Modified COSMIC 12
4.4 Procedures
Behaviour Sampling and Filming
Groups were filmed for around 15 minutes (M = 17.3, SD = 2.6), comprising a 5-minute
teacher-led activity (ACT) followed by 10 minutes of free play (FP). Participating peers were
classmates who had known the child with ASD for at least two months and whose parents had
consented to filming. Participant teachers were either the regular classroom teacher or an
assistant/support worker regularly involved with the child with ASD. Groups were filmed in a
quiet location (e.g., an unused classroom, or whilst other classmates were outside) so that
vocalizations could clearly be heard. The standardised set of highly motivating items provided by
the research team comprised: (for use during the ACT) a colouring book with crayons and a large
jigsaw puzzle; and (for use during the FP) a Jack-in-the-Box; two small cars; a slinky spring; two
tubs of Play-Doh; a marble run; a plastic birthday cake with associated materials (candles, knife,
etc.); a tea set (with cups, bowls, plates, etc.); a textured story book; a bubble gun with liquid;
and two soft balls.
Minimal instruction was given to maintain naturalistic group interaction. Teachers were
asked to first engage the children in the ACT before proceeding to FP4. For the ACT, teachers
were asked to give instructions and feedback as they would in a typical focused classroom
activity, whilst having the children complete the jigsaw or a page of colouring-in. For the FP,
they were asked to allow the children to engage in relatively unstructured play, encouraging
involvement of all the children but minimising direct instruction. Teachers understood that the
aim was to observe the child with ASD in his/her natural setting. While any interruptions to the
protocol were noted, filming was continued as normal, with the exception of two cases where the