Papers reviewed and determined not to be word norm studies. Use the flag icon to report errors or suggest re-inclusion.
16504 papers
Development of interpersonal relationships is a fundamental human motivation, and behaviors facilitating social bonding are prized. Some individuals experience enhanced reward from alcohol in social contexts and may be at heightened risk for developing and maintaining problematic drinking. We employed a 3 (group beverage condition) ×2 (genotype) design (N = 422) to test the moderating influence of the dopamine D4 receptor gene (DRD4 VNTR) polymorphism on the effects of alcohol on social bonding. A significant gene x environment interaction showed that carriers of at least one copy of the 7-repeat allele reported higher social bonding in the alcohol, relative to placebo or control conditions, whereas alcohol did not affect ratings of 7-absent allele carriers. Carriers of the 7-repeat allele were especially sensitive to alcohol's effects on social bonding. These data converge with other recent gene-environment interaction findings implicating the DRD4 polymorphism in the development of alcohol use disorders, and results suggest a specific pathway by which social factors may increase risk for problematic drinking among 7-repeat carriers. More generally, our findings highlight the potential utility of employing transdisciplinary methods that integrate genetic methodologies, social psychology, and addiction theory to improve theories of alcohol use and abuse.
Effects of a Protein Optimized Diet Combined with Moderate Resistance Training on the Postoperative Course in Older Patients with Hip Fracture
There is growing evidence that drugs of abuse alter processing of emotional information in ways that could be attractive to users. Our recent report that Δ⁹-tetrahydrocannabinol (THC) diminishes amygdalar activation in response to threat-related faces suggests that THC may modify evaluation of emotionally-salient, particularly negative or threatening, stimuli. In this study, we examined the effects of acute THC on evaluation of emotional images. Healthy volunteers received two doses of THC (7.5 and 15 mg; p.o.) and placebo across separate sessions before performing tasks assessing facial emotion recognition and emotional responses to pictures of emotional scenes. THC significantly impaired recognition of facial fear and anger, but it only marginally impaired recognition of sadness and happiness. The drug did not consistently affect ratings of emotional scenes. THC's effects on emotional evaluation were not clearly related to its mood-altering effects. These results support our previous work, and show that THC reduces perception of facial threat. Nevertheless, THC does not appear to positively bias evaluation of emotional stimuli in general.
We aimed to investigate whether fear of suffocation predicts healthy persons' respiratory and affective responses to obstructed breathing as evoked by inspiratory resistive loads. Participants (N = 27 women, ages between 18 and 21 years) completed the Fear of Suffocation scale and underwent 16 trials in which an inspiratory resistive load of 15 cmH(2)O/l/s (small) or 40 cmH(2)O/l/s (large) was added to the breathing circuit for 40 s. Fear of suffocation was associated with higher arousal ratings for both loads. Loaded breathing was associated with a decrease in minute ventilation, but progressively less so for participants scoring higher on fear of suffocation when breathing against the large load. The present findings document a potentially panicogenic mechanism that may maintain and worsen respiratory discomfort in persons with fear of suffocation.
Noun with genitive marker in Indo-Aryan language can variously be a child of a noun, a verb or a complex predicate, thus making it an important parsing issue. In this paper, we examine genitive data of Hindi and aim to automatically determine the attachment and relational label of the same in a dependency framework. We implement two approaches: a rule based approach and a statistical approach. The rule based approach produces promising results but fails to handle certain constructions because of its greedy selection. The statistical approach overcomes this by using a single candidate approach that considers all the possible candidates for the head and chooses the most probable candidate among them. Both approaches are applied on controlled and open environment data. A Controlled environment refers to the situation when the relational labels are attested to the input data except for the genitive data; while open environment refers to cases in which the input is only POS tagged and chunked. The rule based and statistical systems produce a high accuracy of 95 % and 97 % respectively for attachment and perform considerably well for labeling in controlled environment but poorly in open environment.
We address the issue of consuming heterogeneous annotation data for Chinese word segmentation and part-of-speech tagging. We empirically analyze the diversity between two representative corpora, i.e. Penn Chinese Treebank (CTB) and PKU’s People’s Daily (PPD), on manually mapped data, and show that their linguistic annotations are systematically different and highly compatible. The analysis is further exploited to improve processing accuracy by (1) integrating systems that are respectively trained on heterogeneous annotations to reduce the approximation error, and (2) re-training models with high quality automatically converted data to reduce the estimation error. Evaluation on the CTB and PPD data shows that our novel model achieves a relative error reduction of 11 % over the best reported result in the literature. 1
Chinese Treebank as a kind of valuable resource in Chinese Information Processing includes rich information on the sentence structure and the constitunent combination.A study on the combination of POS strings is the basic work for the effective use of treebank information.This paper investigates the ambiguous combination in Chinese Treebank,revealing that it largely requires semantic feature to resolve the ambiguous combination and structure in Chinese,and can not be solved simply by grammatical features(such as POS information) of words.
We propose Symbol-Refined Tree Substitution Grammars (SR-TSGs) for syntactic parsing. An SR-TSG is an extension of the conventional TSG model where each nonterminal symbol can be refined (subcategorized) to fit the training data. We aim to provide a unified model where TSG rules and symbol refinement are learned from training data in a fully automatic and consistent fashion. We present a novel probabilistic SR-TSG model based on the hierarchical Pitman-Yor Process to encode backoff smoothing from a fine-grained SR-TSG to simpler CFG rules, and develop an efficient training method based on Markov Chain Monte Carlo (MCMC) sampling. Our SR-TSG parser achieves an F1 score of 92.4% in the Wall Street Journal (WSJ) English Penn Treebank parsing task, which is a 7.7 point improvement over a conventional Bayesian TSG parser, and better than state-of-the-art discriminative reranking parsers. 1
Recent study shows that parsing accuracy can be largely improved by the joint optimization of part-of-speech (POS) tagging and dependency parsing. However, the POS tagging task does not benefit much from the joint framework. We argue that the fundamental reason behind is because the POS features are overwhelmed by the syntactic features during the joint optimization, and the joint models only prefer such POS tags that are favourable solely from the parsing viewpoint. To solve this issue, we propose a separately passive-aggressive learning algorithm (SPA), which is designed to separately update the POS features weights and the syntactic feature weights under the joint optimization framework. The proposed SPA is able to take advantage of previous joint optimization strategies to significantly improve the parsing accuracy, but also overcome their shortages to significantly boost the tagging accuracy by effectively solving the syntax-insensitive POS ambiguity issues. Experiments on the Chinese Penn Treebank 5.1 (CTB5) and the English Penn Treebank (PTB) demonstrate the effectiveness of our proposed methodology and empirically verify our observations as discussed above. We achieve the best tagging and parsing accuracies on both datasets, 94.60% in tagging accuracy and 81.67 % in parsing accuracy on CTB5, and 97.62 % and 93.52 % on PTB.
Este artículo describe investigación sobre los efectos de la desambiguación morfosintáctica usada como un preproceso de un analizador sintáctico profundo basado en\nHPSG, en el contexto del desarrollo de un treebank del español de código abierto, en el\nentorno de DELPH-IN. La anotación treebank se realiza manualmente tomando las decisiones\napropiadas entre las opciones propuestas por el sistema y ordenadas por un módulo\nestadístico. Los experimentos presentados muestran que el uso de un etiquetador reduce\nla ambigüedad de las frases, y contribuye a limitar la cantidad de frases cuyo análisis sobrepasa el límite de tiempo, y ayuda a al módulo estadístico a clasificar el árbol correcto entre los mejores. Por un lado, nuestros resultados validan los beneficios ya reportados en la literatura de tal preproceso de análisis profundo con respecto a la velocidad, cobertura y precisión. Por otro lado, proponemos una estrategia basada en existentes herramientas de código abierto y recursos para desarrollar con alta consitencia treebanks de sintaxis profunda\npara idiomas con limitada disponibilidad de recursos lingüísticos.
The Chinese comma signals the boundary of discourse units and also anchors discourse relations between adjacent text spans. In this work, we propose a discourse structureoriented classification of the comma that can be automatically extracted from the Chinese Treebank based on syntactic patterns. We then experimented with two supervised learning methods that automatically disambiguate the Chinese comma based on this classification. The first method integrates comma classification into parsing, and the second method adopts a “post-processing ” approach that extracts features from automatic parses to train a classifier. The experimental results show that the second approach compares favorably against the first approach. 1
We present two dependency parsers for Persian, MaltParser and MSTParser, trained on theUppsala PErsian Dependency Treebank. The treebank consists of 1,000 sentences today. Itsannotation scheme is based on Stanford Typed Dependencies (STD) extended for Persianwith regard to object marking and light verb contructions. The parsers and the treebank aredeveloped simultanously in a bootstrapping scenario. We evaluate the parsers by experimentingwith different feature settings. Parser accuracy is also evaluated on automatically generated andgold standard morphological features. Best parser performance is obtained when MaltParseris trained and optimized on 18,000 tokens, achieving 68.68% labeled and 74.81% unlabeledattachment scores, compared to 63.60% and 71.08% for labeled and unlabeled attachmentscore respectively by optimizing MSTParser.
In this thesis, an implementation of the MAT algorithm for query learning is improved and tested, specifically by attempted simulation of the MAT oracle using large corpora. In order to arrive at suitable test corpora, algorithms for generating positive and negative examples in relation to regular tree languages are developed.
BACKGROUND: Previous studies addressing teaching and learning in point-of-care ultrasound have primarily focussed on image interpretation and not on the technical quality of the images. We hypothesized that a limited intervention of 10 supervised examinations would improve the technical skills in Focus Assessed Transthoracic Echocardiography (FATE) and that physicians with no experience in FATE would quickly adopt technical skills allowing for image quality suitable for interpretation. METHODS: Twenty-one physicians with no previous training in FATE or echocardiography (Novices) participated in the study and a reference group of three examiners with more than 10 years of experience in echocardiography (Experts) was included. Novices received an initial theoretical and practical introduction (2 hours), after which baseline examinations were performed on two healthy volunteers. Subsequently all physicians were scheduled to a separate intervention day comprising ten supervised FATE examinations. For effect measurement a second examination (evaluation) of the same two healthy volunteers from the baseline examination was performed. RESULTS: At baseline 86% of images obtained by novices were suitable for interpretation, on evaluation this was 93% (p = 0.005). 100% of images obtained by experts were suitable for interpretation. Mean global image rating on baseline examinations was 70.2 (CI 68.0-72.4) and mean global image rating after intervention was 75.0 (CI 72.9-77.0), p = 0.0002. In comparison, mean global image rating in the expert group was 89.8 (CI 88.8-90.9). CONCLUSIONS: Improvement of technical skills in FATE can be achieved with a limited intervention and upon completion of intervention 93% of images achieved are suitable for clinical interpretation.
Cost-sensitive classification, where the features used in machine learning tasks have a cost, has been explored as a means of balancing knowledge against the expense of incrementally obtaining new features. We introduce a setting where humans engage in classification with incrementally revealed features: the collegiate trivia circuit. By providing the community with a web-based system to practice, we collected tens of thousands of implicit word-by-word ratings of how useful features are for eliciting correct answers. Observing humans' classification process, we improve the performance of a state-of-the art classifier. We also use the dataset to evaluate a system to compete in the incremental classification task through a reduction of reinforcement learning to classification. Our system learns when to answer a question, performing better than baselines and most human players.
Syntactic representations based on word-to-word dependencies have a long-standing tradition in descriptive linguistics, and receive considerable interest in many applications. Nevertheless, dependency syntax has remained something of an island from a formal point of view. Moreover, most formalisms available for dependency grammar are restricted to projective analyses, and thus not able to support natural accounts of phenomena such as wh-movement and cross–serial dependencies. In this article we present a formalism for non-projective dependency grammar in the framework of linear context-free rewriting systems. A characteristic property of our formalism is a close correspondence between the non-projectivity of the dependency trees admitted by a grammar on the one hand, and the parsing complexity of the grammar on the other. We show that parsing with unrestricted grammars is intractable. We therefore study two constraints on non-projectivity, block-degree and well-nestedness. Jointly, these two constraints define a class of “mildly” non-projective dependency grammars that can be parsed in polynomial time. An evaluation on five dependency treebanks shows that these grammars have a good coverage of empirical data.
There has recently been some interest among computational linguists in the task of inducing grammar-based "semantic parsers" from sets of paired strings and meaning representations, following pioneering work by Zettlemoyer and Collins (2005). Work of this kind is currently limited by the paucity of datasets for training. The talk reviews the state of the art in this field, then proposes a way to semi-automatically generate much larger language-independent datasets, on the same order of magnitude as syntactic treebanks, using linguistic knowledge that has only recently begun to become available, for use in inducing semantic parsers for under-resourced languages for application in statistical machine translation.
A critical task in Affective Computing is the reliable assessment of emotional states. The two most prominent approaches to classify emotions are categorical concepts of discrete emotions (e.g. OCC) and dimensional models typically using the pleasure - arousal - dominance space (PAD). In current research and applications, however, there is little overlap between these two concepts. A mapping of discrete categories into the dimensional space would offer new possibilities to model the emotional states of users and artificial agents, though. We hence let N=70 healthy subjects place the labels of discrete OCC emotions into PAD space according to their subjective knowledge with a simple visual tool. There was a high inter-subject consistency regarding the positioning of OCC emotions for the dimension of pleasure. However, arousal and dominance ratings showed considerably greater variance. We conclude that global and reliable mappings of OCC emotions into the PAD space can best be provided for the pleasure dimension. The exact positioning of discrete emotions regarding arousal and dominance can only be gained by individual calibration of a given user in a strict within-subject approach.
Freely available statistical parsers often require careful optimization to produce state-of-the-art results, which can be a non-trivial task especially for application developers who are not interested in parsing research for its own sake. We present MaltOptimizer, a freely available tool developed to facilitate parser optimization using the open-source system MaltParser, a data-driven parser-generator that can be used to train dependency parsers given treebank data. MaltParser offers a wide range of parameters for optimization, including nine different parsing algorithms, two different machine learning libraries (each with a number of different learners), and an expressive specification language that can be used to define arbitrarily rich feature models. MaltOptimizer is an interactive system that first performs an analysis of the training set in order to select a suitable starting point for optimization and then guides the user through the optimization of parsing algorithm, feature model, and learning algorithm. Empirical evaluation on data from the CoNLL 2006 and 2007 shared tasks on dependency parsing shows that MaltOptimizer consistently improves over the baseline of default settings and sometimes even surpasses the result of manual optimization.
The neuroendocrine state of new mothers may alter their neural processing of stressors in the environment through modulatory actions of oxytocin on the limbic system. We predicted that amygdala sensitivity to negatively arousing stimuli would be suppressed in postpartum compared to nulliparous women and that this suppression would be modulated by administration of oxytocin nasal spray. We measured brain activation (fMRI) and subjective arousal in response to negatively arousing pictures in 29 postpartum and 30 nulliparous women who received either oxytocin nasal spray or placebo before scanning. Pre- and post-exposure urinary cortisol levels were also measured. Postpartum women (placebo) demonstrated lower right amygdala activation in response to negative images, lower cortisol and lower negative photo arousal ratings to nulliparous women. Nulliparous women receiving oxytocin had lower right amygdala activation compared to placebo. Cortisol levels in the placebo group, and ratings of arousal across all women, were positively associated with right amygdala activation. Together, these findings demonstrate reductions in both amygdala activation and subjective negative arousal in untreated postpartum vs nulliparous women, supporting the hypothesis of an attenuated neural response to arousing stimuli in postpartum women. A causal role of oxytocin and the timing of potential effects require future investigation.
Based on real-text corpora with syntactic annotation, this study quantitatively addressed the following two questions: whether quantitative methods and indexes can point to the diachronic syntactic drifts characterizing the evolution from Latin to Romance languages and whether these methods and indexes can provide evidence to evince the shared syntactic features among Romance languages and define them as a distinctive language subgroup. Our study shows that the distributions of dependency directions are suggestive of positive answers to the above two questions. In addition, the dependency syntactic networks extracted from the dependency treebanks reflect the degree of inflectional variation of a language, and the clustering analysis shows that these parameters, in spite of some imperfections, can also help differentiate Romance languages from Latin diachronically and from other languages synchronically.
Comunicació presentada al 8th International Conference on Language Resources and Evaluation (LREC'12), celebrat del 21 al 27 de maig de 2012 a Istanbul, Turquia.
In the present study, we investigated the effects of empathic paraphrasing as an extrinsic emotion regulation technique in social conflict. We hypothesized that negative emotions elicited by social conflict can be regulated extrinsically in a conversation by a listener following the narrator's perspective and verbally expressing cognitive empathy. Twenty participants were interviewed on an ongoing or recently self-experienced social conflict. The interviewer utilized 10 standardized open questions inviting participants to describe their perception of the conflict. After each of the 10 descriptions, the interviewer responded by either paraphrasing or taking notes (control condition). Valence ratings pertaining to the current emotional state were assessed during the interview along with psychophysiological and voice recordings. Participants reported feeling less negative after hearing the interviewer paraphrase what they had said. In addition, we found a lower sound intensity of participants' voices when answering to questions following a paraphrase. At the physiological level, skin conductance response, as well as heart rate, were higher during paraphrasing than during taking notes, while blood volume pulse amplitude was lower during paraphrasing, indicating higher autonomic arousal. The results show that demonstrating cognitive empathy through paraphrasing can extrinsically regulate negative emotion on a short-term basis. Paraphrasing led to enhanced autonomic activation in recipients, while at the same time influencing emotional valence in the direction of feeling better. A possible explanation for these results is that being treated in an empathic manner may stimulate a more intense emotion processing helping to transform and resolve the conflict.
Compared with nonarousing stimuli, arousing stimuli enhance memory performance. The most robust effects have been reported for negative stimuli, "the negativity effect," although a number of mediating factors prevent definitive conclusions, for example, age, gender, memory task, retention period, and alternative arousal measures. To clarify whether the negativity effect is robust across age, gender, and time, we studied incidental recognition of neutral and negative pictures from the International Affective Picture System (Lang, Bradley, & Cuthbert, 1999) in healthy younger and older adults--women and men--after a 1-year retention interval. Memory performance was related to 2 arousal measures at encoding, skin conductance response (SCR), and intensity rating of unpleasantness. The results showed weaker overall memory performance for older adults compared with younger adults. The negativity effect on accuracy (d') was gender dependent and age independent. In contrast, the negativity effect on response bias (c) interacted with age, but not gender, being weaker for older adults. Despite significant differences in arousal (SCR and arousal rating) between negative and neutral pictures, the correlations between arousal measures and memory performance were weak. Controlling for age and gender, a small negative partial correlation was found between arousal ratings and accuracy. The results extend previous studies by relating long-term recognition to both age and gender as well as to arousal at encoding.
The treatment of the individual database entries is focused principally on the exemplary part (exemplification), which in the Pralex LDB has a specifically defined structure (classification into exemplification subblocks) for each word class. With nouns, we have focused on a detailed description of the syntactic and semantic collocablity. What is important for us in the treatment is the selection of relevant corpus evidence, i.e. examples of the specific usage for the meanings pre-established in the Standard Czech Language (SSJČ), as well as capturing the evidence for new meanings not listed in that dictionary. We illustrate the nouns primarily by uncited evidence, i.e. exemplary collocations in the form of syntagmata, which represent the common collocations of the treated nouns; to a lesser extent we present also sentential citation evidence, i.e. with the citation of the source.
We study constituent parsing of German, a morphologically rich and less-configurational language. We use a probabilistic context-free grammar treebank grammar that has been adapted to the morphologically rich properties of German by markovization and special features added to its productions. We evaluate the impact of adding lexical knowledge. Then we examine both monolingual and bilingual approaches to parse reranking. Our reranking parser is the new state of the art in constituency parsing of the TIGER Treebank. We perform an analysis, concluding with lessons learned, which apply to parsing other morphologically rich and less-configurational languages.
Evaluative conditioning (EC) is the valence change of a stimulus (conditioned stimulus, CS) that is due to the previous pairing with another stimulus (unconditioned stimulus, US). We investigated whether EC can occur also when the CS-US pairings are not experienced directly by the participant but are implied by other events that the participant encounters. In two experiments, positive USs were presented in some trials and negative USs in other trials. Afterwards, participants were given information from which it was possible to conclude that CSs were covertly present during these trials. Finally, the valence of these CSs was registered using both implicit (Implicit Association Test, affective priming) and explicit measures (valence ratings). In line with the assumption that EC effects can be based on CS-US pairings that are not directly experienced, the valence of the CSs changed in the direction of the US with which they were covertly paired. This effect was observed both on explicit and on implicit measures. We argue that several aspects of our results are in line with propositional models of EC and fit less well with association formation models.
Attachment theory describes how people emotionally bond with others, utilize their social interactions to regulate affective needs, and how they differ in their attachment style. However, it remains unknown whether anxious or avoidant attachment is linked to more general differences in emotional processing for negative and positive stimuli, and whether such differences depend on stimulus content. Here we tested how social or non-social positive or negative emotional scenes were rated for pleasantness, arousal, and control, as a function of individual attachment style, in a sample of 54 female participants. Our results show that avoidant attachment was associated with a selective reduction of pleasantness ratings for positive social images, whereas anxious attachment was associated with higher arousal and lower control ratings for negative social images, besides higher arousal ratings for all remaining stimulus categories. These findings reveal that adult attachment style is associated with differences in the perception of emotion-laden stimuli, even when unfamiliar and not directly attachment-related, and such differences may also affect positive scenes, particularly when they contain social information, rather than just socially negative or threatening information. In addition, our results support the notion that anxious attachment is not only associated with hyperactivating tendency during the appraisal of social threat, but may also involve an ambivalence influencing the judgments of both positive and negative information.
The so-called Jehoash Inscription has been the subject of academic and even juridical contention. With respect to the so-called Jehoash Inscription, the most basic outlook that a philologist must adopt, or belief the philologist must form, at the very outset is whether to treat the text as presumably authentic or presumably inauthentic or suspect. The methods of analysis and evaluation that one employs will follow from that initial decision. This chapter reviews a few of the most outstanding deviations from literary and linguistic norms associated with royal building and memorial inscriptions that one encounters in the Jehoash text. It concludes that there is no single, simple, economical way of explaining the abnormalities of the so-called Jehoash Inscription and presents some of the facts concerning the language and form of the Jehoash Inscription. Keywords: authenticity; Jehoash Inscription; linguistic norms; memorial inscriptions
Sentiment analysis has been extensively investigated during the last years mainly for English language. Currently, existing approaches can be split into two main groups: methods based on the combination of lexical resources and Natural Language Processing (NLP) techniques; and machine learning approaches. This paper introduces the use of lexical databases for Sentiment Analysis of user reviews in Spanish for the accommodation and food and beverage sectors. A global sentiment score has been calculate based on the negative and positive words which appear in the review and using the mentioned lexicon database. The algorithm has been tested with short users online reviews acquired from TripAdvisor.
Previous work on treebank parsing with discontinuous constituents using Linear Context-Free Rewriting systems (LCFRS) has been limited to sentences of up to 30 words, for reasons of computational complexity. There have been some results on binarizing an LCFRS in a manner that minimizes parsing complexity, but the present work shows that parsing long sentences with such an optimally binarized grammar remains infeasible. Instead, we introduce a technique which removes this length restriction, while maintaining a respectable accuracy. The resulting parser has been applied to a discontinuous treebank with favorable results. 1
We examine some of the frequently disregarded subtleties of tokenization in Penn Treebank style, and present a new rule-based preprocessing toolkit that not only reproduces the Treebank tokenization with unmatched accuracy, but also maintains exact stand-off pointers to the original text and allows flexible configuration to diverse use cases (e.g. to genreor domain-specific idiosyncrasies). 1 Introduction—Motivation The task of tokenization is hardly counted among the grand challenges of NLP and is conventionally interpreted as breaking up “natural language text [...] into distinct meaningful units (or tokens) ” (Kaplan,
A leading hypothesis to explain the social dysfunction in people with autism spectrum disorders (ASD) is that they exhibit a deficit in reward processing and motivation specific to social stimuli. However, there have been few direct tests of this hypothesis to date. Here we used an instrumental reward learning task that contrasted learning with social rewards (pictures of positive and negative faces) against learning with monetary reward (winning and losing money). The two tasks were structurally identical except for the type of reward, permitting direct comparisons. We tested 10 high-functioning people with ASD (7M, 3F) and 10 healthy controls who were matched on gender, age, and education. We found no significant differences between the two groups in terms of overall ability behaviorally to discriminate positive from negative slot machines, reaction-times, and valence ratings, However, there was a specific impairment in the ASD group in learning to choose social rewards, compared to monetary rewards: they had a significantly lower cumulative number of choices of the most rewarding social slot machine, and had a significantly slower initial learning rate for the socially rewarding slot machine, compared to the controls. The findings show a deficit in reward learning in ASD that is greater for social rewards than for monetary rewards, and support the hypothesis of a disproportionate impairment in social reward processing in ASD.
The annotations of explicit and implicit discourse connectives in the Penn Discourse Treebank make it possible to investigate on a large scale how different types of discourse relations are expressed. Assuming an account of the Uniform Information Density hypothesis, we expect that discourse relations should be expressed explicitly with a discourse connector when they are unexpected, but may be implicit when the discourse relation can be anticipated. We investigate whether discourse relations which have been argued to be expected by the comprehender exhibit a higher ratio of implicit connectors. We find support for two hypotheses put forth in previous research which suggest that continuous and causal relations are presupposed by language users when processing consecutive sentences in a text. We then proceed to analyze the effect of Implicit Causality (IC) verbs (which have been argued to raise an expectation for an explanation) as a local cue for an upcoming causal relation.