This study was conducted to determine whether efficiency in concept attainment can be increased as a function of the interaction of the use of good negative instances and increased attention as defined in terms of risk conditions. Subjects were 36 ninth-grade students of average problem solving ability as determined by IQ. Six treatment groups were formed by combining three types of series of instances, made up of good positive and good and poor negative instances, with a risk condition and a non-risk condition. It was hypothesized that if attention was increased, the combination good positive (GP) good negative (GN) series would be more efficient than GP only series or GP and poor negative PN series. It was concluded that generally a series of GP instances only are more efficient for concept acquisition than a mixed series containing either GN or PN instances. The authors discussed several areas suggested by the present study where further investigation should prove to be productive. (BW/Author)