LOGHOLTOCOM
From EuroVR Knowledge Base
- Full name
- Language origins and grounding: simulating the transition from holistic to compositional languages
- Acronym
- LOGHOLTOCOM
- Website
- Keywords
- Budget
- € 168232.00 euro
- Start date
- 2004/02/01
- End date
- 2006/01/31
Contents |
Partners
Funding Bodies
- EU FP6
- MOBILITY-2.1 Marie Curie Intra-European Fellowships (EIF)
- Marie Curie
Contact
- NOBLE, Angela (Ms)
- Tel: +44-1316509024
- THE UNIVERSITY OF EDINBURGH, SCHOOL OF PHILOSOPHY, PSYCHOLOGY AND LANGUAGE SCIENCES, LANGUAGE EVOLUTION AND COMPUTATION RESEARCH, Old College, South Bridge, UNITED KINGDOM
Objectives
This project involves research into the origins and evolution of human language by using computer simulations. The research will concentrate on the following question: Have the underlying principles of language arisen through biological evolution or through cultural evolution? If language has evolved through biological evolution, the human genome should capture certain aspects of human languages. If it has evolved through cultural evolution, languages must be shaped through the interactions between humans rather than through biological adaptations. Within these two positions, the project focuses on the evolution of two particular aspect of human language, namely its compositional structure and it?s grounding in reality. In language expressions are composed of meaningful parts that can be recombined in other expressions; compare, for instance, the expressions "the red square" and "the blue square", where "the square" is composed with "red" and "blue". This is compositionality. Grounding relates to the question how the meanings of utterances relate to the real world phenomena they stand for. For instance, how does a language user know what "the red square" means? The research will be carried out using computer simulations in which a population of virtual robots will try to develop a language from scratch in order to communicate events that take place in their environment. We will investigate the question of whether language evolved culturally or biologically by developing two studies. In the first study no biological evolution will be used, but the robots will use communicative interactions and individual learning mechanisms to develop and transmit linguistic knowledge from one generation to the next. In the second study, where we investigate biological evolution, we will additionally pass linguistic knowledge genetically (i.e. using artificial DNA) to the offspring of a particular generation.
Main Results
End Users
Evaluation Methodologies
Fill with the evaluation methodologies. Evaluation methods already in the KB can be found in Category:Evaluation Methods
