Marilyn Hughes Blackmon
Research Affiliate • Former Research Associate
Institute of Cognitive Science • Blackmon/Polson Lab

Research Synopsis

My most cited research publications prior to 2018 have reported controlled experiments of independent variables that determine success rates for people searching for information on complex websites.  These experiments use multiple regression models and effect sizes to test predictions of the distribution of attention to the webpage and clicks on available hyperlinks.  Predictions of attention distribution and hyperlink clicks have been based on semantic similarity measured by Latent Semantic Analysis (LSA). The research has produced the automated Cognitive Walkthrough for the Web (AutoCWW) and promoted the comparison and synthesis of two complementary simulation models of Web navigation: a) the Comprehension-based Linked model of Deliberate Search (CoLiDeS), a simulation model of Web navigation based on the Construction-Integration Theory of comprehension, and b) SNIF-ACT, a simulation model of Web navigation based on rational analyses of information-foraging theories and developed with the ACT-R cognitive architecture.