Doctoral Thesis
My dissertation topic is Form-meaning mismatches in language acquisition, and is integrated in the DFG-funded research training group RTG 2636 Form-meaning mismatches.
In the computational part of my work, I investigate (1) how phonological/semantic connectedness of words influences the learning of other similar sounding/meaning words, and (2) how phonological and semantic connectedness interact. I use longitudinal CDI data from wordbank.edu to represent children's growing vocabulary in networks using Python. These are then analysed with R. In the phonology-related sub-project of my dissertation, I collaborates with Michaela Vystrcilová, Laura Pede and Alexander Ecker from the Neural Data Science Group at GAUG.
The experimental part of my work is devoted to the question of what effect (non-)arbitrary word sound and word meaning mappings have on early childhood word learning. For this purpose, I conduct an eye-tracking experiment on young children. In this part of my PhD I collaborate with Ming Yean Sia.
Judith's work is supervised by Nivedita Mani (GAUG, GEMI), Thomas Weskott (GAUG, Seminar for German Philology) and Markus Steinbach (GAUG Seminar for German Philology).
This animation shows the growing vocabulary of a Norwegian child between 17 and 35 months. The connecting lines between the words mean that the connected words are semantically similar. Semantic similarity can be measures in different ways, i.e., associations, features and word embeddings. The semantic similarity used for this animation is based on the University of Florida free association norms (Nelson et al., 2004).
Nelson, D. L., McEvoy, C. L., & Schreiber, T. A. (2004). The university of south florida free association, rhyme, and word fragment norms. Behavior Research Methods, Instruments, & Computers, 36(3), 402–407.