The present study provides normative measures for a new stimulus set of images consisting of 225 everyday objects, each depicted both as a photograph and a matched clipart image generated directly from the photograph (450 images total). The clipart images preserve the same scale, shape, orientation, and general color features as the corresponding photographs. Various norms (modal name and verb agreement measures, picture–name agreement, familiarity, visual complexity, and image agreement) were collected separately for each image type and in two different contexts: online (using Mechanical Turk) and in the laboratory. We discuss similarities and differences in the normative measures according to both image type and experimental context. The full set of norms is provided in the supplemental materials.Electronic supplementary materialThe online version of this article (10.3758/s13428-018-1028-5) contains supplementary material, which is available to authorized users.

Contemporary research on aging has provided mixed evidence for whether older adults are less effective than younger adults at designing and delivering spoken utterances. However, most of these studies have focused on only specific aspects of this process. In addition, they tend to vary significantly in terms of the degree of complexity in their chosen stimuli or task. The present study compares younger and older adults’ performance using a referential production paradigm involving simple everyday objects. We varied referential context such that a target object was either unique in its category (e.g., one shirt), or was accompanied by a same-category object (e.g., two shirts). We evaluated whether speakers’ descriptions provided listeners with sufficient information for identification, and whether speakers spontaneously adapt their speech for different addressee types (younger adult, older adult, automated dialogue system). A variety of measures were included to provide a comprehensive perspective on adults’ performance. Interestingly, the results revealed few or no age differences in measures related to production performance (speech onset latency, speech rate, and fluency). In contrast, consistent differences were observed for measures related to descriptive content, both in terms of informativity and variability in lexical selection: Older adults not only provided more information than necessary for referential success (e.g., superfluous modifiers), but also exhibited greater variability in their selection of modifiers. The results show that, although certain aspects of the production process are well-preserved across the adult lifespan, meaningful age-related differences can still be found in simple referential tasks with everyday objects.