Research
Infants Match Voices and Faces
When we speak, there are facial movements as well as vocal sounds. Both sources of information are important in helping us to understand what a person is saying. A previous study in our lab showed that young children can match the voice of an unfamiliar adult to a video of that same adult. This link between auditory and visual information was recently tested in 6-7-month-old infants. In this study, infants listened to a sample of infant-directed speech. Following that they were presented with 2 silent videos of moms speaking to their infants. One of the videos was the same speaker they heard in the speech sample; the other video was of a mom who matched the speaker on characteristics such as hair color, glasses, ethnicity, etc. Neither mother used the same exact combination of words in the video presentation that the infants heard in the original recorded speech sample. Even so, the infants looked significantly longer at the video of the original speaker than at the video of the other mom. These results indicate that infants are able to match voice characteristics (auditory information) to the facial movements that accompany voice production (visual information). This link between auditory and visual information likely aids infants in the development of language skills
Infants Match Singing Voices and Faces
As mentioned above, we demonstrated that 6-7 month-old infants can match the speaking voice of an unfamiliar adult to a silent video of that same adult speaking; even when the utterances do not match. As a follow-up to that study, we asked whether or not infants can do the same thing – match the voice to the face - when the adult is singing rather than speaking. As in our “talking” study, we first had the infants listen to a sample of a mom singing to her baby. Afterwards, the infants were presented with 2 silent videos of 2 different mothers singing to their babies,one of whom was the mom from the original song sample. Both moms sang a different song in the video presentation than the song the infants heard in the initial recording. During the course of the study the infants gave no evidence that they were able to match the voice of the original mom to her face in the video. This suggests that the link between auditory and visual information that offers clues to our identity when we speak is less strong when we sing. Perhaps untrained singers change their performing style across songs. Interestingly, we found that adults were also unable to perform this task successfully. We are currently investigating if infants are able to identify the mom from the original song sample when she is singing the samesong in the video that she sang in the initial recording.
Children with Cochlear Implants Enjoy Music
Our ability to hear the intricate details of complex sounds such as speech, music, and voices is afforded to us by the remarkable sensitivities of our auditory system. For profoundly deaf children who use cochlear implants (CIs), the hearing experience is drastically different from that of normal hearing children. For them, hearing sensations are partially restored by means of an electronic device that directly stimulates the auditory nerve according to the patterns of speech.
The findings of the following studies, involving children with CIs, help us to understand how the ability to process and understand what we hear is effected by imperfections the hearing itself. In addition, the data will help lead to improvements in the design of cochlear implants so that deaf children using these devices can better experience the range and richness of sounds in our environment.
Although many cochlear implant (CI) children are able to achieve good speech production, many also report enjoyment of music—an unexpected accomplishment given the sparse pitch details transmitted by current CI devices. To gain insight into the nature of their musical experiences, we examined the ability of deaf CI children and hearing children to recognize familiar music. We gathered a selection of popular radio hits and theme songs from children’s television shows, and presented different versions of these songs to both groups of children. The children were asked to identify original versions of songs, instrumental versions without lyrics, and versions which contained only the main melody of the song. Hearing children were able to accurately identify all versions, suggesting that their memory of songs is quite flexible. However, CI children’s ability to identify familiar songs was hampered when the song was changed from its original version; with the melody only version of the songs presenting the most difficulty. These findings suggest that CI children’s memory for songs is less flexible than that of hearing children and that their ability to recognize songs is strongly affected by changes in the features of the songs, such as lyrics and instrumentation.
Familiar Melodies
In this study, we familiarized infants with a short melody by repeating it several times at different pitch levels; sometimes higher, sometimes lower, but always keeping the melody the same. After 2 minutes of this we tested the infants to see if they would show a preference for this now familiar melody over an alternate version of the same melody with one note changed. If the infants showed a preference for the original version of the melody that would tell us that they noticed the change and treated the alternate, changed version of the melody as if it were a new melody. However, the infants did not show a preference for either version of the melody. This does not mean they didn’t notice the changed note. It may mean that although one note changed the infants still considered it to be the same melody, since the rest of it stayed the same. This would suggest that the infants were treating the familiar melody and the melody with one note changed in the same way we treat a melody with variations. For example, we recognize a familiar melody even if it is changed slightly or played with extra notes. In fact, much music is written with one main melody, or theme, which is then changed slightly as the melody repeats. Musicians often refer to this principle as theme and variations. It is possible that young infants are already treating small changes to a melody as a variation on that same melody.
Consonance and Dissonance
Previous studies have shown that the preference for consonant (pleasant sounding) music over dissonant (less pleasant sounding) music is present from birth. However this evidence is based entirely on only two notes played together, which is not typical of music in our culture. We investigated preferences for consonance over dissonance in 6- and 12-month-old infants using simple melodies in one experiment and simple melodies with chord accompaniment in another. In both age groups, infants showed no preference for the consonant melody. When the melody was accompanied by chords, 6-month-olds showed no preference, but 12-month-olds listened significantly less to the consonant music. Our results suggest that the preference for consonance observed in previous studies may be unique to simultaneous combinations of 2 tones. The results also highlight the effects of early exposure to music. Older infants who have had more exposure to music showed a preference for music that was moderately dissonant or distinct from the music ordinarily heard in their environment, suggesting that they may have learned about the structure of the music in their environment.
