All mobile organisms spend much of their lives searching for something, be it food, water, shelter, or others of their kind. Searching uses perception, memory, cognition, and action. It can occur over vast landscapes and periods of time, as in whale migration behaviors. It can occur in under a second on the basis of fleeting bits of sensory evidence, as in frogs searching their visual fields for flies. And it occurs on all scales in between, across organisms but also across different behaviors of a given organism. Human search is especially diverse in this regard because human searches range from subatomic to human to cosmic scales of exploration.
With support from the National Science Foundation, Dr. Christopher Kello is leading a team of researchers at the University of California, Merced, in the study of human search behaviors in three very different but complementary domains: Foraging over large virtual spaces, visual searches over scenes and movies, and memory searches over words and concepts. The generality and evolutionary importance of search suggests that search functions may share common principles across scales and domains. Evidence for this conjecture has been found in "Levy distributions" that describe the frequencies with which segments of different lengths occur in search paths. These frequencies are observed to obey a common scaling law across a wide range of different search behaviors but current evidence is not sufficient to determine the meaning of this law.
With mentorship from three faculty members in Cognitive and Information Sciences, undergraduate and graduate students and a postdoctoral fellow will collaborate on experiments and computational models in each of the three domains of interest, in a unique and highly interdisciplinary education and research experience.
Search is one of the most fundamental and important cognitive functions for humans and other intelligent organisms. Animals and people need to search for food and shelter to survive, for instance, but they also need to search their visual environments and memories for information in order to plan, communicate, and act. All these search functions are kinds of foraging, but researchers do not know if there are principles of cognition common to all of them. Recent theories of search suggest that there may be commonalities among efficient search behaviors--that is, patterns of search that save time or energy or both. However, the generality of these theories is not yet known, partly because researchers have only begun test them in the domains of vision, memory, and communication. With NSF funding, common principles of foraging search functions were investigated by a team of undergraduate students, graduate students, postdoctoral researchers, and professors. The team conducted a range of experiments in visual search, memory search, and communication. They studied how people find conceptual information in long-term memory, and how this search process resembled the way eyes find and act upon desired information in visual scenes. They also created a web-based search video game that allowed to collect search data from thousands of people around the world. The team found that search behaviors and their related dynamics do indeed have common patterns across these different cognitive domains--patterns that are similar to how animals forage for food and other resources. However, these patterns could not be explained by prior theories and models of foraging. Instead, they built upon prior theories to develop a new computational model that was used to simulate how people played the web-based search game. The model accounted for common patterns in people's search behaviors, and it can be applied towards findings from memory search and animal foraging as well. These findings advance our understanding of search functions and may help to develop better search technologies, and better therapies for memory and attentional disorders that impair search processes.
