Neurobiology of Language -
New PhD Students
From top left to bottom right: Sophie, Dick, Julia and Ksenija
A central role of language is to communicate meaning. When hearing or reading an utterance we usually extract the underlying message without much effort. To achieve this, many different types of information must be integrated with each other such as word semantics or syntactic structure. Such integration is not trivial, but the human brain seems to excel at it. In fact, only recently Google’s brain-inspired language translation system outperformed many earlier computational efforts to approach human-like abilities. Nonetheless, we do not yet have complete insight into how semantic and syntactic representations interact at the neural level. Especially in cases of structural and/or semantic ambiguity computer algorithms often fail to extract the correct meaning, when humans are able to do so.
During my PhD I will further investigate the neural representations of semantic and syntactic information during processing of temporarily ambiguous sentences and the temporo-spatial aspects of their interaction. In the past, psycholinguistic research has demonstrated many factors to influence sentence comprehension. Moreover, neuroscientists have identified several regions that are involved in processing certain sentence-level information. However neural representations of semantics and syntax often seem to overlap in time and space, which is why it is difficult to disentangle their contributions to meaning extraction. In my research I will make use of the high temporal resolution of Magnetoencephalography (MEG) brain imaging methods in combination with decoding techniques to leverage the multidimensionality of brain-data and further disentangle time-courses of semantic and syntactic processing in sentences.
Dick van den Broek
My PhD project zooms in at the basic mechanisms of computation in the brain. We want to investigate on the level of networks of 10.000 neurons what it means to represent a word or construct sentence meaning. We have built a simulation environment with a network of spiking neurons, a language input, and a readout mechanism. So far we have been able to get close to perfect performance on a semantic role labeling task (who did what to whom?). The neurons involved made use of adaptive mechanisms to store information over time. Further work will explore the role of synaptic plasticity and dendritic computations.
Experiencing events usually involves dealing with several different domains: locations, people, goals, and timing, just to name a few. Think about watching someone prepare a meal –there’s actually a lot of things to keep track of! Interestingly, people from different language backgrounds have been shown to experience the same event differently and might even remember different features.
In my PhD I build on the idea that language affects thought, by focusing on how the language you speak may shape your experience of events and guide your attention and memory to specific features within them. Using a variety of research approaches (including Virtual Reality and EEG), and comparing different languages, I will study under which circumstances language effects event experience.
When understanding language, people spontaneously form visual mental images that can, in turn, influence their performance in subsequent language and perceptual tasks. However, we still do not know exactly how automatic this engagement in imagery is or how visual these representations are.
In my PhD, I will investigate further mechanisms underlying this language-perception interaction. In particular, my project will focus on language describing motion and its influence on biological and abstract motion perception, respectively. Ultimately, I hope to contribute to the existing body of knowledge about the systematic differences in content and nature of representations people form when understanding language.