Modelling the Noise-Robustness of Infants' Word Representations: The Impact of Previous Experience

Abstract

During language acquisition, infants frequently encounter ambient noise. We present a computational model to address whether specific acoustic processing abilities are necessary to detect known words in moderate noise--an ability attested experimentally in infants. The model implements a general purpose speech encoding and word detection procedure. Importantly, the model contains no dedicated processes for removing or cancelling out ambient noise, and it can replicate the patterns of results obtained in several infant experiments. In addition to noise, we also addressed the role of previous experience with particular target words: does the frequency of a word matter, and does it play a role whether that word has been spoken by one or multiple speakers? The simulation results show that both factors affect noise robustness. We also investigated how robust word detection is to changes in speaker identity by comparing words spoken by known versus unknown speakers during the simulated test. This factor interacted with both noise level and past experience, showing that an increase in exposure is only helpful when a familiar speaker provides the test material. Added variability proved helpful only when encountering an unknown speaker. Finally, we addressed whether infants need to recognise specific words, or whether a more parsimonious explanation of infant behaviour, which we refer to as matching, is sufficient. Recognition involves a focus of attention on a specific target word, while matching only requires finding the best correspondence of acoustic input to a known pattern in the memory. Attending to a specific target word proves to be more noise robust, but a general word matching procedure can be sufficient to simulate experimental data stemming from young infants. A change from acoustic matching to targeted recognition provides an explanation of the improvements observed in infants around their first birthday. In summary, we present a computational model incorporating only the processes infants might employ when hearing words in noise. Our findings show that a parsimonious interpretation of behaviour is sufficient and we offer a formal account of emerging abilities.

Fig 1. The model in learning mode.

Input is presented as speech-meaning pair. After acoustic preprocessing, the memory is adapted to better accommodate the new learning experience.

Fig 2. The model in test mode.

Input (top left) is presented without meaning information, which has to be reconstructed using the fixed internal memory. The resulting activations are transformed into listening preferences.

Fig 3. Simulated listening preferences for experiments where the test speaker is known.

The two different assessment criteria, general matching-based preference of target word over foils and word recognition-based preference are depicted separately. In all panels and for each experiment the left bar depicts listening preferences without added noise, the middle bar corresponds to 10 dB SNR and the right bar to 5 dB SNR.

Fig 4. Simulated listening preferences for experiments where the test speaker is unknown.

The two different assessment criteria, general matching-based preference of target word over foils and word recognition-based preference, are depicted separately. In all panels and for each experiment the left bar depicts listening preferences without added noise, the middle bar corresponds to 10 dB SNR and the right bar to 5 dB SNR.