While many studies investigate how eyewitness identities such as gender, race, and dialect might affect evaluations of the eyewitness’s credibility, lexical dimensions of these statements are often overlooked. Properties of language production such as lexical diversity can reveal in- formation about a person’s language status or other identities, serving as indirect proxies for those attributes when explicit information is unavailable. Previous research has found that a higher number of total (volume) and unique (abundance) words, lower type-token ratio (variety- repetition), and higher proportions of “vivid” descriptors are linked to veracity (e.g., Colwell et al., 2007; Morgan et al., 2013, 2015). However, it is unclear how more complex dimensions of lexical diversity might be related to perceived credibility. In this study, mock jurors (n = 64) evaluated written eyewitness statements for perceived accuracy, credibility, deceptiveness, eloquence, and prestige. The statements, evaluated across six dimensions of lexical diversity (volume, abundance, variety-repetition, evenness, disparity, and dispersion), were written by L1 and L2 English speakers. Quantitative analyses indicated that volume, abundance, variety-repetition, evenness, and disparity predicted perceived integrity (accuracy, credibility, and deceptiveness), while abundance, variety-repetition, evenness, and disparity predicted perceived status (eloquence and prestige). When controlling for lexical diversity, L1 statements were nevertheless rated more positively than L2 statements. These results highlight how juror perceptions may be biased even absent explicit demographic information. Furthermore, they identify the lexical aspects of a text on which jurors may rely when evaluating the integrity of a witness, with implications for the study of human error in veracity judgments.
Kelly Kendro (Wed,) studied this question.