16 16<2> 12<1> 3 10% 2/7 15 10 14<3> 3 12% 2/5 4 12 28<6> 6 15% 3/4 19 15<2> 8<2> 4 17% 4/6 3 17<2> 20<5> 7 19% 0/6 7 11<2> 17<4> 6 21% 2/5 6 17<1> 12<6> 7 24% 1/7 13 8 21<7> 7 24% 4/4 8 16 28<12> 12 27% 5/7 9 12 32<12> 12 27% 6/7 14 7 21<8> 8 28% 4/7 2 10 12<7> 7 32% 1/5 20 6 9<8> 8 53% 2/5 ---------------------------------------------------------------- Only 13 of our 19 mentally normal cases were found to have had the ``Aussage'' Test done so uniformly that results could be fairly compared, as in the above table. The reader will find it easy to refer back to the case for noting other correlations with behavior. In the first double column the items which were given in free recital come first, and in the second part the number of positive responses to questions by the examiner. The coefficients attached to these represent the number of egregious errors or entirely fictitious items given. It should be clearly understood that slight deviations from facts, for instance in color, are not counted as errors for our present purposes. In a later study on this whole topic of the psychology of testimony there will be much more complete itemizing. The errors in which we are particularly interested can perhaps best be called pure inventions. In the next double column is given, first, the total number of incorrect items and, then, the percentage of these to the total number of items reported. In the last column suggestibility is dealt with. We have been accustomed to offer 7 suggestions, asking the individual whether such and such things which might well be in a butcher shop really appeared in the picture. For several reasons not all of the 7 suggestions were asked in every case, therefore the result is best viewed as a statement in fractions-- the numerator being the number of suggestions accepted and the denominator the number of suggestions offered.