Page:Mind (Old Series) Volume 11.djvu/553

 552 J. JASTROW a feeling of certainty ; 0, no decided inclination for one answer over its opposite ; and between the two 1 and 2 naturally found their places. 1 In the following Table, the errors in a set of 10 judgments and the average confidence are recorded for each sense and for each relative difference distinguishing the small lengths from the large ones. Experiments were made on four individuals and include about 1000 judgments. Sense. Absolute Length : 100 mm. Absolute Length : 25 mm. Difference of T^O- Difference of ! ,',,,. Difference of -sV- Difference of 7 u,,. No. of errors in 10. Av. con- fidence. No. of errors in 10. Av. con- iitlence. No. of errors in 10. Av. con- fidence. No. of errors in 10. Av. con- fidence. Eve 75 1.3 3-2 1-6 58 19 2-6 3-5 60 27 7 1-2 3-8 2-1 65 17 1-8 3-3 1-1 2 Hand Arm The order of differential sensibility of the three senses is Sight, Span, Motion. It is to be noted how regularly the confidence rises as the number of errors decreases. This indicates a distinct feeling of the superior sensi- bibility of Sight over Span, 2 of Span over Motion, as well as a distinction between differences of J^- and of y^. As the errors for lengths of 100 mm. differ little from those for lengths of 25 mm., Weber's law may be applicable to these kinds of sensibility. Some experiments were made to test the accuracy of the memory for Sight- and Span-impressions. At an early stage of the experiments, when sight was reproduced by span aud via- n>,-<l, only two lengths were used, one 20 mm., the other 100 mm. long. The blocks were repeatedly reproduced by the eye, and the lines by the thumb and forefinger. Before and after each day's sitting, the subject was required to mark off from memory lines of tlu; two lengths and to guide his thumb and forefinger to the accus- tomed distances on the Span-triangle. In both cases the memory 3 is extremely accurate and is almost as faultless after 1 This method v;is developed in connexion vitli experiments made by Mr. C. S. Peirce and the writer on "Small Differences of Sensation" (Memoirs of the National Arntlt /////, vol. iii.), to which the reader is referred for details. 2 In jud^in^ differences with the thnml) and forefinger, each hand was tested separately (the mean of the two results l.eiii^ ^iven in the Table), with the result that the left hand made fewer error.-. Those experimented upon wen- jiidit-handed. The ri^ht arm alone was tested. 3 Reproduction by memory is evidently analogous to expressing by inches ; the accuracy of these memories favours conclusion (2) on p. J49.