Modified COSMIC 6
per hour in unstructured environments. Furthermore, they found that three low-level
communicative functions (i.e., attention seeking, engaging in social routines, and requesting)
accounted for around 60% of the children’s total communication bids, while four high-level
functions (i.e., giving and seeking information, expressing feelings, and engaging in social
interactions) accounted for only 10% of all bids. However, this measure was restricted to
evaluating children’s spontaneous initiations of communication, ignoring any response
behaviours.
Very few studies have included any assessment of peer interaction. McGee,
Almeida, Sulzer-Azaroff and Feldman (1992) evaluated three young children with ASD
interacting with their teachers and peers. While two forms of communication (gestural
and verbal) and two broad facets of communicative function (initiations vs. responses;
and positive vs. negative interactions) were delineated, the researchers only reported on
the initiation and responding behaviours of their participants. Another such instrument,
the Social-Communication Assessment Tool (S-CAT; Murdock, Cost & Tieso, 2007),
measures social communication in four distinct areas: verbal initiations, verbal responses,
joint attention acts, and non-verbal communication attempts. While useful in its inclusion
of evaluation of peer interaction in everyday settings, this instrument similarly lacks the
detail required for comprehensive measurement of those aspects of social communication
frequently affected in youngsters with ASD (i.e., delineation of both the forms and
functions of acts, along with specification of identity of the interaction partner and the
child’s own role in the interaction).
Although a few naturalistic social-communication measures have been developed for use
with language impaired pre-schoolers without ASD (e.g., Kliewer, 1995; Roberts, Burchinal &