As a psycholinguist, I am fascinated by the observation that humans are able to understand each other so efficiently without any apparent effort. My research interests therefore revolve around the question what the mechanisms are behind successful language comprehension. In my work I have mostly approached this question by looking at the interaction between linguistic and cognitive processes. Over the years, my investigations have also convinced me that the study of language is most insightful when using a broad range of offline and online methods with diverse populations.

The following three research lines illustrate the scope of my work.

1) Predictive processing in various populations

In everyday life, humans constantly predict. For example, they use prediction to ride through busy traffic, to catch a ball, and to finish another person’s sentence. A wealth of research has shown that one of the mechanisms responsible for the fast recognition of spoken language is prediction. Prediction plays an important role in current approaches to sentence comprehension and it has been described as a central ingredient of efficient communication. In my work, I use the visual-world paradigm to examine how efficiently different cues can be used for prediction across several populations (monolingual adults and children, bilingual children, adults with dyslexia) and what the role is of possible mediating factors.

2) Speech processing in challenging listening conditions

In daily life, humans recognize speech under a wide range of challenging conditions such as speech signal and/or environmental degradation. However, most speech recognition theories are built upon evidence collected in optimal situations in controlled conditions. A problem with these conditions is that they often fail to capture the processes involved in everyday speech recognition, thereby underestimating listeners’ flexibility. A detailed understanding of how the speech system operates under challenging conditions can improve the external validity of speech recognition models and enhance our knowledge of the interaction between language processing and cognition. During my PhD I examined how listeners recognize casual speech that often contains large amounts of variation. For example, a Dutch speaker may say the word wedstrijd ‘match’ as wes. This variation in production is called speech reduction, where segments, syllables and even whole words can be changed and/or deleted.

Currently, I examine how listeners recognize speech in noise. In the literature, a distinction has been made between energetic and informational masking. Energetic masking refers to masking at the auditory periphery. In informational masking, the target and the noise may both be audible, but difficult to separate and therefore depends on factors that inhibit or facilitate stream segregation including linguistic and cognitive factors. I use several online and offline techniques to investigate these components of informational masking during speech-in-speech recognition.

3) The role of linguistic and cognitive factors in moral decision making

This research program extends on the previous finding that people tend to make systematically different judgments when they face a moral dilemma in a foreign language (L2) than in their native language (L1). This so-called Foreign-Language Effect entails that an L2 elicits reduced emotional decisions compared to an L1. In the previous work, moral dilemmas were presented visually (i.e. written texts) to bilingual adults. I am currently looking at whether this Foreign-Language Effect also appears when moral dilemmas are presented auditorily to bilingual adults. Moreover, I am presenting these dilemmas in challenging listening conditions such as in background noise or when uttered by a speaker with a foreign accent or a dialect. These results will contribute to our understanding of the influence of reduced signal reliability on decision making processes in adults. This research has also important implications for our globalized world/society, given that many individuals (e.g. immigrants, bilinguals, multinational companies) make moral judgments in both L1s and L2s under several challenging conditions. The results of this project may further provide a starting point for educational research into effects of an L2 and sub-optimal listening conditions on emotionally-driven decision making.