“The whole idea about language learning is changing. A big influence on the field of language development has been the idea by Chomsky who said that language was something very specific. Also, the view on language was quite modular, that is, language consists of different domains that do not interact. On the other hand, brain science tells us that the brain is one big network and everything is interconnected. People are now moving towards a more integrative approach. A lot of researchers are really trying now to understand more about how language relates to other parts of cognition.”

“Of course, there is still the language system that needs to be investigated. The theoretical idea behind how we come to understand the variety of all different languages. You can study that in a very cognitive approach, but you can also study that more theoretically: what is the system? I hope these approaches will become a bit more complementary.”

“After the first year of studying Dutch I already knew that I wanted to do linguistics. One of my professors was Roelien Bastiaanse – a professor in Groningen in aphasia. She always taught her courses in a very interdisciplinary way, she encouraged us to take courses in psychology and neurology. I soon knew that that was what I liked: to study language, but really in relation to cognition, psychology and the brain. And especially: what the outcome is when something in those three, or in their relation, goes wrong.”

“My research lines have been and will be on how children first of all learn a language, but also how the language system develops when we observe that certain language structures are impaired. In children with dyslexia, for instance, who cannot learn to read and write successfully, or in children with SLI (Specific Language Impairment) who have severe difficulties in their oral language development.”

Non-sense words

“I have just finished some work on how children process and learn the forms of lexical representations. Children with SLI, but also children with dyslexia, often have problems accessing the right word form. They for instance have problems with repeating a non-sense word like “snup”. In order to do that, you really have to understand how all these phonemes are sequenced.”

“These children also have problems in short-term memory, and especially phonological short-term memory, which is about accessing and storing the specific phonemes in a certain structure. Now the question is whether it is just due to their memory or whether they have problems accessing lexical representations that are already in long-term memory. The interplay between short-term memory and long-term memory is really important: if they show deficits on a memory task, is it just short-term memory, or does it have something to do with the form of the language that was stored, the mental lexicon.”

“I am also investigating the interaction between phonology and morphology using EEG (Electroencephalography). Here again we find that children with SLI and dyslexia often have morphological problems, for example in inflecting a verb for the past tense, but they also have phonological problems. It might be that the interaction between phonology and morphology is really causing the problems.”

“Recently, I have been doing a few tasks together with Elise de Bree from Utrecht University. In Dutch we have two different surface forms to designate the past-tense, we have bakte and rende, de and te. The phonology of the verb stem determines the morphological form. We asked children to inflect a non-sense verb for the past tense and checked whether they did it right. It turns out that they find it easier and are more accurate when they have to inflect a te-verb. Why are such verbs easier to learn?”

“In Dutch, there is devoicing. We have many instances in which we devoice. For instance at the end of ‘hond’, we pronounce a t while it is written with a d. Maybe children are more tuned in to devoiced elements. It was remarkable that children with SLI, who had problems with inflecting the verbs for the past tense, also showed this advantage for -te verbs, just like children with developmental dyslexia. These children thus showed the same sensitivity to the language characteristics of Dutch. We further wanted to investigate the relationship between morphology and phonology using the EEG methodology. We (Marieke Woensdregt and I) are presenting people sentences in which we violate the surface-form (like for instance bakde and rolte) and see how the brain reacts to such a violation. Does the brain show a response that is typically seen in other phonological violations, or does the brain respond with a morphological response pattern? Again we see here an interaction between phonology and morphology.”

In the field there is discussion about whether Specific Language Impairment (SLI) is single disorder, or an umbrella term for many different things.
“The SLI diagnosis is at this moment based on exclusion. If you do not have general cognitive and perceptual impairments and some other criteria, then you are diagnosed with SLI. But there are very few children with just pure SLI, without any other attention or reading problems and their language problems are often diverse. Some children have more phonological problems, some have more grammatical problems – it is a mixed bag. And there is overlap between different syndromes, such as with developmental dyslexia. Dyslexia is diagnosed based on the presence of severe problems in literacy acquisition, but often these children also show subtle problems with their oral language development. In addition, many children with SLI are also diagnosed with developmental dyslexia. I am not really focused on understanding the comorbidity (the presence of multiple disorders) but I think it is something you have to be very sensitive to when you are studying the language problems of these children.”

“In SLI research there are some people who claim that language is innate, that there are specific areas in the brain for language and that children with SLI have a genetic cause that affects specifically and only language development. That might be true for some children, but in general, if you study children with language problems, they might have normal intelligence, but they often have verbal short term memory problems. There is now also evidence that children with SLI have problems in other executive functioning as well, for instance inhibition. The exact relationship between executive functioning and language needs to be investigated more, but I would opt for the position that language is learned by multiple cognitive mechanisms and that learning in itself is affected, in language, but also in other areas.”

Genes and language

“I think SLI research will in a few years be much more in relation to the brain. There are many studies now on the genetics of language and also on the genetics of SLI. Some time ago we had Simon Fisher visiting Amsterdam. He is now in Nijmegen, but he used to work at UCL in London. He has done a lot of research on the genetics of language and he spoke about the genetics of SLI. We are very far off to understanding a link between a certain gene and language, but he really showed that there may be different types of SLI and they may be related to different types of genetic disorders. People studying comorbidity will move away from the behavioral level, and focus more on the neuronal level.”

“I think also a lot will change in the kind of learning mechanisms that we study. In the eighties and nineties research was really focused on language theory. For example: what does a syntactic tree look like in a child with SLI? I do not think we will see that kind of research much anymore. The question will be much more on why some children do not learn certain linguistic structures and how that is related to learning mechanisms. There is now a big focus on statistical learning: the ability to make use of transitional probabilities for language learning. For instance, research has shown that babies are able to discriminate between different nonwords based on learning that particular sound sequences occur in that combination more frequently than in other sound combinations. I think SLI research will be focused more on the learning mechanisms: language is just one product that you learn through for instance statistical learning. Of course that ties in with the fact that it is a lot more accepted now that language learning involves several cognitive mechanisms.”