The relative effectiveness of positive and negative information feedback in a concept attainment task

simple statistical test was made of the reliability 01' each 01' the two main effccts by applying a sign test (Sicgel. 1956) to the Ilumber of Ss who did and did not show the initial proactivc declillc, and the relative number of Ss who showed the proactive release (Lc., Trial 7 higher than the average of Trials 6 and 8), In both cases, the sign test yielded probabilities lower than ,05 associated with relevant null hypotheses. DISCUSSION The initial decline in performance over the course of the first six problems of each day is interpreted as the result of the development of proactive interference in short·term memory. The effect is short term in the sense that the in terpolated interval of a day between sessions restores performance to its original level. The retentional nature of the finding is suggested in part by the decremental direction of the change. Other (nonretentional) interproblem transfer effects that migh be expected here-learning set or intradimensional shift-are incremental in effecL Further evidence of the retentional nature of this proactive effect comes from a study by Knigh t (J 968), who showed the temporal control of proaction, Very short intertrial intervals prevent proaction from developing. The conclusion is that the proactive effect is on retention rather than learning. The increment on Trial 7 is interpreted as arelease from proactive interference. To make a formal derivation of this effect, a short-term memory model is needed, such as that of Atkinson & Shiffrin (I968), with the added assumption of separate buffer 100 DIMUSIO. I S',? /\ \'- 80 >OE .. . E~~ 70 u • L 0 50 I - --" 3 4 -& • -la 'IOllEMS Fig. 1. Mean percentage correct performance on test trials of successive problems. Proaction develops over the first six problems. is re1eased on Problem 7. and returns on the last three problems. Psychon. Sei .• 1970. Vol. 19 (2) REFERENCES ATKINSON, R, c., & SHIFFRIN, R. M. Human memory: A proposed system and its control processes. In K. W. Spence and J. T. Spence (Eds.). The psychology of learning and /I/II/irQ/ion. Vol. 2. New York: Academic Prc". I96X. Pp. H9-195. BROADBfNT, D. B. Perception and commllilicatioll. Oxford: Pergamon Press, 1958. HOUSE, B. J" & ZEAMAN, D. Miniature experiments in the discrimination learning of retardates. In L. P. Lipsitt and C. C. Spiker tEds.), Adl'ances in child del'elopment and behal'ior, Vol. I. New York: Academic Press, 1963. Pp. 313-374. ~~I(;tl\. \1. S. Thc cffccts 01" inter·trial interval uUfJtion Oll :-.hort-tcrm rl'tl'ntion 01' a t\\'ol'IlOkc \ i ... ual disnimination task by retarded L"ilildfl'n. l' npub lished dortoral dissertation, l'ni\l'I",it\ 01' Connectic"ut. 1968. LOESS. H: Proactive inhibition in short-term memory, Journal of Verbal Learning & Verbal Behavior, 1964, 3, 362-368. SCOTT, K. G. Some parameters of short-term reeall. Unpublished PhD dissertation. University of Connecticut, 1966. SCOTT. K. G .. & SCOTT, M. S. Research and theory in short-term memory. In N. R. Ellis (Ed.), International review of research in mental retardation. Vol. 3, New York: Academic Press, 1968. Pp. 135-157. SHEPP, B. E., & TURRISI, F. D. Learning and transfer of mediating responses in discrimination learning. In N, R, Ellis (Ed.), International review of research in mental retardation, Vol. 2. New York: Academic Press, 1967. Pp. 82-120. SIEGEL, S. Nonparametric statistics for the behOl'ioral sciences. New York: McGraw-Hill, 1956. TURVEY, M, T, Analysis of augmented recal! in short-term memory following a shift in connotation. British Journal of Psychology, 1968,59,131-137. WICKENS, D. D., BORN, D. G., & ALLEN, C. K. Proactive inhibition and item simüarity in short-term memory. Journal of Verbal Learning & Verbal Behavior, 1963, 2,440-445. WOLFF,1. L. Concept·shift and discrimination·reversal learning in humans. Psychological Bulletin, 1967,68,369-408. ZEAMAN, D., & HOUSE, B. J. The role of attention in retardate discrimination learning. In N. R. Ellis (Ed.), Handbook 01 mental deji"cimcl'. New York: McGraw-HiII 1963 Pp. 159:223. ' . NOTE 1. This research was carried out at the Mansfield State Training School, Mansfield Depot, Connecticut. The investigators wish to express appreciation for the active cooperation of Franeis P. Kelley, Superintendent, and Louis Boly, Director of Training. This study was supported by Research Grants M 1099 and K6-MH-HD-20,325 from the U.S. Public Health Service. DIM.I ;.., 0 : o DIM.2 storages for the dimensions of color and form. As the shorHerm storage for one dimension fil1s up and suffers interference, the other remains relatively empty and free of in terference to give rise to the "release" phenomenon. For a more complete account of the performance of retardates on discrimination problems of the type studied here, there is needed a composite of attention and retention models both of wh ich use the same dimensions of stimuli. The relative effectiveness of positive and negative information feedback in a concept attainment task LlNDA S. SIEGEL1 and JERROLD L. DO WN E Y, University of Missouri, Columbia, Mo. 65201 The relative effectiveness of positive and negative verbal feedback was investigated in an attempt to ascertain whether the differential effect was due to reinforcement value of the feedback or to some difference in information gained from the feedback, College Ss solved a concept attainment task under one of two conditions: ( 1) right-nothing (R-N), feedback for correct responses on(v, or (2) nothing-wrong (N·W), feedback for incorrect responses only, The expectation that, when the frequency of "right" and "wrong" feedback was controlled, there would be no difference between conditions was not supported, with N-W Ss performing significantly better than R-N Ss. Investigations regarding the relative effectiveness of verbal feedback combinations in concept attainment have led to the suggestion (e,g., Buss & Buss, lOS 1956) that "wrong" is a stronger negative reinforcer than "right" is a positive reinforcer. However, in these studies, feedback for incorrect answers occurred more frequently than feedback for correct answers, which led Bourne (1966) to suggest that, rather than ascribing the result to a greater reinforcing value for wrong relative to right, the obtained difference might have been due to the more frequent occurrence of wrong and thus a greater amount of informative feedback presented to S in this condition. The present study represents an attempt to contrast the relative effectiveness of "wrong" and "right" as reinforcers with the information value of each controlled. Specifically, the hypo thesis tested was that when the frequency of "right" or "wrong" was equated for all Ss, there would be no difference between right and wrong feedback conditions. METHOD The Ss were 40 students selected randomly from various psychology elasses at Lincoln University. Twenty Ss were assigned to each of two groups: N·W, in which feedback wa (...truncated)