Where accommodation to non-native speakers doesn’t happen

Foreigner-directed speech (FDS) is a classic example of clear speech, and studies have demonstrated accommodation effects like hyperarticulation (Uther 2007), slower speech (Scarborough et al. 2007) or pitch differences (Smith 2007) when addressing non-native speakers. However, earlier work suggests that FDS accommodation does not always occur (Zuengler 1991). This may be because previous work used laboratory settings, often with imaginary interlocutors (Bradlow 2002, Picheny et al. 1986, inter alia); FDS accommodation was found primarily in task-oriented genres, such as giving map directions (Scarborough et al. 2007, Smith 2007). It thus remains an open question whether FDS generalizes to other situations like casual conversations. In this study we examine FDS in an ecologically valid setting. We used a spoken corpus of 991 4-minute speed-dates, creating a balanced sample of over 20 female speakers, each interacting with 4 native speakers of English and 4 non-native speakers. We compared the speech directed to each group; our measurements included rate of speech (words and phones per second, and words per utterance), f0, and word frequency. Controlling for individual variation, all analyses showed no significant difference between the two groups