L’intégration des nouvelles technologies en enseignement supérieur ne devrait pas être vue comme un simple effet de mode, mais plutôt comme un cadre de réflexion sur les orientations nouvelles à donner à l’Université du 21ème siècle, en termes de culture institutionnelle, de ressources organisationnelles et d’opérationnalisation d’objectifs de recherche et d’enseignement (Duderstadt et al, 2002 ; Guri-Rosenblit, 2005). En effet, l’émergence d’une nouvelle tendance de formation post-secondaire, qu’est l’intégration des solutions virtuelles au présentiel, n’est pas sans incidences sur les pratiques enseignantes. Or, la littérature n’offre pas de modèles descriptifs satisfaisants permettant de mieux saisir la pertinence des liens entre l’enseignement en contexte d’apprentissage hybride et le développement professionnel des enseignants universitaires. Aussi avons-nous procédé par des observations participantes de deux cours en sciences de la gestion à HEC Montréal, dans les programmes de certificat de premier cycle ainsi que par des entretiens d’explicitation et d’autoconfrontation, pendant toute la session d’automne 2014, pour respectivement recueillir des données sur les pratiques réelles d’un maître d’enseignement, en sociologie du travail, et d’une chargée de cours, en gestion de projets. Notre analyse du corpus, par catégorisations conceptualisantes, a permis la production d’énoncés nomologiques rendant compte de la dynamique de relations entre ces deux phénomènes. Sur le plan scientifique, elle a apporté un éclairage nouveau sur les processus de construction identitaire professionnelle en pédagogie universitaire, en regard des mutations technologiques, socioculturelles et économiques que subissent l’Université, en général, et les pratiques enseignantes et étudiantes, en particulier. L’approche inductive utilisée a donc permis de définir la structure des interactions des deux phénomènes, selon la perspective des deux enseignants, et d’élaborer des modèles d’intervention enracinés dans leurs pratiques quotidiennes. Aussi sur le plan social, ces modèles sont-ils l’expression d’une grammaire de la pensée et de l’action, ancrée dans les valeurs des enseignants eux-mêmes. Nous avons pris en compte le paradigme de la tâche réelle, versus celui de la tâche prescrite, en termes de mise en œuvre concrète des processus pédagogiques, pour rendre les résultats de cette recherche signifiants pour la pratique. Les modèles, qui ont émergé de notre configuration de la pensée dialogique des participants, peuvent être intégrés à la formation des enseignants universitaires en contexte de bimodalisation de l’Université. / The integration of new technologies in higher education should not be seen as a mere fad, but rather a framework for redesigning the new orientations to be given to the University of the 21st century, in terms of institutional culture, organizational resources and achieving the objectives of research and teaching programs. Indeed, the emergence of a new trend of higher education, that is blended learning, has impacts on the pedagogical practices. Yet, there are no satisfactory models in the literature that describes the link pertaining to the relationship between teaching in the context of blended learning and faculty development. Throughout the whole semester of fall 2014, we collected data, by the means of participant observations and self-confrontation interviews, on the teaching activities of one full-time lecturer, in sociology of work, and one part-time lecturer, in project management, with one undergraduate class of each lecturer in the programs of certificate at HEC Montreal. The data analysis, using the conceptual categorizations, brought about theoretical statements underpinning the dynamics of the relationships between the phenomena of blended learning and faculty development. Scientifically, our research has shed a new light on the processes of professional identity construction in faculty development, regarding the technological, sociocultural and economic changes that the University, in general, and, especially, the teaching and learning practices undergo. Our inductive approach has helped to elucidate the structure of the interactions of the two phenomena, from the point of view of the two lecturers, and to construct models of pedagogical practices rooted in their daily teachings. Therefore, socially, these models translate a grammar of thought and action, anchored in the very values of the lecturers. We took into consideration the paradigm of the actual task, versus that of the prescribed task, in terms of implementation of the pedagogical processes, so that the results of this study be meaningful to the practice. The emerging models, from our redesign of the dialogical thinking processes of the two lecturers, can be used for the training of the faculty integrating online solutions to face-face instructions.

The constitution of the Republic of South Africa has mandated government to keep in touch with the citizens, by regularly updating them on the services available for the improvement of their lives, and further to constantly report back on progress made in implementing government policies. In recent years, the country has been experiencing sporadic service delivery protests especially at the local sphere of government (municipal level). The aim of the study was therefore to examine the success of the communication and information system of the South African government in disseminating information to citizens. The area of the study was the Province of Mpumalanga which is one (1) of the nine (9) provinces constituting the Republic of South Africa. For the study, quantitative research method in the form of a survey was adopted, the hypothesis tested and the findings generalised. In creating a sample frame for the study, both probability and non-probability sampling techniques were adopted. The type of data collection instrument adopted was a structured questionnaire. Single-stage cluster sampling was adopted for heads of government communication whereas judgemental or purposive sampling technique was adopted for citizens. Data was analysed utilising data tabulation, descriptives, and data disaggregation quantitative data analysis procedures. The key findings of the study, amongst others, are: (a) That the current communication and information system being utilised by the South African government has an impact (it has improved communication with citizens); (b) That the citizens prefer government to communicate with them in their own indigenous languages (communicating in all eleven (11) official languages); (c) That the citizens prefer face-to-face communication; (d) That citizens would like to receive government messages via social media (given its immediacy and interactivity); and (e) That the low status of the communication function at the local government level denies communicators access to information to be communicated. Amongst others, the study recommends that (1) communication should be recognised as a strategic management function in all spheres of government; (2) the use of radio as the primary channel of information should be strengthened; (3) face-to-face communication with citizens should be strengthened; (4) information dissemination should be done in all eleven (11) official languages, including in Braille and sign-language; (5) social media and SMS notification should be included in the government communication policy as one of its official information dissemination channels. / Communication Science / M. A. (Communication)

「腦力激盪技術」(brainstorming)是 Osborn(1957)所提出的思考或開會技巧，並且被廣泛地使用，特別是用於進行團體創造。但晚進關於腦力激盪的研究發現；傳統腦力激盪團體（真團體，面對面接觸、互動式團體）的創造力表現不如名義式腦力激盪團體（假團體，結合個別腦力激盪成果）的原因之一是：受試者在真實的團體中會害怕被評價，因此，Camacho(1991)探討評價恐懼(communication apprehension) 與腦力激盪形式的交互影響，而 Dennis 和 Valacich (1993)則設計了一個避免評價恐懼的實驗，將電子暨運用於腦力激盪團體，稱之為「電子腦力激盪」(electronic brainstorming)。 本研究主要是綜合 Camacho (1991)與 Dennis 和 Valacich (1993)的研究為基礎，以 312 名大學生為研究對象，進行 2 （高、低溝通恐懼）×2（互動式、名義式腦力激盪）×2（電子、非電子）之三因子受試者問實驗設計，探討溝通恐懼、腦力激盪形式、採用電子腦力激盪技術與否對於團體創造力與成員對團體歷程感受的交互作用之影響。經分析後所得到的結果如下： 1.當 4 人一組時，低溝通恐懼組的團體創造力顯著高於高溝通恐懼組，且對於團體歷程的感受有：自在、放鬆、興奮，較喜愛在團體中進行腦力激盪，認為其他成員可以刺激自己想出更多的觀念等。但高溝通恐懼者對於團體歷程的感受則有：擔心、煩躁不安、焦慮，較喜愛獨自進行腦力激盪，且在意其他成員認為自己的觀念是「愚蠢的」等。 2.電子式腦力激盪的團體的創造力顯著高於非電子組，且電子組在互動歷程中較機警、有壓力要和別人想出的觀念一樣多，較容易擔心、焦慮。 3.名義式與互動式腦力激盪的團體創造力並沒有顯著不同，但成員對於團體歷程的感受卻有明顯差異。受試者在互動式腦力激盪團體中會有：緊張、煩躁不安、焦慮、不自在、怕難為情等感受，且容易因為擔心其他成員給予自己負面的評價，因而不敢把自己的想法表達出來。相對地：名義式腦力激盪組的受試者對於團體則有機警、鎮定、放鬆、隨時覺察到自己、興奮等感受。 4.不論受試者溝通恐懼的程度如何，採用電子腦力激盪技術時，互動式與名義式腦力激盪組的團體創造力並沒有顯著差異，但互動組顯著出現緊張、煩躁不安、不自在等感受。而當不採用電子技術時，名義式的團體創造力則顯著高於互動式腦力激盪組。 5.對低溝通恐懼者而言，使用電子腦力激盪時，無論腦力激盪形式（互動式或名義式）為何，團體創造力都顯著地較未採用電子技術時好，且成員對於在團體歷程中較容易隨時覺察到自己。而對高溝通恐懼者而言，雖然使用電子技術時，受試者容易對團體歷程產生：緊張、煩躁不安、焦慮、不自在、怕難為情等感受，但只有採取「電子--互動式」腦力激盪時，才能提昇高溝通恐懼受試者的團體創造力。

Online learning has grown exponentially within higher education in the past decade, especially at community colleges. As online course offerings expand community colleges need to assess student learning in order to ensure quality learning experiences for students and for accreditation purposes. The purpose of this study was to compare the assessment methods of community college faculty who teach the same course in both face-to-face and online formats, in order to determine the similarities and differences that exist in assessment practices. This study stemmed from the constructionist and pragmatic perspectives of the researcher. Additionally this study asked how community college faculty perceive the effectiveness of different assessment methods on student learning. The method for this research was a qualitative multiple case study in which community college accounting faculty were interviewed to explore how they assess student learning in both the online and face-to-face formats. Homework problems, exams, quizzes were the most commonly used methods of assessment of student learning in both formats by the study participants. Furthermore, exams and quizzes were believed to be most effective in assessing student learning, among the community college faculty members interviewed. Another discovery from this study included the prevalent use of publisher's software,called My Accounting Lab, among community college accounting faculty in administering assessments to students. Additional concerns around assessing online students effectively, managing online student group work, and mitigating online student cheating were also revealed during interviews with study participants. / Graduation date: 2013

La tesi fornisce uno studio empirico relativo agli elementi linguistici caratterizzanti il parlato faccia-a-faccia e il parlato filmico americano, due domini conversazionali solitamente detti differire in termini di spontaneità, essendo il primo generalmente descritto come la quintessenza del linguaggio parlato (in quanto totalmente spontaneo) e il secondo come non-spontaneo (essendo scritto-per-essere-parlato) e, quindi, non adatto a rappresentare l'uso generale della conversazione. Entrambe le analisi (i.e. quella multi-dimensionale, che offre una panoramica generale dei due domini presi in considerazione, e quella più specifica relativa al comportamento linguistico dell’espressione you know) basate su esempi autentici tratti da corpora dimostrano che, nonostante quanto venga generalmente descritto dalla letteratura a riguardo, conversazione faccia-a-faccia e conversazione filmica hanno molti tratti in comune e confutano l’idea che il linguaggio filmico non possa essere rappresentativo dell'uso generale della conversazione. / The present dissertation examines empirically the linguistic features characterizing American face-to-face and movie conversation, two domains which are usually claimed to differ especially in terms of spontaneity. Natural conversation is, indeed, considered the quintessence of the spoken language for it is totally spontaneous, whereas movie conversation is usually described as non-spontaneous, being artificially written-to-be spoken and, thus, not likely to represent the general usage of conversation. In spite of what is generally maintained by the literature, both the Multi-Dimensional analysis and the micro-analysis of the functions of you know based on authentic data retrieved from corpora show that the two conversational domains do not differ to a great extent and thus confutes the claim that movie language has “a very limited value” in that it does not reflect natural conversation and, consequently, is “not likely to be representative of the general usage of conversation”.

This study explores the training of Psychology tutors at the University of South Africa's (UNISA) learning centres with a view to providing a paradigm shift in tutor training and contributing towards new ways of thinking about education and training in South Africa. It furthermore opens up alternative ways of thinking about the process of facilitation in general. Although the study focuses on tutor training within the discipline of Psychology, the outcomes of the study are applicable to tutor training and tutoring in general, and make a valuable contribution in identifying the basic principles underlying tutor training in other disciplines as well. An ecosystemic approach encourages the trainer-trainee system to become aware of how they affect each other's thinking and behaviour. The basic principles underlying the process of facilitation that emerge in the study, therefore apply equally to the training of tutors and the tutoring of learners by others. These principles provide guidelines in terms of the processes involved in tutor development, as well as inform the issues and themes around which the sharing of ideas can be choreographed in the field of tutor training and tutoring. In conclusion, the information generated in the study focuses specifically on student support in the form of face-to-face tutorials in the distance education setting, the form of tertiary education which is most accessible to many historically disadvantaged learners in South Africa today. More generally, it will make a valuable contribution in the current educational context in South Africa, where student support is of the utmost importance in striving for equality in education, and where the focus is on outcomes-based education. / Curriculum and Instructional Studies / D.Litt. et Phil. (Psychology)

Distance education (DE) is now recognised as an education system independent from conventional face-to-face education. It has self-governing theories and pedagogies. It attracts students with unique characteristics different from those of on-campus students. The present distance learning student differs from the past ones by characteristics, needs and contexts. Not all students registering for distance learning conform to the characteristics of distance students described in theory. It is now acknowledged that DE systems demand special skills including time management, self-regulation and independent learning skills. Yet, few of these students enter into distance learning with prior experiences on its demands. The new student is compelled to learn to deal with challenges that come with DE i.e. the impact of ‘distance’, isolation and time management. ‘The net generation’ and ‘digital natives’ are now in college but with skills not automatically transferable to learning technologies. Information and communication technologies (ICT) providers are mostly focused on the ‘use’ rather than the ‘user’. Universities are continuously adopting new technologies leaving the student bewildered as to the focus; learning or technology training. The internet has ‘everything’; open course ware (OCW), open education resources (OERS), wikis and all web information. Students cannot simply find things for themselves. Furthermore distance learning has no policy on how to engage with the internet and students are left to decide what, which and how much is required for any level of study. Most universities in Africa moving from single to dual mode have not integrated distance learning pedagogy which requires restructuring in the organisation, policy and course development. DE, though spanning over two centuries has been mutually dependent on technology. The present technology demands a paradigm shift from that of correspondence days. These issues have created the need for support strategies that can literally accompany the DE student throughout his/her academic journey. Universities have established DE units, campuses and schools for a variety of reasons. It is required that such universities provide learner support systems for their students. The purpose of this study was to assess the learner support services available for distance learning undergraduate students in two universities in Kenya i.e. Northern University (NU) and Western University (WU). A Learner support system can comprise of numerous components. In this study, nine (9) components/indices were tested as the indices for providing support services. These are:- registration procedures, orientation programme and skills training, technology and learning materials, counselling and mentorship, interactions and communication, feedback, regional centres and library, students association and representation and course progression and satisfaction. This study employed an evaluation research design utilising both quantitative and qualitative methods. Online questionnaires were used for quantitative data collection. For qualitative data collection two (2) instruments were used; an interview schedule for key programme implementers and a documentary analysis tool for documents and websites. The findings indicated that the main indices that distinguished the two universities were registration process, technology and learning materials, counselling/mentorship and regional centres where the t-test showed significant differences. The p values were 0.008, 0.012, 0.036 and 0.015 respectively at 0.05 significance level. In all of them, Northern University (NU) had a relatively high mean score than Western University (WU) except for the index on counselling and mentorship. / Educational Foundations / D. Ed. (Socio-Education)

Syftet med studien är att presentera ett designkoncept för publika diskussioner på nätet med målet att uppnå djupare diskussioner och minska mängden information overload. Arbetet använder metoden Concept Driven Interaction Design (CDID) som går ut på att skapa ett designkoncept utifrån flera olika teorier som appliceras genom interaktionsdesign. Ett designkoncept innehåller tre delar: ett namn, ett syfte och huvudprinciper. CDID innehåller sju steg som arbetet är utformat efter: 1.      Concept Generation – Summeras i en tabell som jämför diskussioner i forum (med flera antagande) och diskussioner med hjälp av metoden Philosophy for Children (P4C), samt annan teori. 2.      Concept Exploration – Åtta olika designaspekter identifieras utifrån jämförelsetabellen. 3.      Internal Concept Critique – Designaspekterna jämförs med två snarlika diskussionsformat. 4.      Design of Artifacts – Designkoncept version 1 skapas. 5.      External Design Critique – Intervjuer utförs för att undersöka intervjupersonerna generella erfarenheter av online diskussioner, testar antagandena från jämförelsetabellen och ge direkt feedback på första versionen av designkonceptet. 6.      Concept Revisited – Ändringar görs baserat på intervjumaterial och direkt feedback. 7.      Concept Contextualization – Designkonceptet kopplas tillbaka till litteraturen. Studien resulterar i ett designkoncept kallat CoFramer och stödjer dessa huvudprinciper som ställs i kontrast till material som identifierats genom intervjuerna: Tabell 1: CoFramers huvudprinciper och faktorer från intervjuer om diskussioner på forum. CoFramers huvudprinciper | Från intervjuer om diskussioner på forum Strukturerad början och slut | Diskussion utan tydligt slut Begränsat antal deltagare | Stor mängd deltagare Minimum antal deltagare | Många inaktiva åskådare Explicita deltagare | Lite information om deltagare Gemensamma förutsättningar | Otydlighet kring deltagares förutsättningar Begränsad informationstäthet per inlägg | Långa inlägg och/eller snabba inlägg Den röda tråden och parallella trådar | Oftast parallella trådar Aktiv samtalsledare | Outredda missförstånd Studien indikerar på att CoFramer bör resultera i mer strukturerade diskussioner och som utsätter användaren för en lägre mängd information overload och mindre grounding cost jämfört med vad som vanligtvis uppstår i publika diskussioner online. / The focus of this essay is to develop a design concept for online public discussions with deep discussions and low information overload. The method used is Concept Driven Interaction Design (CDID) which involves constructing a design concept by applying a variety of theories in tangible interaction design. A design concept has three basic parts: a name, high-level goals and outlines generic principles. CDID includes seven steps which is used in this work: 1.      Concept Generation – Formatted into a table that compare forum discussion (with several hypothesis) compared to discussions with the method Philosophy for Children (P4C) and other theories. 2.      Concept Exploration – Eight distinct design aspects are identified from the comparison table. 3.      Internal Concept Critique – The design aspects are compared to three similar discussion formats. 4.      Design of Artifacts – Design concept version 1 is created. 5.      External Design Critique – Interviews are conducted to explore their general experience of online discussions, investigate the hypothesis from the comparison table and to seek direct feedback on the first draft of the design concept. 6.      Concept Revisited – Changes are made to the design concept based on the interview material and the direct feedback. 7.      Concept Contextualization – The design concept is related to the original literature. The result of the study is a design concept named CoFramer. CoFramer’s generic principles are summarized and contrasted against factors identified from the interview material in this table: Tabell 2: CoFramer’s generic principles and interviews about online discussions. CoFramer’s generic principles | From interview material on forum discussion Organized start and ending | Discussions without clear ending Limited number of participants | Large number of participants Minimum number of participants | Large number of inactive spectators Explicit participants | Low information about participants Common conditions | Vagueness in participants conditions Limited information density per post | Long posts and/or fast posts The red thread and parallel threads | Often parallel threads Active facilitator | Unresolved misunderstandings The study indicates that CoFramer would create more structured discussions with a lower amount of information overload and less grounding cost compared what normally arise in public online discussions.

In disseminating information about health issues, government health departments and NGOs use, inter alia, written health texts. In a country like South Africa, these texts are generally written by medical experts and thereafter translated into the languages of the people. One of these languages is Zulu, which is spoken by the majority of South Africans. A large percentage of Zulu speakers are illiterate or semi-literate, especially in the rural areas. For this reason, Zulu translators have to use ‘simple’ language that these readers would understand when translating English texts into Zulu. Translators are expected to use strategies that can deal with non-lexicalized, problematic or other related terms that appear in health texts, as well as geographical and cultural constraints. This study focuses on the strategies used by Zulu translators in an attempt to make translated Zulu health texts accessible to the target readership. The investigation includes the use of self-administered questionnaires for respondents from two of South Africa’s nine provinces, where Zulu speakers are found (Gauteng and KwaZulu-Natal), to determine whether the health texts do reach the target readership. Focus groups, semi-structured interviews and other complementary techniques were used to collect data from the selected respondents. Furthermore, a parallel concordance called ParaConc was used to extract and analyse data from the corpus as compiled for the present study, in an attempt to investigate the strategies used to make the translated health texts easier to read. The study uncovers various strategies which are used when translating English health texts into Zulu. These strategies include the use of loan words, paraphrasing, cultural terms and so on. In future, the use of ParaConc can be broadened to investigate newly discovered translation strategies, with the aim of making health texts more accessible to the target readers. Furthermore, this software programme can also be used to study translation strategies as used in other types of texts, for example journalistic texts. / Linguistics / D. Litt. et Phil. (Linguistics (Translation Studies))

L'interaction face-à-face représente une des formes les plus fondamentales de la communication humaine. C'est un système dynamique multimodal et couplé – impliquant non seulement la parole mais de nombreux segments du corps dont le regard, l'orientation de la tête, du buste et du corps, les gestes faciaux et brachio-manuels, etc – d'une grande complexité. La compréhension et la modélisation de ce type de communication est une étape cruciale dans le processus de la conception des agents interactifs capables d'engager des conversations crédibles avec des partenaires humains. Concrètement, un modèle de comportement multimodal destiné aux agents sociaux interactifs fait face à la tâche complexe de générer un comportement multimodal étant donné une analyse de la scène et une estimation incrémentale des objectifs conjoints visés au cours de la conversation. L'objectif de cette thèse est de développer des modèles de comportement multimodal pour permettre aux agents artificiels de mener une communication co-verbale pertinente avec un partenaire humain. Alors que l'immense majorité des travaux dans le domaine de l'interaction humain-agent repose essentiellement sur des modèles à base de règles, notre approche se base sur la modélisation statistique des interactions sociales à partir de traces collectées lors d'interactions exemplaires, démontrées par des tuteurs humains. Dans ce cadre, nous introduisons des modèles de comportement dits "sensori-moteurs", qui permettent à la fois la reconnaissance des états cognitifs conjoints et la génération des signaux sociaux d'une manière incrémentale. En particulier, les modèles de comportement proposés ont pour objectif d'estimer l'unité d'interaction (IU) dans laquelle sont engagés de manière conjointe les interlocuteurs et de générer le comportement co-verbal du tuteur humain étant donné le comportement observé de son/ses interlocuteur(s). Les modèles proposés sont principalement des modèles probabilistes graphiques qui se basent sur les chaînes de markov cachés (HMM) et les réseaux bayésiens dynamiques (DBN). Les modèles ont été appris et évalués – notamment comparés à des classifieurs classiques – sur des jeux de données collectés lors de deux différentes interactions face-à-face. Les deux interactions ont été soigneusement conçues de manière à collecter, en un minimum de temps, un nombre suffisant d'exemplaires de gestion de l'attention mutuelle et de deixis multimodale d'objets et de lieux. Nos contributions sont complétées par des méthodes originales d'interprétation et d'évaluation des propriétés des modèles proposés. En comparant tous les modèles avec les vraies traces d'interactions, les résultats montrent que le modèle HMM, grâce à ses propriétés de modélisation séquentielle, dépasse les simples classifieurs en terme de performances. Les modèles semi-markoviens (HSMM) ont été également testé et ont abouti à un meilleur bouclage sensori-moteur grâce à leurs propriétés de modélisation des durées des états. Enfin, grâce à une structure de dépendances riche apprise à partir des données, le modèle DBN a les performances les plus probantes et démontre en outre la coordination multimodale la plus fidèle aux évènements multimodaux originaux. / Face to face interaction is one of the most fundamental forms of human communication. It is a complex multimodal and coupled dynamic system involving not only speech but of numerous segments of the body among which gaze, the orientation of the head, the chest and the body, the facial and brachiomanual movements, etc. The understanding and the modeling of this type of communication is a crucial stage for designing interactive agents capable of committing (hiring) credible conversations with human partners. Concretely, a model of multimodal behavior for interactive social agents faces with the complex task of generating gestural scores given an analysis of the scene and an incremental estimation of the joint objectives aimed during the conversation. The objective of this thesis is to develop models of multimodal behavior that allow artificial agents to engage into a relevant co-verbal communication with a human partner. While the immense majority of the works in the field of human-agent interaction (HAI) is scripted using ruled-based models, our approach relies on the training of statistical models from tracks collected during exemplary interactions, demonstrated by human trainers. In this context, we introduce "sensorimotor" models of behavior, which perform at the same time the recognition of joint cognitive states and the generation of the social signals in an incremental way. In particular, the proposed models of behavior have to estimate the current unit of interaction ( IU) in which the interlocutors are jointly committed and to predict the co-verbal behavior of its human trainer given the behavior of the interlocutor(s). The proposed models are all graphical models, i.e. Hidden Markov Models (HMM) and Dynamic Bayesian Networks (DBN). The models were trained and evaluated - in particular compared with classic classifiers - using datasets collected during two different interactions. Both interactions were carefully designed so as to collect, in a minimum amount of time, a sufficient number of exemplars of mutual attention and multimodal deixis of objects and places. Our contributions are completed by original methods for the interpretation and comparative evaluation of the properties of the proposed models. By comparing the output of the models with the original scores, we show that the HMM, thanks to its properties of sequential modeling, outperforms the simple classifiers in term of performances. The semi-Markovian models (HSMM) further improves the estimation of sensorimotor states thanks to duration modeling. Finally, thanks to a rich structure of dependency between variables learnt from the data, the DBN has the most convincing performances and demonstrates both the best performance and the most faithful multimodal coordination to the original multimodal events.