![]() For instance, some words co-occur (i.e., collocate) frequently with specific other words. In everyday language, many words keep specific company ( Firth, 1957 Sinclair, 1991). Next, we develop these arguments in more detail and report two experiments, using evaluative priming procedures ( Fazio et al., 1986) as a measure of implicit associations.īiases in Language and Associative Learning If the company words keep is sufficient to elicit implicit associations, words like cause should elicit implicit negative evaluations despite their neutrality on explicit measures of valence. For example, the verb cause is rated as neutral in explicit measures of valence but typically occurs alongside negative words (the most common noun collocates within four words to the right are death, problems, damage, pain, cancer, trouble, concern, disease, effect, harm Hauser and Schwarz, 2016). The present studies avoid this ambiguity by drawing on semantically prosodic words that satisfy two criteria: (1) human raters consider them neutral words that do not have clear positive or negative valence, thus indicating that humans do not endorse these words as being positive or negative, but (2) corpus analyses indicate that these words overwhelmingly keep valenced company. This ambiguity renders it difficult to determine whether word embeddings exert a causal influence on implicit associations, as Caliskan and Lewis (2020) highlighted. Unfortunately, words’ biased company is also related to propositional knowledge, and consciously endorsed beliefs may guide both language use and implicit associations. This observation invited the hypothesis that biased company in language may be a key contributor to implicit bias in human minds (for a review, see Caliskan and Lewis (2020)). For instance, female-related words are more likely to occur in family related contexts than are male-related words, whereas male-related words are more likely to occur in science-, technology-, engineering-, and math-related contexts than are female-related words ( Caliskan et al., 2017 Lewis and Lupyan, 2020). In natural language use, the company many words keep is biased in ways that mirror human implicit biases. ![]() “You shall know a word by the company that it keeps.”Īs Firth (1957) noted, words tend to keep certain company, and the company they keep guides how we understand them ( Landauer and Dumais, 1997 Mikolov et al., 2013). Societal implications of the causal embedding hypothesis are discussed. That neutral concepts with valenced collocates parallel the influence of valenced concepts suggests that their collocations in natural language may be sufficient for fostering implicit bias. Throughout, neutral primes with positive (negative) collocates facilitated the evaluation of positive (negative) targets much like strongly valenced primes, whereas neutral primes without valenced collocates did not. We use such semantically prosodic concepts to test the influence of collocation on the emergence of implicit bias: do neutral concepts that frequently collocate with valenced concepts have corresponding implicit bias? In evaluative priming tasks, participants evaluated positive/negative nouns (Study 1) or pictures (Study 2) after seeing verb primes that were (a) strongly valenced (e.g., hate and comfort), (b) neutral in valence but collocated with valenced concepts in corpora (e.g., ease and gain), or (c) neutral in valence and not collocated with valenced concepts in corpora (e.g., reply and describe). For example, when asked, most people evaluate the concept “cause” as neutral, but “cause” is frequently followed by negative concepts (e.g., death, pain, and trouble). Concepts that are themselves neutral in valence but nevertheless collocate with valenced concepts provide a unique opportunity to address this question. It is unknown whether such collocations merely reflect societal biases or contribute to them. White American) names collocate with negative (vs. science-) related concepts, and African American (vs. Societal biases often emerge in these collocations e.g., female (vs. In everyday language, concepts appear alongside (i.e., collocate with) related concepts.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |