Four studies conducted over a 2-year period systematically evaluated the effectiveness of interactive instructional practices in the vocabulary learning and reading comprehension of learning disabled students. The design of the four studies was similar in that comparisons were made among three different interactive teaching strategies (semantic mapping, semantic feature analysis, and semantic/syntactic feature analysis) and a teaching strategy relying on direct, definitional instruction. During the first year, subjects consisted of 42 bilingual, learning disabled students in upper elementary grades studying social studies and 61 learning disabled junior high and middle school students studying science. During the second year, subjects consisted of 47 upper elementary and 53 junior high school students. All subjects were selected randomly from a metropolitan southwestern school district. Although specific instructional materials and students varied across the 2 years of the study, the major distinguishing characteristic between the two studies was that researchers instructed the students the first year while teachers resumed responsibility for instruction the second year. Upon completion of instructional intervention (and again 4 weeks later), students completed a multiple choice test and a written recall activity. Results indicated: (1) evidence to support the short and long term effectiveness of the interactive strategies; (2) transfer of this understanding to a written recall task was less encouraging; and (3) teachers were as effective as researchers in providing instruction. (Four tables of data are included; 24 references are attached.) (RS)