Edited by: Elena Nava, Università degli Studi di Milano-Bicocca, Italy
Reviewed by: Naiqi G. Xiao, Princeton University, United States; Przemyslaw Tomalski, University of Warsaw, Poland
This article was submitted to Developmental Psychology, a section of the journal Frontiers in Psychology
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
During the first year of life, infants undergo perceptual narrowing in the domains of speech and face perception. This is typically characterized by improvements in infants’ abilities in discriminating among stimuli of familiar types, such as native speech tones and same-race faces. Simultaneously, infants begin to decline in their ability to discriminate among stimuli of types with which they have little experience, such as non-native tones and other-race faces. The similarity in time-frames during which perceptual narrowing seems to occur in the domains of speech and face perception has led some researchers to hypothesize that the perceptual narrowing in these domains could be driven by shared domain-general processes. To explore this hypothesis, we tested 53 Caucasian 9-month-old infants from monolingual German households on their ability to discriminate among non-native Cantonese speech tones, as well among same-race German faces and other-race Chinese faces. We tested the infants using an infant-controlled habituation-dishabituation paradigm, with infants’ preferences for looking at novel stimuli versus the habituated stimuli (dishabituation scores) acting as indicators of discrimination ability. As expected for their age, infants were able to discriminate between same-race faces, but not between other-race faces or non-native speech tones. Most interestingly, we found that infants’ dishabituation scores for the non-native speech tones and other-race faces showed significant positive correlations, while the dishabituation scores for non-native speech tones and same-race faces did not. These results therefore support the hypothesis that shared domain-general mechanisms may drive perceptual narrowing in the domains of speech and face perception.
The first year of an infant’s life is characterized by a fast attunement of perceptual mechanisms to the specific sensory inputs that infants encounter in their daily life. This process, known as perceptual narrowing, leads to a decline in the ability to discriminate or recognize stimuli that are not present or not relevant in the infant’s environment. So far, perceptual narrowing has been observed for visual as well as acoustic perception. In speech, very young infants can discriminate all kinds of speech sound contrasts but this sensitivity declines for many non-native sound contrasts within the first year of life (e.g.,
Similar patterns have emerged in research on infants’ face perception. For instance, the face-sensitive N170 signal showed different properties for upright and inverted faces in adults, but similar properties for both orientations in infants (
The similarities between the perceptual narrowing processes in speech and face perception have led to the suggestion that these domains share some underlying developmental mechanisms (
The present study investigated relations between perceptual narrowing in these two domains by testing the effects of perceptual narrowing in both domains in Caucasian monolingual infants: We tested 9-month-old German learning children on their ability to discriminate same-race and other-race faces, as well as non-native Cantonese tone contrasts in separate experiments using an infant-controlled habituation-dishabituation paradigm.
Fifty-three healthy, full-term Caucasian infants of German origin (
Our study was conducted in accordance with the German Psychological Society (DGPs) research ethics guidelines. The experimental procedures and informed consent protocols were approved by the Offices of Research Ethics at the Universities of Giessen and Potsdam. Written informed consent was obtained from all parents of the infant participants prior to their participation in experiments.
Non-native speech stimuli consisted of Cantonese CV syllables (/tɕhi/) with mid-level (tone 33) or high-rising (tone 25) tone variants taken from the study by
Face stimuli consisted of colored photographs of six other-race Asian (Chinese origin) and six same-race Caucasian (German origin) women, on a white background. All faces were presented in three poses: frontal, ¾ to left, and ¾ to right. On each photo the women looked straight at the camera with the hair, neck, and shoulders being visible. We edited the photos in Photoshop CS3 to make them matched in head size, and also made the skin-tone, eyes, head, and lip-color as similar as possible. Each photo was presented in the middle of the screen, appearing as 12.5 cm (10.98° visual angle) wide and 16.5 cm (14.47° visual angle) tall. Faces were paired within each ethnicity according to similarity ratings collected in a pilot experiment. A follow-up pilot experiment confirmed the occurrence of the ORE in 9-month-old infants using these face pairs.
Parents were informed about the general purpose of the study and the experimental procedure, but were blind to the hypotheses. Parents gave written consent for their child’s participation. During the experiment infants sat on their parent’s lap at a distance of approximately 65 cm from the 23.8″ display with a resolution of 1920 pixels × 1080 pixels, and an integrated Tobii tx300 eye-tracker with a sampling rate of 300 Hz. Parents were instructed to close their eyes and stay silent during the experiment. Each testing started with a 5-point infant calibration procedure. The calibration was repeated until it was successful for all five points for up to four maximum attempts. The data from infants who failed the calibration procedure were excluded from the final sample.
We used an infant-controlled habituation-dishabituation procedure for both the speech and face tasks. Within both habituation and test trials, stimuli were presented until infants looked away from the screen for 2 s, or until a maximum trial length of 40 s was reached. The average looking time during the first three habituation trials served as the baseline for the habituation criterion. The habituation phase continued in sets of three trials, until the average looking time for a set of three trials decreased to below 50% of the average from the first three trials. The habituation phase continued until either this habituation criterion was reached, or until a total of 18 habituation trials had been presented. Infants who failed to habituate (
After habituation, infants proceeded to the test phase, where they were sequentially presented with the habituated stimulus and a novel stimulus of the same type, with each infant being randomly assigned to see the habituated stimulus either first or second. E-Prime version 2.0 (Psychology Software Tools, Pittsburgh, PA, United States) was used for stimuli presentation.
During the speech task infants were habituated to one of the two tones. A silent rotating animation of a colorful circle presented on the screen was used as an attention getter between trials. During test trials, the infants heard repetitions of the habituated tone, and the second (novel) tone in a sequential random order in order to eliminate order effects. A checkerboard pattern was presented on the screen whenever infants heard the tones during both habituation and test trials.
During the face task, infants were habituated to photos of one person in three different poses, alternating in random order in sequences of three. To direct infants’ attention to the screen, a neutral audio signal was played as an attention getter before the start of each habituation trial. During the test trials, the previously habituated and a novel face of the corresponding condition were shown sequentially (in the frontal pose) in random order.
Dishabituation scores for each infant in the speech and face tasks (
Dishabituation scores for the face and speech tasks.
Finally, to determine whether there was a relation between infants’ dishabituation scores in the speech and face condition, we calculated the Pearson correlations between the dishabituation scores in the speech and face tasks for each group of infants.
The
Next, we tested the Pearson correlations (
Correlations between dishabituation scores in the non-native speech and other-race face task
Agreeing with previous research, our results confirmed that 9-month-old monolingual infants were not able to discriminate between non-native tones (
Most interestingly, our results showed that the dishabituation scores of infants for non-native tones and other-race faces were highly correlated, while the dishabituation scores for non-native tones and same-race faces showed no correlation. The positive correlation between the ability to discriminate between non-native tones and other-race faces indicates that infants who are weak in discriminating other-race faces are also weak in discriminating non-native speech and vice versa. Most importantly for the interpretation of this effect, no correlation was found between the discrimination of non-native tones and same-race faces, which indicates that the correlation between the dishabituation scores for non-native tones and other-race faces is not merely an effect of general tasks requirements (e.g., attention, memory, or habituation speed). Our results therefore support the hypothesis that the developmental trajectories of perceptual narrowing in speech and faces share some underlying mechanisms that drive these processes and can affect the speed and/or the outcome of these processes across both domains within an individual. It could well be that these domain-general processes are involved in applying statistical learning to the stimuli surrounding infants, allowing for the specialization of their perceptual systems to the stimuli classes which appear most often (
AK contributed to the design of the work, acquisition and analysis of the data, and drafting of the manuscript. AG contributed to the design of the work and revising of the manuscript. BH contributed to the design of the work, drafting and revising of the manuscript. GS contributed to the design of the work, analysis of the data, and revising of the manuscript. All authors approved the final version and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.