Audiovisual integration in noise by children and adults

A Self-Organized Artificial Neural Network Architecture for Sensory Integration with Applications to Letter-Phoneme Integration

“…What is thus shown is the fraction of simulation runs when the MMSON stabilized with a correct classification. When combining the information in this plot with the information in Figure 10, one sees also the qualitative similarity between the last finding of Barutchu et al (2010) and the results from our architecture; the classification rate of the bimodal module remains high even though the outgoing activity of the model decreases as noise is moderately increased (up to 20% to 30%, say).…”

“…This is a well-known psychophysical result. Barutchu et al (2010) examined the effect of audiovisual integration in noise on reaction times. This recent study contains three interesting findings that can also be put in close relation to our artificial neural network architecture's operation; reaction time length seems to be an increasing function of the level of noise, reaction times are shortened with multisensory stimuli even when they are noisy, and at moderate levels of noise, when multimodal information is available, the main effect is this lengthening on reaction time while accuracy remains high.…”

Audiovisual Integration in Children Listening to Spectrally Degraded Speech

“…Nevertheless, whether additional visual cues improve children's perception of degraded (e.g., noise-vocoded) speech is one key area that is yet to be explored. Based on evidence showing that audiovisual integration in noise develops as a function of age (Barutchu et al, 2010;Ross et al, 2011), the same may be true when the auditory speech signal is spectrally degraded.…”

“…Although speech perception is improved in children from 4 years of age when visual information is also present in a quiet environment (Massaro, 1984;Massaro, Thompson, Barron, & Laren, 1986), the magnitude of gain is smaller than that seen in adults (Desjardins & Werker, 2004;Hockley & Polka, 1994;Massaro et al, 1986;McGurk & MacDonald, 1976;Sekiyama & Burnham, 2008). Similarly, typically developing children up to 14 years of age benefit less than adults from observing visual articulations when speech sounds are presented in noise (e.g., Barutchu et al, 2010;Ross et al, 2011;Wightman, Kistler, & Brungart, 2006), but the benefit grows with age and continues to develop into late adolescence. Ross et al (2011) suggested that this developmental trajectory reflects not only increasing exposure to audiovisual speech in noisy environments but also the maturation of other perceptual and cognitive abilities that emerge concurrently with the development of audiovisual integration skills.…”

“…In listeners with normal hearing, we expected to replicate existing findings that young children (between 4 and 7 years of age) require greater spectral resolution (i.e., more frequency bands) to identify the speech signals at the same level as adults and older children ages 8-11 years (Eisenberg et al, 2000). In addition, if the same underlying mechanisms support audiovisual integration when the speech signal is either masked by background noise or is itself degraded by noise vocoding, we expected the following: On the basis of the finding that children benefit less from accompanying visual information when auditory speech is presented in noise (e.g., Barutchu et al, 2010;Ross et al, 2011), young children would show a reduced amount of benefit in comparison to both older children and adults when noise-vocoded sentences were presented simultaneously with corresponding mouth movements (Petersen & Posner, 2012;Posner & Petersen, 1990). …”