The essential act of reading is converting a visual pattern into meaning. Most people read effortlessly every day, but how is this achieved? To answer this question, our lab uses a very multidisciplinary approach that includes elements of computation, information theory, behavior and neurophysiology.
Sometimes context provides useful clues to what the next word will be. In cognitive psychology, this is often explored using the lexical decision task with a prime word (e.g., “DOCTOR”) preceding a target word (e.g., “NURSE”). If the prime alters the speed with which the target is recognized, then we infer that the mental representation of the prime and target words interact in some way. Priming can occur between two words that are semantically linked, but it may also occur because one word frequently co-occurs in text with another word. One current research focus in the lab is the measurement of word-word and word-sequence co-occurrences with large samples of text (Brants and Franz’s (2006) terabyte Google Dataset). Based on these measurements, we are examining how expectancy generated from a word or word sequence facilitates recognition of a subsequent word.
A second research line may, at first, seem tangential to the investigation of statistical regularities in text. This research focus is on the role of sex hormones and gender in word recognition. There are several reasons that this research interest is an important and natural part of our focus on word recognition. First, we are keenly interested in creating a concrete model using general principles but applying them to a specific semantic domain. For instance, the conditional probabilities described in Hahn & Sivley (2011) and Hahn (in press) provide a natural weighting scheme for a distributed network model of a semantic domain that could describe and predict behavior. Sex is an obvious choice as a target semantic domain because 1) it may be subjected to large-scale differences between genders and across hormone levels and 2) the Google Dataset provides a description of sex-related text that is uncensored by social desirability. Social desirability is an important challenge when collecting sex-related data in a lab and is apparent when free association norms suggest that ass has a stronger link to donkey than butt (Nelson et al., 2004). As a result, modeling semantic memory for sex based on free associations is unlikely to be successful. Developing a network model of sex and aggression based on statistical regularities is an ongoing project in the lab. Testing hypothesized changes in word recognition produced by hormonal changes is another ongoing project in the lab.