MODELACT Conference



Call & Topics
Keynote Speakers
Venue & Accomodation


Rome, 6-7 June 2016
CNR (National Research Council)
Aula Marconi - P.le Aldo Moro 7 (see in Google Maps)



In the last decade the relations between language, action, and cognition have been investigated from a wide range of perspectives, including semantics, neurosciences, psychology, philosophy, computational linguistics, and robotics.

As a matter of fact, action verbs bear the basic information that should be understood in order to make sense of everyday sentences, but their lexical encoding remarkably differ from language to language. For this reason, both the semantic disambiguation and the mutual translation of action verbs in natural languages are problematic.

In language acquisition (and specifically for sign languages, which rely on a visual-gestural modality) the link between the conceptualization of actions and their physical performance through the body become crucial, since gestures anticipate linguistic behaviors in the early stages of child development.

From another point of view, the processing of action related sentences involves the activation of the motor system in the brain, leading to an embodied representation of the linguistic meaning. More in general, neuroscientists investigated the systems which underlie both language and sensorimotor faculties, and developed models in which linguistic, cognitive and motor abilities strictly interact.

Researches on recognition, categorization, and execution of actions have also been exploited in the field of robotics, in order to conceive artificial systems that can understand instructions and perform actions in accordance to a human-centered model. These researches have been often carried out in connection to computer vision tasks, which turn out to be essential for action recognition and for anchoring words (referring to the objects implied in the actions) to the real word entities.

In recent years, several important European projects and initiatives have been developed in order to explore the connections among actions, cognition, and linguistic abilities, with a particular attention to robotics and automatic systems (Poeticon++Robot-EraTHE Hand EmbodiedACat).

Among them, the MODELACT Project (based on the IMAGACT linguistic ontology of action) aims to identify how cognitive and linguistic action categories are defined in humans, and to develop models that can be used to take account of these facts. The project board promotes the MODELACT Conference on "Action, Language and Cognition" (to be held in Rome on 6-7 June 2016), which intends to represent a meeting point for the widening community that is concerned with the above-mentioned research topics and fields.