Investigating deception in second language speakers: Interviewee and assessor perspectives

Abstract

Purpose. The first of two experiments investigated the effect that speaking in a nonnative language has on interviewees' perceptions of their interview experience. A second experiment investigated evaluators' perceptions of the credibility of interviewees who spoke in their native or non-native language.Method. For the first experiment, 52 participants told the truth or lied about their identity during a mock border control interview. All of the participants were interviewed in English, for half of the sample this was their native language, and for the other half of the sample English was not their native tongue. Post-interview, all participants completed a self-report questionnaire relating to their perceptions of their interview experience. For the second experiment, 128 participants evaluated the credibility of interviewees from the first experiment. The modality of presentation of interview clips was varied and included 'Visual and Audio', 'Visual Only', 'Audio Only', and 'Transcript Only'.Results. Non-native speakers were more likely than native speakers to report being nervous and cognitively challenged during their interviews and were more likely to monitor their own behaviour. Overall, evaluators were better able to distinguish between truth tellers and liars who were speaking in their native language than between truth tellers and liars who were non-native speakers. Relative to native speakers, there was a smaller truth bias for evaluations of non-native speakers. When evaluators were considering the non-native speakers, they achieved higher discrimination accuracy when they were exposed to 'Visual Only' or 'Transcript Only' presentations than when they were shown the 'Visual and Audio' or 'Audio Only' interview clips.Conclusions. Self-reported experiences of a mock border control interview differed dependent on whether interviewees were speaking in their native or non-native language. Discrimination accuracy was better for native speakers than it was for non-native speakers and was at its worst when evaluators heard the accents of the non-native speakers.With increasing global threats to security and high levels of migration, border control interviews are commonplace and it is very often the case that interviewees are required to communicate in a language other than their native tongue. The first experiment presented here focuses on the experiences of truthful and lying interviewees who spoke in their

Language and eyewitness suggestibility

“…Eyewitnesses who give their statement in a non‐native language often need to exert more effort than native speakers. In a recent study by Akehurst, Arnhold, Figueiredo, Turtle, and Leach (), participants interviewed in a non‐native language in a mock border patrol interview reported feeling more nervous and cognitively strained compared with participants interviewed in their native language. Abutalebi (), summing up past research on the production and control of a second language, stated that one explanation of this language effect is that lexical terms in the native and non‐native language compete with each other.…”

Looks like a liar? Beliefs about native and non‐native speakers' deception

“…When expectations about how a speaker should behave are violated, suspicion can be aroused (e.g., Bond et al, ). Non‐native speakers report and overtly display signs of nervousness and cognitive demands during interviews (e.g., speech hesitations and phrase repetition; Akehurst et al, ; Elliott & Leach, ; Gregersen, ). Thus, observers' negative biases when judging non‐native speakers (e.g., Da Silva & Leach, ; Evans & Michael, ) could be explained by this failure to adjust expectations in keeping with speakers' language proficiencies.…”

“…The same consideration does not appear to be afforded to non‐native speakers, however. Across several studies, non‐native speakers are viewed less positively than native speakers, either because they are less likely to be labeled truth‐tellers or they tend to be labeled lie‐tellers (e.g., Akehurst et al, ; Da Silva & Leach, ; Evans & Michael, ).…”

“…In a new phase in the field, researchers have begun to examine the factors unpinning these differences in decision‐making. For example, speakers' accents (Akehurst et al, ), the demands of the deception task (Evans et al, ), and observers' familiarity with non‐native speech (e.g., Evans & Michael, ; Leach, Snellings, & Gazaille, ) have been tested as moderators of proficiency effects. The role of cognitive load has also received significant empirical attention (e.g., Evans et al, ).…”

“…The cognitive load approach to deception detection predicts that differences between lie-and truth-tellers should be exacerbated under these conditions (Vrij, Fisher, Mann, & Leal, 2008). Thus, non-native speakers might simply exhibit more cues to deception due to the heightened cognitive demands associated with both lying and speaking a less proficient language (e.g., Akehurst et al, 2018;Elliott & Leach, 2016). The Psychologically-based Credibility Assessment Tool (PBCAT)-which is employed by minimally trained observers to identify 11 cues to deception-has yielded the most consistent findings about the leakage of non-native speakers' deception (Evans et al, 2013).…”