Modified COSMIC 7
Bailey, 1994), the lack of comprehensiveness in such tools is also apparent. Kliewer (1995)
employed qualitative methodology but provided only descriptive data on the use of the children’s
pragmatic communication in inclusive school settings. Roberts et al. (1994) assessed
communication in unstructured play settings but only considered communicative form in terms
of the number of different words used by children and their mean length of utterance. The need
for a comprehensive tool evaluating forms, functions, roles, and social partners across a range of
everyday settings remains.
3. Development of the COSMIC and rationale for modification
Recognising this need, Pasco, Gordon, Howlin and Charman (2008) recently
developed the Classroom Observation Schedule to Measure Intentional Communication
(COSMIC) to assess effectiveness of a PECS intervention (Howlin, Gordon, Pasco,
Wade, & Charman, 2007). COSMIC sampled the behaviour of (mainly) non-verbal, low-
IQ children with autism in special education classrooms, across 15-minute periods of
video-taped interactions. Codes included various forms and functions of communication,
and delineated social interaction partners and the children’s own roles within the
communication acts. Unlike existing tools, COSMIC coding delineated each of these
facets independently1, considering both teacher and peer interaction. Pasco et al. (2008)
assessed 91 children aged 4 to 11 during snack time and various other activities
(including one-to-one and group teaching, free play, etc.). COSMIC codes were
compared to ADOS-G item scores for the assessment of concurrent validity, with
moderate to high correlations evident between some but not all corresponding items.
COSMIC rates of Initiation acts and Use of PECS were found to be sensitive to change
following implementation of the PECS intervention (Howlin et al., 2007). Predictive