Developing the productive emotion vocabulary test (PEVT) for adult speakers of English as an additional language

Abstract

Abstract Emotion vocabulary (EV) is integral to language proficiency, yet it is rarely assessed, and existing tools are largely decontextualised and unvalidated for adult users of English as an additional language (Lx). We introduce the Productive Emotion Vocabulary Test (PEVT), a vignette-based free-recall task eliciting up to three responses. Twenty-two pictorial-text vignettes, redesigned from materials validated with German children, were administered to 107 adults, alongside LexTALE and a language background questionnaire. Eighteen vignettes elicited the target emotion or a close synonym as the dominant response, with substantial naming agreement. In mixed-effects analyses, context length did not affect accuracy, whereas response order, LexTALE and produced-word frequency did: first responses were more likely to be accurate than subsequent responses, and higher LexTALE scores and higher-frequency labels were associated with greater accuracy. Findings provide preliminary validity evidence for the PEVT as a research instrument capturing productive EV breadth and a narrowly operationalised facet of depth: contextual naming alignment.