This ability is maintained only in children that grow up in a bilingual environment, which indicates that bilingualism favours the retention of certain innate perceptive capacities. The study, by Núria Sebastián Gallés, Salvador Soto-Faraco and Jordi Navarra in collaboration with researchers at the University of British Columbia, Vancouver (Canada), has been published in Science on May 25 (DOI: 10.1126/science.1137686).

It is known that human faces are one of the stimuli that most attract the attention of infants, and that, like adults, children, from an early age, can distinguish one language from another when they hear it. But the question arises as to what extent the facial movements associated with talking are relevant to infants who are beginning to perceive language. It was generally held that such “visual” information was fairly irrelevant to verbal communication, since auditory information is sufficient to process language. The research at GRNC shows that this is not the case, since infants of less than six months can discriminate between two languages (for example, English and French) only by observing how the speaker articulates the sounds when speaking one language or the other, in the complete absence of sound.

This capacity to distinguish between languages on the basis of visual information changes over time and in the linguistic context in which the infant grows up. Thus the study shows that the older infants (eight months) do not retain this capacity, with the notable exception of those who grow up in a bilingual environment (English/French). According to Salvador Soto-Faraco “the loss of this capacity for visual discrimination of languages by monolingual children of eight months is a curious result, but at the same time expected, since it is a phenomenon that is also found in other areas of initial language development. For example, they also lose the capacity to distinguish between certain phonemes (when they do not belong to the mother tongue) or the capacity to distinguish between two languages on the basis of the sound (when they are not the mother tongue)”.

According to the authors of the study, this decline with age reflects the fact that certain innate perceptive capacities are only retained if they constitute an advantage for learning the mother tongue. In the case of children who grow up in a bilingual environment, the ability to distinguish between faces that speak English or French is an adaptation to the demands of their particular environment.

Before reaching theses conclusions, the researchers performed a study that consisted in showing the infants ( 4, 6 and 8 months old) a series of silent video clips, in which various speakers recited extracts from the story “The Little Prince” (Saint-Exupéry) in French or English. They used a process of “habituation”, in which initially all the video clips presented corresponded to only one language. When the initial interest of the infants in the images decreased by 60%, they were considered to have become habituated to the visual information. Then, without any pause, the test phase was begun, in which the same faces were shown, in the same order, but reciting phrases from the story in the other language.

To assess the interest shown by the infants in the test phase, the researchers measured the time during which each one paid attention to the screen when a change of language was introduced. They then compared this with the attention time in the control condition in which the language did not change. The recordings were made from a room which was separate from the test room, to which the researchers had access via closed-circuit television cameras. The infants were sitting on the lap of one of their parents, who was wearing dark glasses to prevent them from interfering with the attention of their children.

The results of the study showed that the infants aged less than six months already realised that the speakers had changed language, because their attention time for the video clips was significantly longer when there was a change of language than when there was no change.

This research is part of a global study aiming to identify the capacities of language processing (language discrimination) on the basis of visual information, which has been carried out in joint collaboration between the researchers at GRNC and the University of British Columbia, Vancouver (Canada). A similar study was performed in Barcelona to test this capacity for visual discrimination of languages in monolingual and bilingual adults, using Catalan and Castilian in the tests. The results of this first study were published in Perception and Psychophysics last April.