Some groups of preliminary year indigenous students at the University of Papua and New Guinea were Ss for an investigation of concept selection strategies among students from a non-Western culture. Some of the methods of Bruner, Goodnow, and Austin (1) were replicated. It was found that in forming conjunctive categories the students were consistent in maintaining a definite strategy, that more students adopted a scanning strategy, and that focusers were the most successful. Mixed strategists attained no success. The abstraction of disjunctive concepts provided more difficulties, as it did with their Western counterparts. The Ss, especially focusers, did not maintain their strategies. Focusers and scanners on the conjunctive problems did not interchange roles on the disjunctive problems, but some of each adopted mixed strategies with the latter problems. Mixed strategists and focusers were more successful than scanners, though focusers either solved all or no problems. The Ss generally found the use of negative examples difficult. The investigation is the first part of a series of studies. © 1971 Taylor & Francis Group, LLC.
For accurate and reliable automatic control the organisation of computer systems in aircraft and applications in other multi-task computer systems is more likely to be based on a number of "cooperative" computers in an integrated manner , than a main central computer. Such a cooperative design should provide for the normal work load of each computer plus checking the data processing of another computer and have the ability to continue operation in case of a malfunction in one processor. To ensure that these requirements are met in a data driven operating system the basic system design parameters need careful assessment for optimised system performance especially when deadlines must be met in some control functions and an order of priorities preserved. The present paper reports progress on a computer aided system design of such linked cooperative computers under the above conditions to determine basic system parameters such as the desirable character rates for the several central processors (C.P's) and optimum cycle length of C.P. to C.P. transfers. The methods also overcomes objections  to some systems design procedures by allowing the results of sampled data theory to be made use of in the systems design by incorporating the sampling rate-although calculated separately  in one of the prime design parameters of the system. © 1971.