Max Planck Research Group "Neurotypology"
Language is surely the most sophisticated communication system known to us. However, the precise characterisation of this uniquely human skill is rendered difficult by the high diversity of its manifestations: there are over 6000 languages in use in the world today, each with its own unique characteristics. These distinctive properties apply to all levels of a language, beginning with its sounds, over the form of its words to the manner in which individual words are combined to make up a sentence and how sentences may then be used in a discourse.
Despite these striking differences, the most common assumption in neurolinguistics – i.e. the discipline concerned with the brain bases of language – has been that, as far as the brain is concerned, all languages work in essentially the same way. This means not only that the same brain regions are thought to engage in language understanding, but also that their function has been hypothesised to be very similar across languages. However, recent findings challenge this hypothesis of a cross-linguistic uniformity in the neural basis of language. Rather, they indicate that even very closely related languages such as English and German may differ fundamentally with respect to the neurocognitive mechanisms involved in language understanding. Conversely, it also appears that unrelated languages such as English and Finnish may behave similarly in certain respects.
Taking these observations as a point of departure, the research of the Independent Junior Research Group “Neurotypology” adopts a new approach to the examination of cross-linguistic similarities and differences in the neurocognitive bases of language comprehension. Specifically, we assume that the grounding of language in the – presumably universal – higher cognitive abilities of humans results in basic underlying mechanisms that are common to all languages. However, depending on the properties of an individual language, these commonalities will then be subject to language-particular specialisation. One example for a presumably universal processing mechanism is the establishment of hierarchical relations between sentential arguments. Thus, in an English sentence such as The policeman comforted the little boy, the comprehension system must determine that the policeman is the Actor of the event being described and the boy is the Undergoer of that event. While English assigns these participant roles (the Actor-Undergoer hierarchy) on the basis of linear position (Actors typically precede Undergoers), other languages employ morphological information to the same purpose. For example, the German sentence Den kleinen Jungen tröstete der Polizist, ‘[the little boy]ACC comforted [the policeman]NOM’ has the same meaning as the English example above even though the order of the participants is reversed. Despite the overt differences between these two comprehension strategies, which are supported by numerous empirical findings, we assume that they are manifestations of the same underlying processing mechanism, albeit in a language-specific form. Indeed, converging evidence for this hypothesis stems from the finding of very similar neurophysiological and neuroanatomical correlates of processing for these seemingly diverging domains in the two languages.
The aim of the Junior Research Group is to examine cross-linguistic similarities and differences of this type using neurocognitive techniques – i.e. event-related brain potentials and, in certain cases, functional magnetic resonance imaging – and to model these findings within a cross-linguistic neurocognitive model of language comprehension, the “extended Argument Dependency Model”, eADM (Bornkessel, 2002; Bornkessel & Schlesewsky, submitted, to appear; Schlesewsky & Bornkessel, 2004). To this end, we are currently studying Turkish, Hindi, Japanese, Icelandic and Chinese.