TABLE III
| x = no judgments of that kind given. | % of mistakes in judgments of certainty (1). | % of mistakes in judgments of certainty (2). | % of mistakes in judgments of certainty (3). | % of mistakes in judgments of certainty (4). | |
| Subject (in giving shapes) | (a) | 7% | 10% | 0% | 100% |
| A (in giving colors) | (b) | 2 | 14 | 23 | 0 |
| (in giving letters and numerals) | (c) | 0 | 0 | x | x |
| Subject | (a) | 2 | 10 | 10 | 0 |
| B | (b) | 3 | 6 | 20 | 25 |
| (c) | 4 | 50 | 0 | x | |
| Subject | (a) | 1 | 8 | 10 | 0 |
| C | (b) | 4 | 14 | 14 | 0 |
| (c) | 1 | 0 | 16 | 0 | |
| Subject | (a) | 3 | 4 | 0 | 16 |
| D | (b) | 1 | 6 | 0 | 0 |
| (c) | 5 | 0 | 0 | 0 | |
| Subject | (a) | 2 | 10 | 25 | 50 |
| E | (b) | 1 | 33 | 40 | 0 |
| (c) | 0 | 0 | 0 | 0 | |
| Subject | (a) | 1 | 25 | 7 | 25 |
| F | (b) | 4 | 15 | 29 | 26 |
| (c) | 3 | 0 | 0 | x | |
| Subject | (a) | 3 | 6 | 5 | 0 |
| H | (b) | 6 | 2 | 15 | 15 |
| (c) | 4 | 0 | 0 | 20 |
TABLE IV
| General % of mistakes in judgments of certainty (1). | % of mistakes in judgments of certainty (1) about attended-to groups. | General % of mistakes in judgments not of certainty (1). | % of mistakes in judgments not of certainty (1) about attended-to groups. | |
| Subject A | 4% | 1% | 17% | 27% |
| Subject B | 3 | 2 | 10 | 7 |
| Subject C | 2 | 3 | 9 | 24 |
| Subject D | 3 | 4 | 4 | 0 |
| Subject E | 1 | 2 | 22 | 34 |
| Subject F | 3 | 1 | 21 | 23 |
| Subject H | 4 | 6 | 6 | 10 |
The results given in Tables III and IV were compiled from the same records as those of the two Tables just discussed. They give the relation of error to certainty and to attention, as that relation was developed in this experiment. No experiments were conducted with these relations of error primarily in view, but the results developed in connection with the problem of the effect of attention upon certainty in one's judgments.
Both Tables show again marked individual variation. They suggest to me, in the first place, a further line of investigation in the same field and for the same purpose as those investigations which L. William Stern outlines in an article[101] entitled Aussagestudium. This further line is the testing subjects to learn the probable relative correctness of the judgments made with different degrees of confidence. Although a comparison of the first and third columns in Table IV makes it clear that the proportion of mistakes for the highest grade of confidence is lower than for the other grades taken together, there is a very marked difference among the subjects to be noticed. The difference in the two percentages is, for instance, very slight in the cases of D and H, and very great in the case of E. It is interesting to notice with regard to E that while he has the lowest percentage of mistakes for certainty (1), he has the highest percentage for the group of certainties (2), (3), and (4). In the more detailed percentages given in Table III we see further that in certain fields and sometimes in all fields (as with subject C) judgments made with the lowest grade of confidence were invariably correct. Such Tables might be of help in a case where the evidence of eye-witnesses conflicted. We might perhaps learn that witness N made a large proportion of mistakes where he was absolutely certain, whereas witness M was seldom wrong in judgments in which he had a low degree of confidence. Even when the probity of both was unquestioned, we should not then assume that N was more probably right because he had so much more confidence in his judgments than M had in his. A much longer and more comprehensive set of experiments would be necessary before we could feel that we had at hand a table from which to work in this way.
The question of the effect of voluntary attention upon error, for answering which Table IV was compiled, brings out again the marked individual variation among these seven subjects which has shown itself in practically all parts of the experiment. Some effect seems to have been produced always, but this was sometimes to give a larger percentage of mistakes in the attended-to groups and sometimes a smaller. With A, B, and F the percentage of mistakes in certainty (1) was lower for the groups attended to than for the total number of judgments of that order. Only with subject B, however, is this true of the group of lower grades of certainties also. On the other hand, with subjects C, D, E, and H the percentage is greater for certainty (1) in the groups attended to than for certainty (1) in the collection of all the judgments of certainty (1) taken together. Here, too, in the case of subject D, the results with regard to the lower grades of certainty reverse those for certainty (1). Thus all four possibilities as to the kind of influence of voluntary attention upon certainty appear. We cannot say that the place of voluntary attention will tend to affect the percentage of error in any given way. We can only say that apparently it made some difference with each subject. It might be found by further experimenting that the character of this difference is associated with some other characteristic of either attention or the feeling of certainty, as, for instance, with the ease with which attention is held to the chosen field or with the type of the subject's certainty.
Like all experiments, these open up further questions quite as much as they answer those toward which they are aimed. To repeat something of what has already been said, I feel that what it has established is (1) that introspection develops distinct grades of certainty in the case of every individual, (2) that the particular characteristics of the feeling of certainty vary markedly among individuals; (3) that the feelings of certainty associated with the different senses are not, as feelings of certainty, to be distinguished from each other; (4) that the judgments of the highest degree of certainty which are made about the constitution of any visual field just seen will not be confined to the group in that field toward which the attention is directed; and (5) that such fixing of the attention will, nevertheless, materially alter the subject-matter of such judgments of greatest certainty. The rather vague statement that the percentage of error is not surely less with the judgments of a group because attention is fixed on that group may perhaps be added as a sixth conclusion. The most interesting and promising of the problems which the experiments seem to me to raise are: (1) the problem, are there such definite types of the feeling of certainty that people may be classified according to their types, and, if so, what are the types and what their relation to other psychological characteristics of the individual? (2) the problem, what will be the result of careful and trained introspection as to the relation of so-called logical and psychological certainty and in what fields do these appear for different individuals? (3) the problem, how can a test for grading the probable percentage of error in the judgments of different grades of certainty made by any one person be constructed? and (4) the problem, how are such facts as those given in Table IV to be connected with the effort required for attention, the type of certainty of each subject, etc.? Other problems could, of course, be suggested, but these, I feel, mark the steps that naturally follow the experiments described here.
