Abstract
The ability to form rich semantic representations for thousands of words is a defining feature of human cognition. In order to better understand these representations and the way they contribute to lexical processing, researchers have applied various methodologies that produce large-scale databases that quantify the different ways we associate semantic information with words.
We extend this line of research by investigating variables that capture socially meaningful dimensions of semantic representation, in order to map out the socio-semantics of thousands of Czech words. We present ongoing research that aims to collect normative ratings for nouns, adjectives and verbs, along five theoretically distinct dimensions: gender, political alignment, location, valence and age.
Czech has been relatively understudied in terms of the availability of normative ratings, with current datasets limited in terms of size and scope. Moreover, in the past 30 years the population has experienced dramatic social shifts, making it an ideal candidate to study differences and similarities in semantic representations between populations with varying social profiles.