In today’s Academic Minute, Dr. T. Florian Jaeger reveals how language is universally shaped by the inner workings of the human brain.
T. Florian Jaeger is the Wilmot Assistant Professor of the Sciences at the University of Rochester. His research examines how production and comprehension complexity can influence a speaker’s choice in language variation. His findings have been widely published in a number of peer-reviewed journals and he earned his Ph.D. at Stanford University.
Dr. T. Florian Jaeger – The Brain and Universal Grammar
Languages across the world share many similarities. But there’s heated debate among linguists and cognitive scientists about the source of these similarities. Some think that cognitive bias that predisposes human speech toward these similarities. Others think that shared structures found across languages are the result of random accidents of culture, geographic proximity, or distant common origins?
To answer this question, we set up an experiment to observe how people learn a new language. We taught native English speakers a novel miniature language —(we told them this was an ‘alien’ language).
Study participants had to figure out the structure of the language for themselves by watching a series of videos and listening to descriptions of the actions in the new language. For example, they might see cartoons of an alien chef hugging an alien cowgirl accompanied by the following sound clip “Glim klamen kah daf”. After enough examples, separated over just four days, most participants learned to understand the alien language and were able to speak it.
Crucially, however, they changed the language and these changes were not random – we had intentionally created little deficiencies in the language, but participants removed those deficiencies and, remarkably, ended up speaking a version of “alienese” that was better suited for efficient information transfer than the input they received!
Specifically, we found that learners preferred an efficient balance between clarity and brevity. It thus seems like our brain is indeed biased towards languages with certain properties and that efficient information transfer is one of them.