The ability to learn rules is at the heart of the ability to learn language. This thesis is a collection of papers tackling rule learning from various perspectives and domains – including... Show moreThe ability to learn rules is at the heart of the ability to learn language. This thesis is a collection of papers tackling rule learning from various perspectives and domains – including visual, auditory, and speech domains – in both infants and adults. Using both simple XYX-, XXY-, or XYY-type rules, and more complex Lindenmayer grammars, we were able to gain insights into the rule learning processes of young infants and of adults. While we were unsuccessful in attempted replications and extensions of previous studies, it was precisely these failures that helped to provide a more nuanced picture of rule learning: even the simplest type of rule learning is far from straightforward. For infants, we find evidence for a repetition bias in both the visual and speech domain that is difficult to overcome, while for adults we show that the learning environment – the task used, the instructions, types of testing stimuli – are all highly influential in determining whether a simple rule can be learned or not. Furthermore, by studying patterns found in babbling we were able to hypothesize for the first time about the parallels between production and perceptual abilities with respect to rule learning. Show less
This doctoral dissertation reports five studies investigating the processing of prosody and music with cochlear implants or simulations thereof (vocoders) in Dutch. Cochlear implants are... Show moreThis doctoral dissertation reports five studies investigating the processing of prosody and music with cochlear implants or simulations thereof (vocoders) in Dutch. Cochlear implants are implanted hearing prostheses that partly restore hearing for profoundly deaf individuals by presenting an electrical reconstruction of sound to the hearing nerve. Prosody is the melody and rhythm of speech and is crucial in spoken communication. Important functions of prosody include the conveying of emotions (emotional prosody) and the marking of new or old information in utterances (linguistic prosody). These functions are realized by speakers, among other ways, by means of variation in intonation and the duration of parts of an utterance. Of these forms, the perception of intonation variations, but not particularly of duration variations, is notoriously difficult for cochlear implant users. This difficulty is caused by limitations of the device and the interface between cochlear implants and the hearing nerve. Possible limitations in perception and production of linguistic and emotional prosody by actual and simulated cochlear implant hearing had never been systematically compared. This thesis shows that cochlear implant users might have more difficulty discriminating emotional than linguistic prosody and that they rely relatively much on intonation cues for emotional prosody but on duration cues for linguistic prosody. Tests with vocoders showed that sharpening the slopes of spectral filters (simulating reduced spectral smearing) improves prosody perception up to values much extremer than tested before. Taken together, this set of experiments discusses issues to take into account when studying the perception and production of prosody by cochlear implant users and with vocoder simulations. Show less
This dissertation is concerned with the development of grammar. Starting from a usage-based perspective, which holds that children use domain-general learning mechanisms to acquire the... Show moreThis dissertation is concerned with the development of grammar. Starting from a usage-based perspective, which holds that children use domain-general learning mechanisms to acquire the grammatical patterns of their mother tongue, Beekhuizen shows how to operationalize various concepts from this tradition in a computational model. In order to arrive at a sound set of assumptions, Beekhuizen compares and criticizes various earlier usage-based modeling approaches and scrutinizes the concepts of a usage-based theory of language acquisition from the perspective of a computational modeler. As the model should be able to produce utterances on the basis of a meaning to be expressed, as well as to interpret utterances, the availability of meaning from the situational context is studied empirically. The resulting model, the Syntagmatic-Paradigmatic Learner, simulates an increasing ability to understand utterances on the basis of a grammar of constructions, as well as to produce utterances on the basis of this grammar. Several developmental effects are simulated and the internal states of the model are carefully examined. Show less
The Feature Co-occurrence Constraint theory proposed in this dissertation provides a means to capture the development of the language-learning child's segment inventory. It does this by combining a... Show moreThe Feature Co-occurrence Constraint theory proposed in this dissertation provides a means to capture the development of the language-learning child's segment inventory. It does this by combining a growing set of features with constraints that are automatically activated as soon as these features are acquired. Representation and derivation go hand in hand, and develop together during acquisition. The Feature Co-occurrence Constraint theory builds on a minimal view of phonology, where the inventory is seen as epiphenomenal rather than a mentally ‘real’ object, features are few and monovalent, and the constraint set is limited to no more than two types. The theoretical consequences of the proposal for both feature theory and constraint theory are worked out in detail and a thorough discussion of phonological acquisition is provided, making this book of interest to both theoretical phonologists as acquisitionists. Show less
The cochlear implant (CI) gives severely hearing impaired to profoundly deaf children access to auditory speech input and consequently stimulates their oral language development. However, speech... Show moreThe cochlear implant (CI) gives severely hearing impaired to profoundly deaf children access to auditory speech input and consequently stimulates their oral language development. However, speech perception with a CI is still not optimal. Therefore, these children develop oral language based on reduced auditory speech input. This dissertation aims at enhancing our knowledge of whether a CI provides sufficient access to auditory speech input to acquire verbal morphology. It takes a new perspective regarding the research on morphology acquisition in CI children. The outcomes of the children under investigation are not only compared to those of their normal hearing peers, but also to those of their specific language impaired (SLI) peers. The latter group is known to be particularly delayed in their acquisition of verbal morphology. One of the major findings of this dissertation is that CI children outperform their SLI peers in the production of verbal morphology. Remarkably, they are even able to catch up with their normal hearing peers. Nevertheless, their spontaneous speech samples contain more verb inflection errors as compared to similar samples from normal hearing peers. In this respect CI children compare to their SLI peers. The results of the CI children are further analyzed as a function of their age at implantation and hearing age. Additional emphasis is given to the role of perceptual salience in the acquisition of morphology. Show less
