Semantic memory updating due to the COVID-19 pandemic from a large-scale word association study

LAURINO JULIETA; DE DEYNE, SIMON; CABANA, ALVARO; KACZER LAURA

Congreso; Congreso de la Sociedad Argentina de Neurociencias; 2021

We are interested in analyzing how semantic networks are modified when memory is updated at the collective level. In particular, we will focus on a recent example, from the COVID-19 pandemic, which generated a large number of new meanings for familiar words, such as 'bubble' (now used to refer to a group of people with whom one has frequent social contact). A widely used tool to study internal representations and processes involved in word meaning is the word association task, in which a word (i.e., the cue) is presented and you have to respond with the first word that comes to mind (i.e., the response). In this study, we performed a large-scale experiment of word associations that already had pre-pandemic data, Small World of Words, SWoW (De Deyne et al., 2019) and analyzed the modifications in the representations of 1000 words. First, we determined the frequency and entropy for ‘pandemic’ words (words that incorporated new meanings or changed their frequency) and control words that were given as responses, at two different times: before pandemic (t0) and during pandemic (t1). Preliminary results show that pandemic words present greater differences in frequencies and entropy between t0 and t1, compared to control words. In addition, we will measure semantic similarity within each word at the two time points implementing a decaying random walk process. We predict that pandemic words will present smaller similarities between the two time points, compared to control words. (aclararía por qué esperamos esto).