Natural language and mixed modality task presentations in the human-computer interaction using programmable logic controllers

This dissertation focuses on the human-computer interaction elements of task presentation and task structure in the programming of Programmable Logic Controllers (PLCs). Task presentation methods that were explored included a natural language (GE Fanuc's State Logic), and a mixed modality language (GE Fanuc's Logicmaster Relay Ladder Logic). Task Structure elements utilized in this study were linear procedural tasks and conditional branching tasks.;A two-level factorial analysis of covariance determined that task presentation was a significant contributor to programming development times. Task structure and the interaction between task presentation and task structure did not yield significant p-values at the .05 alpha level to allow the rejection of their associated null hypotheses. The covariates utilized in the study included gender, age, status, computer experience, and English. None of the covariates appeared to prove significant in influencing program development times.;The population utilized in the quasi-experiment consisted of 26 general university students from Iowa State University with no prior PLC programming experience. The subjects underwent two hours of videotape training on how to program a single state or rung of code prior to the one hour testing portion of the study. The instruction consisted of basic education of a PLC, input and output functions and how they relate to associated hardware, and software navigation.;Significant contributions of this research include the distinction of a natural language from the alphanumeric categorization. Previous research documents the alphanumeric categorization as one that does not distinguish a difference between a natural language and a syntactical language. This study indicates that the use of a natural language is superior to the use of a mixed modality language in programming development times for PLCs. Previous studies had documented that mixed modality platforms were superior to alphanumeric (syntactical elements only) or pictorial languages alone.;Industry should strongly consider the implementation of a natural language into its process control. In addition, it is strongly recommended that a natural language be considered its own classification instead of lumping into the alphanumeric category. This study warrants this based on the natural language's performance superiority.