Finding the link between language and perception

Wednesday, 22 April, 2015
Language is at the heart of everyday communication. But we don’t all understand language in the same way, with prior information, age and cognitive ability playing a key role. An EU-funded network is studying how people in different ability and age groups perceive language. Its projects are developing tools and training software that could help people with disorders like autism communicate better.

Perception and language have traditionally been two distinct fields of study. Now, for the first time, the Marie Skłodowska-Curie Initial Training Network LanPercept is examining their relationship in tandem.

"Our main aim is to produce a new generation of scientists who can advance our understanding of the interaction between these two cognitive processes," says Mila Vulchanova from the Norwegian University of Science and Technology, coordinator of the LanPercept network.

The researchers are testing how visual objects, events and actions shape language understanding. They are also studying how language influences the attention we pay to the visual environment. For example, pronouns like ‘this’ and ‘that’ can influence our perception and memory of distances. This usually means that we perceive ‘this glass’ as being closer to us than ‘that glass’ even if the actual distances are the same.

Local and global processing

LanPercept is an umbrella for a range of projects. 'The development of local and global processing: from perception to language' is one of them and involves Dorota Smith – a PhD student and Early Stage Researcher from Aston University. Smith’s project explores the link between local and global processing in perception and language in individuals with and without developmental disorders such as autism.

Many people with autism have extraordinarily detailed perception but find it hard to perceive and understand the overall context – a so-called 'global processing' ability. Early research from the 1980s concluded that this is an impairment in people with autism, while later research rejected this idea, explaining the difference as a bias towards local processing. Smith wanted to know what factors lead to those contradictory results.

“We hope to shed light on this issue by looking at how typically developing individuals and those with autism respond in a range of local-global visual and language tasks and questionnaires,” she says.

Her project has already shown that in volunteers without autism, biases in processing visual information can be altered by providing additional information about the task. The next step is to clarify whether the same will hold true for people with autism.

According to Smith, knowing the relationship between processing in visual perception and language could potentially make it easier for clinicians and teachers to identify individuals who struggle with communication, allowing for earlier intervention. After her doctoral research, Smith plans to combine standard psychological assessments with brain imaging to make further strides in understanding this condition.

Various industrial partners within the network are collaborating with the researchers, providing secondments, entrepreneurial training and commercial guidance. By the time LanPercept finishes in 2016, the researchers involved will have acquired the skills and knowledge to improve their chance of a successful career in academia – or in the private sector. Moreover, by working together, the team expects to be able to provide new expert knowledge and solutions to the wider research community.

LanPercept gathers 10 partners (2 from the private sector and 8 universities). Five additional partners (all from the private sector) are associated to the network. Eight different countries (Norway, UK, the Netherlands, Germany, Spain, Denmark, Sweden and Finland) are involved in LanPercept.

Language and Perception
Project Acronym: