distinctive characteristics, such as frequent co-occurrence of the word with pragmatic markers, reflexive discourse and unfilled pauses, all indicative of its relatively tenuous link with the neighbouring portions of text. As regards the functions of the unconventional jakby, the word emerges as a pragmatically multifunctional yet no longer hedging device, capable of, among others, facilitating floor-holding/-grabbing, helping to plan discourse, marking register clash and introducing elaboration on prior thought. Rich in pragmatic functions and syntactically more detached from the adjacent textual material than its canonical base, the investigated jakby appears to fit into the category of propositionally empty yet strategically salient pragmatic markers.

Building on the theoretical frameworks of frame and stance, this paper aims to demonstrate how play framing is manipulated in culturally meaningful contexts of Japanese conversations among friends and to show the consequences it brings to social life. This study particularly focuses on speech style shifts across speakers as one of the linguistic play-framing devices. The notion of “complementary stylistic resonance” as a special kind of pragmatic resonance is introduced to investigate how speech participants meta-linguistically signal their common stance of constructing a play frame. It was observed that in play they characteristically use the speech style of each imagined persona in a complementary social relationship such as “teacher and student,” “husband and wife,” and “American male and female in the dubbing register.” The ideologies of those dichotomized social roles are spontaneously evoked between the speakers through meta-language practice, resulting in solidifying their ideologies. Furthermore, in play, speech styles of those social roles are exaggerated and maximally contrasted within the pairs so that their identities are easily recognized by the speech partners to successfully co-construct the play at hand and to enhance its humorous effects. Although there may be a gap between ideology and reality, complementary stylistic resonance in play helps speech participants reconstruct their language ideologies of socially salient roles in local language practice, which serves as the concrete and dynamic ground for the process of recreating a larger cognitive and interactive dimension of culture.

This article explores the relationship between immigrant and non-immigrant Asian American youth identities and the use of language to manage this relationship. Focusing on everyday interactions at a high school in Texas, the analysis examines how fluent English-speaking Korean and Filipino American students draw on linguistic resources associated with Asian immigrants, thus attending to generational identity, an important, though often oversimplified, social dimension in transnational contexts. According to the present analysis, salient generational differences may exist between Asian American youth, yet their linguistic practices complicate simple binaries of opposition. Specifically, this article focuses on how fluent English-speaking students both accommodate toward and mock Asian immigrant speech and notes that these ostensibly divergent practices exhibit linguistic overlap. It is argued that the convergences and divergences of these practices can be productively examined by distinguishing between the levels of frame and ideology, thus explaining how speakers interpret Asian immigrant revoicings as accommodation, mocking, or, in some cases, an ambiguous linguistic act that hovers in between. Keywords: Accommodation; Asian Americans; Identity; Mocking; Stereotypes; Youth.

In an influential paper, Fox & Thompson (1990) argue that the grammar of relative clauses in spoken American English is affected by interactive and cognitive factors pertaining to the communication situation. Existential sentences containing a relative clause as well as an overt locative expression figure prominently in their analysis. The present paper examines Fox & Thompson’s analysis of such sentences in the light of a wide range of data. It is shown that the generalizations they make on the basis of their limited corpus (25 tokens) rest on false premises. Their analysis fails to take account of some of the most salient properties inherent in existential sentences in all varieties of English; it also disregards relevant cross-linguistic data. An alternative analysis is offered.

An echo answer is an answer that repeats elements of the question. This response form occurs after yes/noquestions and “statements about B-events”. The current study is based on data from native/non-native institutional interaction, and echo answers are shown to play an important role in certain types of repair that are characteristic of such interaction. Echo answers have two main usages. The first is to appropriate a candidate formulation and integrate it into one’s own turn in progress. This often happens when native speakers attempt to assist non-native interlocutors in expressing themselves. The other is to claim a strengthened commitment to the answer. This is especially salient in cases where a minimal agreement might project a potential dispreferred response. Echo answers may occur alone or with an initial or final response word, and these different response formats are shown to index the relative epistemic authority of the interlocutors.

This study re-examines Mandarin Placement verbs from a lexical-constructional perspective and redefines the class with semantic-to-syntactic properties pertaining to lexicalization patterns in Mandarin. It aims to show that Placement verbs lexicalize a cognitively salient causal chain that extends from an agentive motion to locational change and to resultant spatial configuration. The event chain serves as the conceptual basis for linking motion-triggered events and states that are syntactically distinct in profiling the three contingent stages: caused to move → caused to be → spatially grounded. Although English Placement verbs (put, hang, etc.) are typically taken to be exemplars of the caused-motion construction, this study shows that Placement verbs may be distinguished syntactically and semantically from pure Caused-Motion verbs and posture-based Spatial Configuration verbs. While the three classes of verbs may be viewed as demonstrating respectively the individuated stages of the proposed event chain, Placement verbs are the only class that encompasses all three event types in their meanings and are associated with a wide range of semantically compatible constructions. The three stages are discussed with graphical representations and collocational distinctions. Further sub-classifications of the Mandarin Placement verbs are provided with different semantic profiles for each subclass. Crucial to the analysis is the fact that location-profiled uses of Placement verbs outnumber path-profiled uses in Mandarin, indicating a categorical shift from motional to locational predication. By teasing out the language-specific and class-specific lexicalization patterns that are collo-constructionally definable, the study demonstrates the usefulness of a lexical-constructional approach in fine-tuning verbal semantic distinctions for cross-linguistic and cross-categorial comparisons.

This study demonstrates that two types of language ideologies (linguistic nationalism and feminine language normativity) influence how Japanese contemporary novels represent non-Japanese characters’ speech. It investigates the role of gender and observes that novelists only infrequently assign highly gendered utterance-final forms to non-Japanese characters when they speak in Japanese. This tendency is more salient among the representations of male non-Japanese characters. Masculine expressions seem to belong to a set of linguistic resources that are considered available only to the Japanese. This exclusivism, i.e., linguistic nationalism, might explain the lack of highly masculine forms among non-Japanese characters in novels. As for the relatively frequent assignment of gendered language for female characters, the normativity of feminine language makes it part of the basic language of all female speakers including non-Japanese individuals. In addition, feminine expressions are not as strongly associated with authenticity as masculine expressions.

In this paper, we explore how multiple accountability (MA) can enable an independent regulatory agency to deal with multiple conflicting public values in a complex and politically salient decision-making process. We examined the decision-making process of the Dutch National Health Care Institute on quality regulation of emergency care in the Netherlands. Using insights derived from ethnography, document analysis, and interviews, we show that MA resulted from strategic interactions between the Institute’s vertical and horizontal accountability forums. We argue that MA impeded efficiency but also enabled the Institute to deal with multiple conflicting public values.

Facial expression recognition has been studied broadly, and several works using local micro-pattern descriptors have obtained significant results. There are, however, open questions: how to design a discriminative and robust feature descriptor?, how to select expression-related most influential features?, and how to represent the face descriptor exploiting the most salient parts of the face? In this paper, we address these three issues to achieve better performance in recognizing facial expressions. First, we propose a new feature descriptor, namely Local Shape Pattern (LSP), that describes the local shape structure of a pixel's neighborhood based on the prominent directional information by analyzing the statistics of the neighborhood gradient, which allows it to be robust against subtle local noise and distortion. Furthermore, we propose a selection strategy for learning the influential codes being active in the expression affiliated changes by selecting them exhibiting statistical dominance and high spatial variance. Lastly, we learn the size of the salient facial blocks to represent the facial description with the notion that changes in expressions vary in size and location. We conduct person-independent experiments in existing datasets after combining above three proposals, and obtain an improved performance for the facial expression recognition task.

The use of depth information, acquired by depth sensors, for salient object detection (SOD) is being explored. Despite the remarkable results from recent deep learning approaches for RGB-D SOD, they fail to fully incorporate original and accurate information to express details of RGB-D images in salient objects. Here, we propose an RGB-D SOD model using a three-input multilevel fusion network (TMFNet), which differs from existing methods based on double-stream networks. In addition to RGB input (first input) and depth input (second input), the RGB image and depth map are combined into a four-channel representation (RGBD input) that constitutes the third input to the TMFNet. The RGBD input generates multilevel features that reflect details of the RGB-D image. In addition, the proposed TMFNet aggregates diverse region-based contextual information without discarding RGB and depth features. Thus, we introduce a cross-fusion module, and benefiting from rich low- and high-level information from salient features, feature fusion enables the improvement of localization of salient objects. The proposed TMFNet achieves state-of-the-art performance on six benchmark datasets for SOD.

Incipient faults in running gear systems corrupt the overall performance of high-speed trains, increasing the necessity of fault detection and diagnosis whose purpose is to maintain the safe and stable operation of high-speed trains. For this purpose, a novel data-driven method, that utilizes Hellinger distance and slow feature analysis, is proposed in this study. By integrating Hellinger distance into slow feature analysis, a new test statistic is defined for detecting incipient faults in running gear systems. Furthermore, the hidden Markov method is developed for performing reliable fault diagnosis tasks. The salient strengths of the proposed method lie in its satisfactory fault detectability on the one hand and the considerable robustness against high-level noises on the other hand. Finally, the effectiveness of the proposed method is verified through a numerical example and a running gear system of high-speed trains under actual working conditions.

RGB salient object detection (SOD) has made great progress. However, the performance of this single-modal salient object detection will be significantly decreased when encountering some challenging scenes, such as low light or darkness. To deal with the above challenges, thermal infrared (T) image is introduced into the salient object detection. This fused modal is called RGB-T salient object detection. To achieve deep mining of the unique characteristics of single modal and the full integration of cross-modality information, a novel Cross-Guided Fusion Network (CGFNet) for RGB-T salient object detection is proposed. Specifically, a Cross-Scale Alternate Guiding Fusion (CSAGF) module is proposed to mine the high-level semantic information and provide global context support. Subsequently, we design a Guidance Fusion Module (GFM) to achieve sufficient cross-modality fusion by using single modal as the main guidance and the other modal as auxiliary. Finally, the Cross-Guided Fusion Module (CGFM) is presented and serves as the main decoding block. And each decoding block is consists of two parts with two modalities information of each being the main guidance, i.e., cross-shared Cross-Level Enhancement (CLE) and Global Auxiliary Enhancement (GAE). The main difference between the two parts is that the GFM using different modalities as the main guide. The comprehensive experimental results prove that our method achieves better performance than the state-of-the-art salient detection methods. The source code has released at: https://github.com/wangjie 0825/CGFNet.git.

Code review is a common software engineering practice of practical importance to reduce software defects. Review today is often with the help of specialized tools, such as Gerrit. However, even in a tool-supported code review involves a significant amount of human effort to understand the code change, because the information required to inspect code changes may distribute across multiple files that reviewers are not familiar with. Code changes are often organized as commits for review. In this paper, we found that most of the commits contain a salient class(es), which is saliently modified and causes the modification of the rest classes in a commit. Our user studies confirmed that identifying the salient class in a commit can facilitate reviewers in understanding code change. Inspired by the effectiveness of machine learning techniques in the classification field, we model the salient class identification as a binary classification problem and extract a number of discriminative features from commit to characterize the salience of a class. The experiment results show that our approach achieves an accuracy of 88%. A user study with industrial developers shows that our approach can really improve the efficiency of reviewers understanding code changes in a reviewing scenario without using comment.

Behavioral interventions are an increasingly popular tool for encouraging socially desirable behavior. While they are expressly designed to seize people's attention, little consideration has been given to the costs of doing so. We estimate the consequences of seizing attention in the context of a widespread highway traffic safety campaign that displays roadside fatality counts on highway dynamic message signs (DMSs). We exploit detailed data on DMS and crash locations, DMS log files, and a unique setting in Texas where fatality messages are displayed only for one week each month. We find that this behavioral intervention significantly increases the number of traffic crashes. This increase is immediate, dissipates over 7 km, and increases with the displayed fatality count. Furthermore, drivers do not habituate to these messages, even after five years, and the effects do not persist beyond the treated weeks. Crashes increase statewide during treated weeks, inconsistent with any benefits. Our results show that behavioral interventions, designed to be salient, can crowd out more important considerations, causing interventions to backfire with costly consequences.

Abstract. According to previous research, the insula is important for processing salient and emotional stimuli, but its precise role remains elusive. By combining high spatial and temporal resolution, intracranial electroencephalography (iEEG) might contribute to filling this gap. Four drug-resistant epileptic patients with intracranial electrodes in the insula were instructed to watch and rate pictures with sexual content and neutral pictures. Event-related potentials (ERPs) were computed separately for both types of stimuli. Ninety-three percent of the anterior insula (AI) and 85% of the posterior insula (PI) contacts showed differences between ERPs. AI-positive deflections tended to have an earlier onset than PI-positive deflections. The results suggest that the AI generates a P300-like response and contributes to the early phase of the late positive potential, both components found enhanced while viewing emotional stimuli in the ERP literature. The present findings are interpreted as congruent with the role of the AI in maintaining attention to salient stimuli.

Adverse weather conditions seriously threaten the traffic safety, especially for rainy days with the ponding water on the road surface, which potentially result in vehicle crashes, person injuries and crash fatalities. Automatic splashed water detection based on surveillance videos is an attractive way to effectively prevent the traffic accidents. However, surveillance videos exhibit great variations with lighting changes, illumination conditions and complex backgrounds, which pose great difficulties in automatic recognition. In this paper, a novel deep learning based approach is proposed to detect the splashed water. To the best of our knowledge, this is the first work on this topic based on deep learning. An effective semantic segmentation network, called SWNet, is novelly proposed to extract the potential splashed water regions. An encoder-decoder structure is designed to capture the visual characteristics of splashed water. SWNet achieves high efficiency by reusing pooling indices and adopting the light-weight decoder. With the multi-scale feature fusion structure, SWNet integrates the coarse semantic information and detailed appearance information, which significantly boosts the accuracy and refines the edge segmentation. A weighted cross entropy loss for splashed water is adopted to cope with the unbalanced distribution between splashed water and backgrounds. Moreover, a splashed water attention module is designed to focus on the salient regions of moving vehicles and splashed water, by performing attention mechanism to integrate global contextual information in semantic segmentation. Experiments conducted on a newly collected splashed water dataset demonstrate the effectiveness and efficiency of the proposed approach, which outperforms the state-of-the-art methods.

In this paper, we survey state-of-the-art research outcomes in the burgeoning field of reconfigurable intelligent surface (RIS) in view of its potential for significant performance enhancement for next generation wireless communication networks by means of adapting the propagation environment. Emphasis has been placed on several aspects gating the commercially viability of a future network deployment. Comprehensive summaries are provided for practical hardware design considerations and broad implications of artificial intelligence techniques, so are in-depth outlooks on salient aspects of system models, use cases, and physical layer optimization techniques., 22 pages, 18 figures