176 -SECTIONAL ADDRESSES 



shown serially to individual subjects, each card being exposed for thirty 

 minutes. Immediately after the presentation the subjects were asked to 

 write a full report of the cards. Without warning they were asked a 

 month later to report all that they could then recall of the pictures ; a 

 third report was called for at the end of another month, and finally, in 

 some cases, a fourth report was written after the lapse of a period varying 

 from a year to nineteen months. The results of these experiments 

 showed that certain pictures had been uniformly well remembered and 

 others ill. The question arose how far this might be due to the position 

 of the cards in the series, rather than to the intrinsic character of the cards. 

 The reports also suggested problems about the influence of one recall 

 upon another. 



Further experiments have been made with the same material. Eleven 

 new subjects took part, and the range of their scores for immediate 

 recall show them to be a group comparable to that of the previous experi- 

 ment. Some of the hypotheses suggested a year ago receive further 

 support. 



Position in Series and Intrinsic Character. 



In the experiments of the present year the pictures which had been 

 worst remembered were put in the positions occupied by the pictures 

 yielding the best scores, and vice versa. To be first in the series would 

 seem undoubtedly to be advantageous. The best score now as number i 

 attaches to a card which only possessed a fair record previously as 

 number 4. There is also something to be gained by being last in a series. 

 As last picture old number 2 has now"a fair score. It previously had a 

 very low one. 



But position will not explain everything. Old number 5 which was 

 put in the position of good number 3 remains very low in score . Number 3 , 

 though assigned the position of old low-scoring number 2, still yields a 

 high score. Old number i is not so high now that it occupies the fourth 

 place, but it still obtains a good score. 



One may contrast the two cards that retain their former respective high 

 and low scores. I hazarded the suggestion last year that this was due to 

 their intrinsic character, and in particular to the spatial organisation. In 

 present number 2, foreground, middle distance and background make a 

 single whole — each contributes to one scene. In present number 3 there 

 are three scenes unrelated, whether viewed in terms of perceptual organi- 

 sation or in terms of meaning. 



This year's reports contain evidence of the same confusions as last 

 year's. In meaning there is a relation between ' Chopping Logs ' and 

 ' Felling Trees.' The right side of number 3, representing the latter 

 activity, is imported into number i, where log-chopping is in the centre 

 of the picture. The hut of number i and the hut of number 5 are 

 interchanged ; each has a feature of similar appearance — viz. a wooden 

 upright supporting the roof. The principles of Gestalt psychology as 

 well as the doctrine of meaning may be evoked to explain the data. 



The recall of the picture wherein ' pig killing ' is the central episode 

 may find its right explanation in the emotional value of the scene, and here 



