FRANKFURT. Anyone who has ever looked for milk in somebody else's house knows that they have to go into the kitchen, open the refrigerator and look in the compartment in the fridge door. Even little children know which objects to find where within a room. Cognitive Psychologist Melissa Vo concerns herself with the question of how this scene processing ability develops. The 33-year old professor for Cognitive Psychology was recently appointed to Goethe University from Harvard Medical School. In addition, as a grant recipient of the German Research Foundation’s Emmy Noether Program, she has established the "Scene Grammar Lab" at the Psychology Institute.
"Most people take the ease with which they orient themselves in their environment, perceive and interact with objects, for granted." says Melissa Vo. She first learnt that this is by no means the case when pursuing her dissertation at the Ludwig Maximilian University in Munich, where she worked, among others, with physicists and engineers on artificial intelligence for technical systems. "While a child can find their favourite stuffed animal under a blanket without any difficulty, this presents an almost insurmountable challenge to a robot or a computer vision algorithm," she says.
While natural scenes are complex, their structure follows certain regularities that the human brain clearly learns very early. Perception is therefore greatly impacted by the knowledge of the arrangement of objects in space. For instance, most objects lie on a horizontal plane. For this reason, if experimental subjects are shown images of floating objects, the resulting irritation is expressed in altered brain signals. "The EEG then shows responses similar to when subjects hear or read a grammatically incorrect sentence," says Vo. Similarly, she was able to determine marked deviations in gaze patterns with the help of eye-tracking systems.
These observations suggest that scene processing might rely on mechanisms similar to knowledge about grammatical structures in language or even music. Melissa Vo hopes, therefore, to develop diagnostic tools for the early recognition of children who suffer from reading and writing difficulties, for example, by testing their implicit scene processing abilities before they enter formal schooling. Moreover, her work is also significant for the development of technical assistance systems that could, for example, be intended to support the elderly at home. Together with her 3 doctoral candidates, she is currently setting up a perception laboratory in Frankfurt. The experimental subjects will not only be tested in front of computer screens, as previously, but will also have to search actively for objects and interact with them in real-world scenarios using mobile eye-tracking glasses.
The Munich born and bred researcher is the daughter of a Vietnamese father and American mother. Frankfurt is an interesting research location for her because of the many co-operation partners in town, e.g. at the Institute of Psychology, the Frankfurt Institute for Advanced Studies (FIAS), or the Ernst Strüngmann Institute (ESI) for Neuroscience in Cooperation with the Max Planck Society. She has other partners at the Universities of Darmstadt, Gießen and Marburg. Vo, who has worked in the USA for the past 5 years, also has established an extensive network of contacts at American Universities, which her students can benefit from. "When I was still doing my undergraduate work, my Professor arranged a research internship for me at the Columbia University in New York City. In my opinion, the experiences from that time were extremely formative and gave me more perspective," says the researcher. Together with her American colleagues, she has also organized the international OPAM conference (on Object Perception, Attention, and Memory) to promote young scientists in the field of Cognitive Psychology.
Information: Prof. Dr. Melissa Vo, Cognitive Psychology I, Campus Westend, Tel.: (069) 798 35342, mlvo@psych.uni-frankfurt.de.
Diese Pressemitteilung in deutsch lesen »
