Human and computer estimations of Predictability of words on written language

BRUNO BIANCHI; FACUNDO CARRILLO; DIEGO F SLEZAK; JUAN E KAMIENKOWSKI; DIEGO E SHALOM

CABA

Conferencia; XLIII Jornadas Argentinas de Informática e Investigación Operativa (43JAIIO); 2014

When we read printed text, we continuously predict the follow words in order to integrateinformation and direct future eye movements to forthcoming words. Thus the Predictabilityhas become one the most important variables when explaining human behavior and informationprocessing during reading. In this study we present results of word predictability inlong Spanish texts, estimated from human responses in a massive web-based task. We usedLatent Semantic Analysis (LSA) as a way to estimate human-based predictability valuescomputationally. We validated the human estimation of predictability with local and globalproperties of the text, and we showed that LSA-distance on adequate timescale capturessome semantic aspects of the prediction.