Toddlers are notoriously difficult to understand, yet like accented speakers, their productions tend to systematically differfrom adult productions. Thus, we hypothesize that listeners with routine exposure to toddlers (perhaps even toddlersthemselves) should comprehend toddlers best. Three listener groups were tested on their comprehension of toddlersutterances in an eye-tracking study; forty-eight toddlers (Mage= 33 months), sixteen undergraduates with little experiencearound children (Mage= 18 years), and sixteen mothers of young children (Mage=38 years). All listeners looked longerto targets than distractors (p ¡ 0.05), with mothers and undergraduates target fixations significantly greater than toddlers(p ¡ 0.001). Mothers target fixations (78%) did not differ significantly from undergraduates (74%; p = 0.17). Thesepreliminary findings suggest a complex picture regarding the role of experience in comprehending toddlers utterances.Clearly, however, toddlers do not outperform adults in understanding toddlers.

Young childrens speech pronunciations deviate systematically from adult forms. For example, onsets are often simplified(e.g., stop becomes top), unstressed syllables frequently deleted (e.g., spaghetti becomes getti), and certain segments arecommonly replaced with other ones (e.g., rice becomes wice). The current study examined how well adults and a popularautomatic speech recognition system (i.e., Siri) deal with these deviations. The same 12 children were recorded producing32 words in isolation at three ages: 2.5, 3.5, and 5.5 years. 12 adults were also recorded. These recordings were presentedto 48 young adults, 7 mothers, and Siri for transcription. All listeners performed worst with 2.5-year-old productions,and humans outperformed Siri with all ages (p ¡ 0.001). Mothers demonstrated the highest accuracy with 2.5-year-oldproductions (86%). Additionally, Siri made distinctive transcription errors with childrens speech. These errors may reflectthe systems lack of training with young childrens voices.