Infant Adaptations to Bilingual Environments

• In Event: Attention, learning and memory in infancy

Abstract

Infants adapt to the external world by sampling (exploring) it. The more variable the environment, the more sampling (exploration) is required. Infants who regularly hear two or more languages are likely to be exposed to more varied language input than infants who regularly hear one language. These ‘bilingual’ infants are also likely to receive less input from each language. Given these exogeneous sampling constraints, how do bilingual infants keep pace with their monolingual peers? We propose that they do so by placing more weight on gathering information (exploring) and rapidly orienting towards speakers. Whereas monolingual infants may be drawn to familiar stimuli so they can build detailed representations of their environment, bilingual infants may err on the side of exploration and collect more samples from their more varied environments. These would include visual as well as auditory samples, because visual input facilitates language learning and is easier for preverbal infants to manipulate than auditory input. In support of our proposal, we found that infants raised in bilingual homes switch attention between visual stimuli faster and more frequently than infants raised in monolingual homes. But this difference could be due to non-language (e.g., cultural) differences, rather than language environment per se. To rule this out, we investigated whether infant measures of attention switching could be associated with a continuous measure of language experience. We hypothesised that the less exposure to English infants raised in the UK have (1) the faster they abandon attention to a visual stimulus in order to shift attention to a novel stimulus and (2) the more often they switch attention between two visual stimuli.

Fifty-six 7-9-month-olds from bilingual homes were administered two eye tracking tasks: the gap-overlap task (Atkinson et al., 1988; Johnson et al., 1991) and a graded change detection task. In the gap-overlap task, infants were presented with a central fixation stimulus followed by a peripheral target. The time it took the infant to saccade to the peripheral target was measured. At the onset of the peripheral target, the central stimulus either remained onscreen (the overlap condition) or disappeared (the baseline condition). In both conditions, the infant had to switch attention. But in the baseline condition, the infant also had to disengage attention from the central stimulus in order to shift it. Therefore, by subtracting saccadic reaction times (SRTs) in the baseline condition from SRTs in the overlap condition, we can measure how quickly infants abandon visual processing of one stimulus in order to shift to a new stimulus. In the second task, two visual stimuli were presented. One stimulus gradually changed across trials. The number of times the infant switched looking between stimuli was measured.

Infants whose language environment was dominated by English were slower at disengaging (Table 1) and switched attention less often (Table 2). Crucially, we will present the moment-by-moment exploratory looking data and time series analyses to augment our understanding of the relationship between early language experience and attentional development – and to elucidate the mechanisms through which infants’ developmental trajectories diverge.

Authors

• Dean D'Souza, City University of London
  Presenting Author

• Dan Brady

• Jennifer X Haensel

• Hana D'Souza, Cardiff University