Feeling the Way to Words: Parents’ Speech and Touch Cues Highlight Word-To-World Mappings of Body Parts

ABSTRACT

Touch cues might facilitate infants’ early word comprehension and explain the early understanding of body part words. Parents were instructed to teach their infants, 4- to 5-month-olds or 10- to 11-month-olds, nonce words for body parts and a contrast object. Importantly, they were given no instructions about the use of touch. Parents spontaneously synchronized the location and timing of their touches on the infant’s body with the body part nonce word that they were teaching. Their unrelated, mismatched words and touches did not show synchrony. Moreover, when parents used this synchronized speech and touch input, infants looked more frequently to the target locations on their bodies and to their parent’s faces compared to when the input was mismatched or only speech. Similar to other research on parents’ use of speech and gesture input, these results show how parents use multimodal input and raise the possibility that infants may use touch cues to segment and map early words.

Acknowledgments

Jess Morra, Skyler Finning, Brittany Bowers, and a team of undergraduate research students assisted with recruiting and data collection. We thank the families for participating, the community locations that allowed us to recruit, and the anonymous reviewers who provided very helpful comments on previous versions of the manuscript.

Disclosure

No potential conflict of interest was reported by the authors.

The underlying research materials for this article can be accessed at DOI 10.17605/OSF.IO/VSWAF.

Notes

¹ Touching the infants’ hand while labeling the object was categorized as a Mismatch given our primary focus on the body part targets, elbow and knee, and that the “body” definition included the hand. These occurrences might create particular cases of cross-situational word learning, for example, the parent labels the object, while the infant is holding the object, and touches the infant’s hand and the object. The complexity of these cases, and the variability across parents in how they used the object, led us to take an extra-cautious approach to categorizing touches to the infant’s hand during the object trial.

² Our method for measuring the synchrony between touches and word tokens is broadly similar to the method used by Gogate and colleagues in their structured tasks. One specific difference is that our measure of synchrony is calculated using a Praat script that references the midpoint of the word token relative to the onset and offset of touch tokens instead of using judgments of informed observers about the onsets of touch (and gesture) and word tokens. In contrast to these structured tasks, Nomikou & Rohlfing (2011) used a strict coding method in their study of mothers changing diapers on their 3-month-olds. Specifically, they only examined segments during which information from both modalities, action, and speech, was present; they did not examine delays between onsets and offsets. This same study, however, also included a rich qualitative analysis in which informed observers identified more complex cases of potential synchrony (see also Zukow, 1989; Zukow-Goldring, 1997). Future research should consider the practical consequences and theoretical impact of using the different approaches and combining them when appropriate.

Additional information

Funding

Funding was provided by Bucknell University and Agence Nationale de la Recherche [ANR-10-IDEX-0001-02 PSL*, ANR-10-LABX- 0087 IEC].