Multimodal Integration in Statistical Learning: Evidence from the McGurk Illusion

Recent advances in the field of statistical learning have established that learners are able to track regularities of multimodal stimuli, yet it is unknown whether the statistical computations are performed on integrated representations or on separate, unimodal representations. In the present study, we investigated the ability of adults to integrate audio and visual input during statistical learning. We presented learners with a speech stream synchronized with a video of a speaker's face. In the critical condition, the visual (e.g., /gi/) and auditory (e.g., /mi/) signals were occasionally incongruent, which we predicted would produce the McGurk illusion, resulting in the perception of an audiovisual syllable (e.g., /ni/). In this way, we used the McGurk illusion to manipulate the underlying statistical structure of the speech streams, such that perception of these illusory syllables facilitated participants' ability to segment the speech stream. Our results therefore demonstrate that participants can integrate audio and visual input to perceive the McGurk illusion during statistical learning. We interpret our findings as support for modality-interactive accounts of statistical learning.

Linguistic Stylistics

Marina Katnic-Bakarsic. Linguistic Stylistics The practical, i.e. educational, objective of this research was to produce lectures on linguistic stylistics for the students of Sarajevo University, while the theoretical one was to produce a monograph on the subject. This monograph, which can also be used as a university textbook, includes twenty-nine chapters, an index of topics, a bibliography and a list of sources. The theoretical postulates are followed by examples from texts in various functional styles in Bosnian, Croatian, Serbian, and in some cases Russian or English. Linguo-stylistic problems were investigated from both the structuralist and post-structuralist points of view. Linguistic stylistics is therefore understood as a discipline which studies expressive, stylistically marked language units on all language levels, functional-stylistic language variation and various aspects of intertextuality and metatext. The author introduces a notion of stylistic competence. The stylistic competence of a speaker is directly proportional to his/her knowledge of different varieties of language (i.e. subcodes) and to the successful switching from one subcode to another. Stylistic creativity is a special segment of stylistic competence as a feature of individual style. A new classification of functional styles has also been introduced. This includes six primary styles (scientific, colloquial, administrative, publicistic, journalistic and literary-artistic) and five secondary styles (oratorical, the style of advertisements and commercials, that of comics, that of essays and that of screenplays). A special place is given to the analysis of the style of hypertext and hypermedia, which can be understood only within the post-structuralist theory of text deconstruction and intertextuality. The project also analysed some new topics, including reregistration in literary texts, gender and style of dialogue, and citations as metatextual signals and their role in different types of text. The results therefore offer a new approach to the study of linguistic stylistics both in Bosnia and Herzegovina and in the field in general.

Offline grammar-based recognition of handwritten sentences

This paper proposes a sequential coupling of a Hidden Markov Model (HMM) recognizer for offline handwritten English sentences with a probabilistic bottom-up chart parser using Stochastic Context-Free Grammars (SCFG) extracted from a text corpus. Based on extensive experiments, we conclude that syntax analysis helps to improve recognition rates significantly.

106th Commencement

Video of Clark University's 106th Commencement. Commencement speaker was Alan Khazei.

Opening Keynote: History, Politics, and Education: Teaching about Holocaust and Genocide

Format: 5 minute introduction, 20 min per speaker, 30 minute discussion Moderator: Debórah Dwork, Strassler Center, Clark University

Panel One: Formal Education and Non-formal Education

Format: 5 minute introduction, 15 min per speaker, 70 min discussion Moderator: Johanna Vollhardt, Clark University

Panel Four: Politics and Policy

Format: 5 minute introduction, 15 min per speaker, 70 minute discussion Moderator: Shelly Tenenbaum, Clark University

Panel Five: Looking ahead

Format: 5 minute introduction , 7 min per speaker, 1 hour discussion Moderator: Ken MacLean, Clark University Debόrah Dwork, Strassler Center, Clark UniversityThomas Kühne, Strassler Center, Clark UniversityTaner Akçam, Strassler Center, Clark University

Panel Three: Challenges and Best Practice

Format: 5 minute introduction, 15 min per speaker, 70 minute discussion Moderator: Raphael Rogers, Clark University

Panel Two: Content Orientation versus Civic Education

Format: 5 minute introduction, 15 min per speaker, 70 minute discussion Moderator: Eric DeMeulenaere, Clark University

Monotone Constrained Tensor-product B-spline with application to screening studies

When different markers are responsive to different aspects of a disease, combination of multiple markers could provide a better screening test for early detection. It is also resonable to assume that the risk of disease changes smoothly as the biomarker values change and the change in risk is monotone with respect to each biomarker. In this paper, we propose a boundary constrained tensor-product B-spline method to estimate the risk of disease by maximizing a penalized likelihood. To choose the optimal amount of smoothing, two scores are proposed which are extensions of the GCV score (O'Sullivan et al. (1986)) and the GACV score (Ziang and Wahba (1996)) to incorporate linear constraints. Simulation studies are carried out to investigate the performance of the proposed estimator and the selection scores. In addidtion, sensitivities and specificities based ona pproximate leave-one-out estimates are proposed to generate more realisitc ROC curves. Data from a pancreatic cancer study is used for illustration.

A BAYESIAN HIERARCHICAL MODEL FOR CONSTRAINED DISTRIBUTED LAG FUNCTIONS: ESTIMATING THE TIME COURSE OF HOSPITALIZATION ASSOCIATED WITH AIR POLLUTION EXPOSURE

Numerous time series studies have provided strong evidence of an association between increased levels of ambient air pollution and increased levels of hospital admissions, typically at 0, 1, or 2 days after an air pollution episode. An important research aim is to extend existing statistical models so that a more detailed understanding of the time course of hospitalization after exposure to air pollution can be obtained. Information about this time course, combined with prior knowledge about biological mechanisms, could provide the basis for hypotheses concerning the mechanism by which air pollution causes disease. Previous studies have identified two important methodological questions: (1) How can we estimate the shape of the distributed lag between increased air pollution exposure and increased mortality or morbidity? and (2) How should we estimate the cumulative population health risk from short-term exposure to air pollution? Distributed lag models are appropriate tools for estimating air pollution health effects that may be spread over several days. However, estimation for distributed lag models in air pollution and health applications is hampered by the substantial noise in the data and the inherently weak signal that is the target of investigation. We introduce an hierarchical Bayesian distributed lag model that incorporates prior information about the time course of pollution effects and combines information across multiple locations. The model has a connection to penalized spline smoothing using a special type of penalty matrix. We apply the model to estimating the distributed lag between exposure to particulate matter air pollution and hospitalization for cardiovascular and respiratory disease using data from a large United States air pollution and hospitalization database of Medicare enrollees in 94 counties covering the years 1999-2002.