Can robots learn language? Is understanding a language depending on how we see the world and does a Spanish speaker see the world in the same way as an English one?
Linguistic and cognitive experts are going to argue those issues when they arrive at Northumbria University next week for the fifth annual ‘Embodied and Situated Language Processing 2012’ conference August 28-30.
Research funded by the Economic and Social Research Council recently focused on the spatial demonstratives – words, like ‘this’ and ‘that’, which are associated with how far or near an object is to the speaker – and how they vary across languages. In a series of experiments conducted in English and Spanish, Professor Kenny Coventry, Director of the Cognition and Communication Research Centre at Northumbria University and colleagues found that the different versions of ‘this’ and ‘that’ across languages reveal distinctions that are also important for how speakers use ‘this’ and ‘that’ in English.
They discovered that, in both languages, there was a strong association between an individual’s perception of where an object was positioned and their use of spatial demonstrative words.
“Different languages carve the world up in different ways,” said Coventry. “In English we use the word ‘this’ for objects that are close to us or in our hands, and ‘that’ for objects that are distant. Other languages make other distinctions. For example, in Sinhalese – spoken in Sri Lanka – there are two different types of ‘this’: one describing an object you can see and another for an object hidden from view. We wanted to find out whether a person’s language and how it is used to interact with the world, causes them to think in a different way to people with another language.”
The Embodied and Situated Language Processing 2012 conference will focus on how languages are learned and whether robots could become intelligent enough to truly learn, and gain an understanding in, a language in the same way that a human does.
“We discovered that even though languages are different, speakers describe spatial distinctions in similar ways and perceive them in the same way," says Coventry. “The words people use to describe the proximity or distance of an object corresponds with their perceptions of where an object sits in a space.”
Their findings were published in Cognition. Coventry is coordinating the conference with colleagues Dr Larry Taylor and Dr Paul Engelhardt.
Seeing Is Believing? No, Saying Is Perceiving
Comments