Study Projects


Study projects in progress
Study projects completed

The master programme includes an obligatory one-year study project in which up to 10 students work together: they define their research topic, make their own project plans, work on the topic and present their results.

For the conceptual design of study projects, please see "Study Projects in Cognitive Science"

Study projects in progress

  • Mind-Reading and Social Cognition

    The expression "Mind-reading" stands for our ability to attribute mental contents to both oneself and other individuals. So called "theory-theorists" maintain that the ability to explain and predict mental behaviour is underpinned by a folk-psychological /theory/. According to "simulation-theorists" on the other hand what lies at the root of our mind-reading abilities is not any sort of theory, but rather a /non-conceptual ability/ to project ourselves imaginatively into another person's perspective ("simulating" their mental activity with our own). Since several years "simulation"-approaches are the "hot new game" in the mindreading-debate in cognitive sciences, while the "theory" -approach is characterized as "cold" and as "overintellectualising" our mind-reading capacities. The mindreading theme is an actual interdisciplinary cognitive science research topic with contributions from philosophy, neurosciences, primatology, developmental psychology, clinical psychology and linguistics. The aim of the study project will be to evaluate the range of "simulation"-approaches. After a shared introductory phase, research will be divided among 5 subgroups of 2-3 persons, each of which will be concerned with newest empirical research results concerning mindreading-abilities.

Study projects completed

  • ASADO: Analysis and Structure of Aviation Documents(seit September 2004)
    Website: http://www.cogsci.uni-osnabrueck.de/~ASADO/

    Aus größerer Distanz und aus der Perspektive der Anwendungssituation heraus lässt sich das ASADO-Projekt in 3 Bereiche unterteilen: Archivierungsaufgaben, Knowledge-Management und Benutzungsanforderungen.

    Vor diesem Hintergrund lassen sich drei Aufgabenstränge in ASADO erkennen. Auf der einen Seite geht es darum, die vorhandenen Dokumente unter Verwendung vektorieller Repräsentationen in Gruppen mit ähnlichen Dokumenten zu clustern. Auf der anderen Seite geht es darum, aus den vorhandenen Dokumenten die relevante Terminologie zu extrahieren, Begriffe semantisch zueinander in Beziehung zu setzen (Thesaurus) und daraus ein formales Modell für die zugrundeliegenden konzeptuellen Strukturen (Ontologie) abzuleiten. Beide Arbeitsstränge ergänzen sich systematisch und können als wechselseitiges Korrektiv aufgefasst werden. Die Visualisierung beschäftigt sich dann insbesondere damit, inwieweit die Cluster und das konzeptuelle Wissen recherchierenden Personen graphisch näher gebracht werden kann. Dieser Arbeitsbereich wird durch eine Bedarfsanalyse untermauert werden.

  • Building a Dialog System (seit April 2004)

    The goal of this study project is to build a state of the art dialog system with the markup language VoiceXML. The domain is carsharing: in order to book a car for a certain time it is necessary to phone up a human agent in a call centre. To substitue such routine booking conversations, we are up to construct an automatic telephone system.
    One of the aims of our project is to model dialogs as human-like as possible and use information about the user in order to adapt the system to him/her. This will be done stepwise by developing different, increasingly user-friendly versions. Every step will be evaluated and at the end we will be able to determine which feature caused the most positive impact on user acceptance.
    Since speech synthesis and recognition is taken over by hosting platforms in the internet, we can focus on our research questions. These questions deal with facets of human-machine interaction, user adaptiveness and programming issues. An important aspect is the information flow during the dialogue and how it can be enabled and simplified by the architecture.

  • "feelSpace": Can we create a new sensory modality by providing systematic stimulation with a vibrotactile device? (seit April 2004)

    Former experience with sensory substitution devices has shown that strong integration of new information into perception is possible when fair training is assured. These experiments rely on already existing senses. We innovate the idea by integrating input that does not correspond to natural sensory modalities. This should lead to the creation of a new sensory modality.
    We use longterm stimulation with orientation information on humans. The medium is a custom made belt equipped with a compass, a micro-controller and 13 vibrators. The user feels his orientation in space via vibrotactile stimulation. Thereby, we augment the user’s perceptual space with information which is usually given only indirectly. We expect the tactile perception to disappear whilst conscious information processing is displaced by direct access to a new ”sense of direction”.

  • MINI: Mechanisms in Neuropsychological Issues (April 2003 - April 2004)

    Website: http://www.cogsci.uni-osnabrueck.de/~mini

    The purpose of the project is to have a hands on experience in developing and implementing paradigms regarding cognitive functions as can be used with persons who have recent neurological insults due to tumor, stroke, and other disease processes. In this study project we are currently developing two lines of study. The first is a new paradigm for examing affective priming, in persons with right hemisphere damage, using both picture primes and targets. The second involves the use of a virtual Morris Water Maze to investigate theories of spatial navigation and memory in persons with medial temporal lobe impairments. We hope to have some Theory of Mind components in the project, as well as our practical needs of computer science for programming testing enviroments, and neuroscience and neuropsychology in terms of implementing
    testing and the development of research questions from the grounding theories involved.

  • Higher Level Language Processes in the Brain
    (April 2003 - April 2004)

    When people read texts, they usually build two kinds of interrelated cognitive representations - a representation of the text per se and a representation of the referential situation addressed by the text (situation model). Recent research has suggested that the left hemisphere contains the text representation whereas situation representations seem to be distributed across both cerebral hemispheres.
    The goal of this project is to conduct an experimental study using functional magnetic resonance tomography (fMRI)to further investigate the neural correlates of text representations in the human brain.

  • MAPA: Mapping Architecture for People's Associations
    (Oktober 2002 - Oktober 2003)

    Ziel dies MAPA-Projektes ist die Entwicklung einer neueen Art von Knowledge Structuring Environment. Unter Berücksichtigung von kognitiven Wissensrepräsentationtsprozessen wird ein intuitiv benutzbares Software-Tool entwickelt. Die zu implementierende Anwendung soll den Benutzer bei der Erstellung eines persönlichen Netzwerks von Information Items unterstützen. Der Schwerpunkt liegt auf der Bereitstellung von adaptiven Visualisierungen eines unabhängigen Objekt-Modells. Zunächst wird das System im Bereich des Sprachlernens getestet.

  • PARADIGM: Preference And Reference: Analysing Demonstratives in GerMan (Juli 2002 - Juni 2003)

    Poster presented at EuroCogSci 2003

    This project deals with the distribution and reference of the german demonstrative pronoun "der" and its case and gender variants, in comparison with the behavior of personal pronouns. After investigating the relevant facts about both pronoun types, we develop our "Complementarity Hypothesis", which claims that demonstratives differ from personal pronouns mainly in that they prefer less salient referents. The hypothesis is tested against a newspaper corpus and is found to be fundamentally correct. Two psycholinguistic experiments are conducted.

  • EROSAL: Empirical RObot Study on Animal Learning
    (October 2002 - September 2003)

The study projects listed below were carried out within the Computational Linguistics and Artificial Intelligence study programme, which has since been discontinued.

  • KOKS: Korpusbasierte Kollokationssuche
    (Oktober 2000 - Dezember 2001)
  • ORCA: Osnabrücker Robo-Cup Agenten
    (Oktober 2000 - September 2001)
  • BiKInI: Benutzerbasierte Intelligente Klassifikation von Informationen aus dem Internet
    (ab Wintersemester 1999)
  • PLOT: Prolog-Lernsystem mit Online Tutor
    (ab Wintersemester 1997)
  • Semf: Semantikbasierte Fehlerbehandlung in einem CALL-System
    (ab Wintersemester 1995)
  • MoPs: Modellierung orthographischer Prozesse
    (ab Wintersemester 1994)

  • Promise: Projekt Mediengestütztes Interaktives Sprachenlernen Englisch
    (ab Wintersemester 1993)
    Weitere Informationen (PDF, 112kB)
  • Spock: Stücklisten -Planung, -Organisation und Computergestützte Konfiguration
    (ab Wintersemester 1993)
  • Cavol:Computer Assisted Vocabulary Learning (ab Wintersemester 1992)
    Weitere Informationen (Postscript, 121kB)
  • Peep: Presupposition Explication and Evaluation Project (ab Wintersemester 1992