Contact: lea.hofmaier(a.t.) uni-tuebingen.de
• 2010: Abitur at the Geschwister-Scholl-Schule Tübingen
• 2011-2014: B.Sc. Cognitive Science at the Eberhard Karls Universität Tübingen
• 2014-2019: M.Sc. Cognitive Science at the Eberhard Karls Universität Tübingen
• Since March 2020: Research Assistant and PhD Student at the Research Training Group 1808: Ambiguity – Production and Perception at the Eberhard Karls Universität Tübingen
Abstract: "Modeling Ambiguity Production and Perception in an Instruction Task" (working title)
Is ambiguity an error in communication or does it serve a purpose? There is a lot of evidence that suggests that ambiguity is being used deliberately in communication, perhaps in order to save time, or maybe to fulfill societal expectations (e.g. it is considered impolite to explain things that they already know in detail to your conversational partner). It is now my goal to examine how exactly and why ambiguity arises in communication by creating an artificial model of a suitable problem.
My dissertation project is concerned with the communication between two artificial agents in a task-based context. The goal of the task is that one (speaker) agent phrases instructions in such a manner that another (hearer) agent is capable of correctly executing them. By limiting the permitted sentence length (i.e. the transmitted information) ambiguity arises in the instruction. I now want to investigate whether the speaker will be able to phrase the instructions (i.e. choose the transmitted information) in such a way that arisen ambiguity can be disambiguated within the context. Furthermore, I want to examine to what extent social aspects (e.g. politeness in conveying too much or too little information) have an influence on communication and the utilization of ambiguity.
• Predictive cognitive models
• Speech processing of simulated agents
• Artificial Intelligence
• Mental representation
• Ambiguity in communication