Theses

Theses (Bachelor/Master) - Winter semester 2018/19 and summer semester 2019

Multisensory perception of children and adolescents between 4 and 20 years of age

In the context of a research project funded by the DFG (Deutsche Forschungsgemeinschaft), we perform studies with blind children who regain eyesight by means of clinical surgery. Most of the time these children suffer from cataract, which can easily be removed surgically. As opposed to the western world, such surgery is very rarely offered in developing countries. Therefore, those children often show reduced perceptive abilities for their entire life. In order to develop appropriate methods for rehabilitation we perform similar studies with children of the same age in Germany. The tasks in these studies consist of simple multi-sensory perception and learning experiments, the goal of which is to analyze the participants’ response.

Vision for action in children and adolescents between the age of 4 to 20 years

In the context of a research project funded by the DFG (Deutsche Forschungsgemeinschaft), we perform studies with blind children who regain eyesight by means of clinical surgery. Most of the time these children suffer from cataract, which can easily be removed surgically. As opposed to the western world, such surgery is rarely offered in developing countries. Therefore, those children often show deficits not only regarding perceptive abilities but also when executing simple motor tasks that are supported by visual feedback, like grasping objects or walking from one place to another. In order to develop adequate methods for rehabilitation as well as to improve our understanding of the basis of plasticity and learning with regard to visually controlled tasks, we perform studies with children of the same age in Germany. The experiment tasks comprise simple actions, such as grasping and handling objects, walking and navigating, or throwing and catching. The behavior of the participants will be recorded and analyzed afterwards.

Immersive 3D video games without falling

Realistic simulations and 3D games are becoming increasingly popular with the proliferation of Head Mounted Display (HMDs), such as the Oculus Rift or the HTC Vive. The problem is that persons using HMDs often fall as the visual world in which they are immersed is too immersive and affects the sense of balance. In this project, we investigate ways to reduce or avoid these effects by means of multisensory stimulation. Together we develop 3D environments, stimulate the sense of balance via various sensory cues and analyze the entire body movement of the participants.

Are we walking in circles when we’re lost in the desert?

The answer to this question is: yes. We have confirmed this in several studies. The circles are partly very small. The reason why we’re walking in circles is, however, not yet fully resolved. That is the reason why we want to study how participants find their way in different environments, and to which extent they are able to walk straight. The available information will be changed and the walking behaviour in several environments (forest, field, desert) will be analysed.

How do I find my way home in an unknown environment?

Humans have the impressive ability to find their way in almost every unknown environment. In order to do that, they use multiple strategies, as for example navigation based on landmarks or path integration. Of course we also make characteristic mistakes. These provide information about the mechanisms that we use to find our way around. In this project, we use both actual and articially generated environments (in virtual reality) to analyse the processing of the sensory information that is needed to successfully solve navigation tasks. 

Costs and benefits of learning new motor skills

We must learn various motor skills, such as cycling, skiing, playing the piano, etc. with some effort. Some things can be relatively quickly learned (cycling), with other skills taking years to optimise (playing the piano). In this project, we investigate the factors influencing this learning behaviour with the aim of better understand the plasticity of the brain and possibly to develop suitable training paradigms. To do so, we examine the behaviour of the subjects in learning new motor actions and analyse their changes.

Auditory Perception in Motion

How do we know where a sound comes from, how fast the sound source moves towards us or whether it has a single or multiple origins? Compared to vision, where spatial information is presented topographically already on the retina, our auditory system needs to derive the spatial topology from the frequency spectrum of the perceived sound. For that, the expectation of a sound source is a major factor which for example depends on the likelihood of occurrence  for specific tones or sequences of tones.
This project is about motion perception in audition and the impact of the expectation on spatial hearing. Therefore, we investigate the auditory perception of contestants in a anechoic chamber.

Movement and sense of touch

The sense of touch is perhaps the most important sense we have, and which guarantees our survival. Nevertheless, it is perhaps the least explored of all the senses. For example, we know extremely little about how we perceive simple things such as movement, hardness, roughness, etc. with the help of touch. The aim of this project is to improve our understanding of tactile motion perception. This requires the integration of body movement and tactile stimuli. Therefore, the behavior of the participants will be stimulated and analyzed with tactile input devices developed especially for this purpose. Telephone companies and the automotive industry have recognized this deficit in our understanding and have initiated major research projects that aim to bring tactility back into smartphones and intuitive interaction while driving. This project will also contribute to this.

Joint learning of human and machine

Machines are constantly becoming more intelligent and are learning to adapt themselves better to humans. In turn humans are also learning to adapt to machines and are becoming more confident in working with them. However, these learning processes come with a tradeoff between the potential of solving difficult tasks and the risks of working with machines. Therefore the goal of the project is to investigate and analyze the dynamics of these learning processes so that joint learning can be achieved in a fast and safe way. If a machine can learn quicker or more than a human being depends on several conditions. These conditions need be identified and based on them appropriate learning paradigms will be developed.