Iconicity Influences How Effectively Children with Autism Use Pictures As Symbols in a Search Task
In the absence of spoken language, many low-functioning children with autism (CWA) are taught to communicate using pictorial symbols. Previous studies examining the learning and generalisation of word-picture relations suggest that CWA may have difficulties understanding the symbolic nature of pictures. Here we investigate the ability of CWA to contextualise symbolic information communicated by pictures in a search task that did not involve word learning. Based on evidence from typically developing children (TDC), we anticipated that this ability may be influenced by iconicity (the extent that a picture resembles its referent) and/or access to linguistic representations corresponding to depicted referents.
Our objectives were a) to test whether CWA and TDC differ in their ability to contextualise pictures in a symbolic search task, and b) to identify whether their success is mediated by pictorial iconicity or access to labels for depicted referents.
Sixteen CWA (M age = 9.9 years) were matched to 16 TDC (M age = 3.6 years) on receptive language (CWA: M = 3.6 years; TDC: M = 3.6 years). Children were introduced to a small toy character and a unique set of 4 occluders (various upturned buckets) that were individuated by familiar nameable objects or unfamiliar unnameable objects. The experimenter demonstrated the toy hiding underneath each occluder before presenting 4 pictures, one depicting each occluder. The experimenter highlighted the picture-referent correspondences and clearly asserted the intended purpose of the pictures. Out of the child’s view, the toy was concealed underneath one of 4 occluders. Children were shown a picture of the hiding location and then searched for the toy. Over 3 sessions, children completed trials with 3 picture types – colour photographs, black-and-white line drawings and abstract colour pictures.
The results revealed zero group differences; neither CWA nor TDC were influenced by familiarity of depicted referents, and both groups’ errorless retrieval rates (ERR) were above-chance with colour photographs (CWA ERR: 74%; TDC ERR: 81%), line drawings (CWA ERR: 58%; TDC ERR: 61%) and abstract pictures (CWA ERR: 62%; TDC ERR: 67%). However, both groups made significantly more errorless retrievals in the most-iconic photograph trials (F = 12.93, p < .001), and performance was universally predicted by receptive language. Children with relatively low receptive language (M = 2.9 years) only performed above-chance in photograph trials, while children with relatively high receptive language (M = 4.6 years) performed above-chance with all 3 picture types.
In favourable experimental conditions, CWA can contextualise mental representations of pictures and use them to adaptively guide their behaviour. However, this ability is significantly influenced by iconicity and receptive language. The lack of a referent familiarity effect shows that CWA were not reliant on labelling depicted referents, suggesting that their linguistic development had a passive, yet critical, influence on their pictorial understanding. Overall, these results indicate that the picture comprehension difficulties experienced by CWA relate primarily to word-picture-object mapping, rather than picture-object mapping per se. Furthermore, our findings provide a data-ground rationale for delivering picture-based communication training with colour photographs.