Children with autism spectrum disorder (ASD) who have limited spoken communication skills are commonly taught to mand by exchanging picture symbols. Competent manding via symbol exchange relies on conditional discrimination--that is, selecting the picture symbol that corresponds to the desired reinforcer. Picture Exchange Communication System (PECS) is a popular symbol exchange intervention for children with ASD that teaches conditional discrimination (i.e., phase IIIb) after teaching simple discrimination (i.e., phases I through IIIa). However, the initial simple discrimination phases of PECS may not offer meaningful communicative benefits and may delay acquisition of conditionally discriminated mands for some children. This dissertation research included a systematic literature review and experimental study to better understand this issue. The systematic review of single-case PECS research was conducted to determine how often and how quickly individuals with ASD and other developmental disabilities (DD) progress through PECS phases. A total of 2,001 published and unpublished records were screened, and 18 studies were coded for details about which PECS phases participants mastered and the number of trials (or sessions) to mastery. Results indicated most participants mastered the first two phases of PECS. However, approximately half of participants did not appear to master phase IIIb, and the number of trials to master phases I and II showed considerable variability. These findings suggest PECS may inconsistently or slowly produce conditionally discriminated mands. This is concerning because conditional discrimination is necessary for meaningful exchange-based communication. An adapted alternating treatments design was used to investigate whether an abbreviated symbol exchange (ASE) intervention resulted in quicker acquisition of conditionally discriminated mands when compared to PECS phases I through IIIb for children with ASD and other DD. ASE applied conditional discrimination procedures from the start of instruction, whereas PECS progressed from simple (i.e., phases I through IIIa) to conditional (i.e., phase IIIb) discrimination procedures. Participants were four young children (aged 3 to 5 years) with ASD or other DD who had limited or no functional speech. ASE, PECS, and no-treatment control conditions were rapidly alternated. Conditionally discriminated mands were recorded and analyzed as percentage of correct responses. Generalization probes were initiated for all participants after control condition responding improved for the first participant. Generalization probes were used to detect generalized identity matching (i.e., untaught conditionally discriminated mands). Results indicated ASE was more efficient than PECS for teaching conditionally discriminated mands to three of four participants. However, different patterns of responding were evident for different participants. Two participants rapidly acquired conditionally discriminated mands from ASE, and generalized identity matching appeared to emerge. One participant acquired conditionally discriminated mands from ASE, but results did not support generalized identity matching. Finally, one participant did not acquire conditionally discriminated mands from ASE or PECS. Findings from this study offer preliminary evidence that ASE can more rapidly produce conditionally discriminated mands than PECS for some children with ASD and other DD. Results also suggest generalized identity matching may emerge from instruction for some learners. Further research is needed to determine the variables that contribute to different treatment outcomes from ASE and PECS. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://bibliotheek.ehb.be:2222/en-US/products/dissertations/individuals.shtml.]