1. What was the main question in your dissertation?
My PhD research shone a spotlight on how native speakers of Dutch differ in their grammatical knowledge and proficiency. Specifically, I wondered how these differences were related to diversity in life-long reading experience, which varies considerably from person to person.
2. Can you explain the (theoretical) background a bit more?
Most books are not written the way we speak. Writers tend to express ideas in longer, more elaborate sentences, often using grammatical constructions that you don’t come across in everyday conversation. If you read more, you will get more acquainted with these characteristics that make ‘book language’ different from most spoken language. By adulthood, an avid reader has accumulated a vast amount of exposure to ‘book language’ compared to a reluctant reader, who might hardly encounter ‘book language’ in their daily life. Thus, avid readers may be more experienced with using and/or understanding complex grammatical constructions which can often only be found in books. I wanted to find out whether it is indeed the case that people who tend to read more have higher grammatical proficiency. To try to answer this question, I looked at how reading experience contributes to a) adult native speakers’ knowledge of grammatical structures, and b) their processing of grammar in spoken language.
3. Why is it important to answer this question?
Young people in the Netherlands are reading less and less, according to the Programme for International Student Assessment (PISA). While educators, parents, and policy makers agree that all of us should read more, not much is understood about the ongoing benefits of reading for other language skills in adulthood, especially when it comes to grammatical proficiency. I use “grammatical proficiency” as a general term to refer to the knowledge and processing skills that are needed to make sense of grammatical structures and use them effectively, both in writing and in speech. My research aimed to address this relative gap in our understanding, using a variety of experimental methods.
4. Can you tell us about one particular project?
My final project focused on how people use grammar to predict what is coming up and if people who read a lot and therefore have high grammatical proficiency, are faster predictors. Grammar is essentially a language’s rulebook for language use, so the rules can be used to predict what someone is going to say. For example, in some languages, nouns have grammatical genders and an article that must match said gender. For example, in Dutch you have the article ‘de’ for feminine and masculine nouns and the article ‘het’ for neuter nouns. You can use these rules of grammatical gender to predict what someone is going to say. So if someone is holding a shirt (het overhemd) and a sweater (de trui) and they say “Ik draag vandaag de [Today I’m wearing the] …” you know that they will have to say “trui” next.
In my study I looked at this kind of grammatical prediction in real time using eye tracking. Eye Trackers are very useful tools, because we know that eye movements are triggered within milliseconds of recognising, or even anticipating, a spoken word. So even before you hear “trui” in the example above, your eyes would have moved to the sweater after you heard the article ‘de’.
For a new experiment, I invited participants who reported in my previous study to have a lot of reading experience and participants who reported having less experience with reading. I tracked their eye movements as they listened to passive sentences in Dutch, like “Het raam wordt inderdaad gebroken door een stier” (The window is indeed broken by a bull). The target word to be predicted was the agent of the action, so “stier” (bull) in this example. While listening, participants saw a visual display with four pictures, for example a window, a bull, a bath, and a belt. Only one of the pictures was a plausible agent of the action mentioned.
To be able to anticipate the upcoming agent before it is mentioned, the listener has to be sensitive to the cues in the unfolding sentence that signal the grammar of the sentence and thus the likely identity of the final word. The first possible cue is the passive auxiliary verb “wordt”.
Overall, the participants who had reported to be highly experienced with reading looked more often to the plausible agent before it was mentioned in the speech than the participants who had reported to be less experienced with reading. This suggests that the people who have a lot of reading experience made better use of their grammatical knowledge in order to predict how it would end.
5. What was your most interesting/ important finding?
Most excitingly, I found evidence that people who read more are better at using their grammatical knowledge to predict upcoming words while listening to speech.
6. What are the consequences/implications of this finding? How does this push science or society forward?
Our ability to predict upcoming language is thought to support our language comprehension and make it more efficient. If avid readers predict more while listening due to their heightened grammatical proficiency, it suggests that long-term written language experience has functional benefits for the processing of spoken language. In other words, in case we needed any more convincing that reading isn’t a waste of time, it appears that long-term exposure to ‘book language’ can lead to improvements in grammatical proficiency and language ability in general.
7. What do you want to do next?
I am currently exploring opportunities in applied research, with a special interest in education and adult literacy.
– Link naar dissertatie
Interviewer: Merel Wolf
Editing: Julia Egger
Dutch translation: Cielke Hendriks
German translation: Natascha Roos
Final editing: Merel Wolf