Recognizing voice: the child with autism spectrum disorder

Abstract

Purpose This paper assesses the ability of the minimally verbal child with autism to recognise their own voice. The rationale for this study rests in recent advances in technology aimed at making the voice of speech generating devices (SGDs) sound more like the child using them (van Santen and Black, 2009). The purpose of this paper is to investigate the child’s ability to actually recognise the sound of their own voice in a series of short experiments using computer-based methodology. Design/methodology/approach Using a voice-face matching computerised paradigm, the performance of 33 children with autism was compared to that of 27 children with developmental delay (DD), and 33 typically developing (TD) children. The children were matched for verbal and non-verbal ability and a training period was conducted prior to the main test to ensure children’s understanding of what was expected of them. Findings The findings of this study suggest that the child with autism recognise the sound of their own voice at test, but with much greater difficulty than age-and-ability matched comparison groups. The implications of this finding are useful for researchers in the field of speech mimicry technology and manufacturers of SGD software packages. The paper also provides empirical insights about how the child with autism may process voice in their everyday social interactions. Research limitations/implications Some limitations to this study exist, for instance, there were only a small number of presentations involving self-voice in this task. This may have over simplified the process for the young TD children and the children with DD. Nevertheless, it is striking that despite being matched for non-verbal mental age, the children with autism performed significantly less well than either of the other two groups of children. However, future studies would benefit from adjusting the number of presentations of voice and face accordingly. It is also important to note that for some children with autism the simultaneous presentation of faces and voices may act more as an interference effect (Cook and Wilding, 1997; Joassin et al., 2004) than a facilitation effect (Molholm et al., 2002). Future studies may wish to test a subgroup on voice recognition without the aid of visual prompts. Practical implications The paper includes implications for the type of voice children with autism may prefer to use when communicating via a SGD. The authors suggest that if the child does not recognise or prefer the sound of their own natural voice on such devices, partial or complete abandonment of the SGD may occur. Originality/value This paper fulfils an identified need to research how children’s abilities and preferences can be taken into account at the point of decision making for particular communication tools.