|
|
|
|
|
|
|
|
|
|
|
|||||||||
|
|||||||||
|
Stefano
Baldassi, PhD Natural life has evolved by using vision to provide the most useful cues for a successful adaptation of the vast majority of living organisms. Foraging and prey hunting, as well as partner selection, are in fact often based on individuals searching around their environment in order to find the right visual configuration that matches a given template. Specific patterns of textures, motion, colors and other visual features are stored somewhere in the brain to be matched with contingent visual experience for recognition and reaction. Such a template, that I assume to be activated in a top-down fashion, is in many cases genetically inherited and is both flexible and rigid to guarantee reliable rates of success. This might be true for classes of living organisms far away from the complexity of mammals, like reptiles or insects, or even inferior species. Visual search is so strongly rooted in nature that the amazing phenomenon of camouflage has evolved to balance it. When dealing with complex backgrounds and multiple items, the voluntary or automatic activation of selection processes can bias the processing of the scene in advantage of some items at the expenses of others. The mechanisms that cause such bias is generally called "Visual Attention". Such a general framework provides a general clue to what my main research interests are. I have been carrying on with this topic since 1997, right before my PhD, when I was research assistant at the lab of Michael Morgan, at the University College London. In my thesis project, done in Istituto di Neuroscienze del CNR and Rome University “La Sapienza” under the supervision of David Burr, I dealt with the problem of the integration of information in multi-element displays. After my PhD, I worked under the supervision of Preeti Verghese, at the Smith-Kettlewell Eye Research Institute of San Francisco, in research project generated from our respective earlier work, with a common ground in search and attention experiments. In our first project we modeled visual search with multiple targets, generating a version of the SDT Max model called the "Signed-Max" model. Then we started out a new project on the basic mechanisms of attention probed by visual masking which is still continuing. Starting January 2003 I moved back to the Pisa lab based on my position as a research scientist at the University of Florence, Department of Psychology. Currently, my main focus is on a project aimed to develop a novel psychophysical technique for mapping the internal response distributions that determine the observers responses with sparse and crowded multi-element displays. This technique has allowed to demonstrate an apparent paradox of visual decisions, that the uncertainty derived by complex displays relates positively with the proportion of high-confidence errors. |