The relationship of cognitive style variables and conditions of test administration was investigated in cognitive assessments of hearing-impaired children, aged six through eleven. One hundred-twenty children were given the Raven Coloured Progressive Matrices (CPM) and a Piagetian battery under one of six conditions of testing: (1) standard; (2) simple feedback; (3) verbalization after problem solution; (4) verbalization during and after solution; (5) differential feedback; and (6) conditions 4 and 5 combined. Kagan's Matching Familiar Figures Test (MFFT) and the Nowicki-Strickland Locus of Control Scale were also administered. Performance under conditions 5 and 6 was greater than performance under conditions 3 and 4, which in turn exceeded performance under conditions 1 or 2. Criterion-related validity was significantly enhanced under all - elaborative conditions. MFFT latency, MFFT errors, and locus of control scores were differentially related to performance in the different testing conditions. The overall prediction equation explained CPM and Piagetian test performance under conditions 2, 4, and 5. All analyses indicated consistency of findings across both dependent measures. (Author/MH)