Analogical effects on linking elements in German compound words

Abstract

This paper examines whether the selection of linking elements for novel German compounds can be better explained in terms of a single or a dual-route model. Previous studies had focused on the predictability of linking elements by rules. We investigate a single-route model by focusing on the paradigmatic analogical effect of the compounds sharing the left (right) constituent with the target compound, i.e., the left (right) constituent family. A production experiment reveals an effect of the left, but not of the right constituent family. Simulation studies of the responses, using a computational model of paradigmatic analogy, show that the left constituent and its phonological and morphological properties (rime, gender, and inflectional class) simultaneously codetermine the selection of linking elements. We show how these results can be accounted for by a single-route approach, and we outline a symbolic interactive activation model that merges the factors into one psycholinguistically motivated processing mechanism.

Acknowledgments

We are grateful for the help of Arne Fitschen and Ulrich Heid for providing us with a list of 34,000 German compounds. We also thank Kathrin Delhougne for her help in preparing these compounds for further analyses. This study was financially supported by the Dutch National Research Council NWO (PIONIER grant to the third author), the University of Nijmegen (the Netherlands), and the Max Planck Institute for Psycholinguistics (Nijmegen, the Netherlands).

Notes

¹Testing rime, gender, and inflectional class as co-variates in an overall analysis is not possible because the responses in the three sub-experiments were categorised differently, namely as -s-responses vs. others, -en-responses vs. others, and -Ø-responses vs. others.

²For a description of the model's similarity metrics, see Daelemans et al. (2000) and Krott et al. (2001).

³For all reported prediction accuracies, the following parameter settings were used: similarity algorithm: IB1; feature metrics = weighted overlap; features weighed by information gain values; size of best neighbour set = 1. Different settings do not change the pattern of results. For detailed information about the parameters, see Daelemans et al. (2000).