c. STUMPF'S TOSPSYCHOLOGIE, i. an objective import. And the credibility of such a judgment considered with reference to the correlative objects is distinguished as " objective credibility ". In discussing the conditions of such credibility we have to distinguish between two classes of sense- judgment : (1) such as are responses to questions to which either alternative answer may according to circumstances be true or false; e.<j., TVhich of these two tones is the higher? (2) those corresponding to questions to which an affirmative answer is in all cases wrong and a negative one right ; e.g., Are these two tones perfectly like or equal in pitch ? Is this interval a perfectly pure one ? The conditions of the case exclude the possibility of this, both in the external processes and the related internal sensa- tions. The objective credibility of the first class of judgments is identical with their probability, and is determined by the number of right and wrong answers in a series of judgments. That of the second class is identical with the degree of approximation of the affirmations to the truth. In all cases the objective credi- bility of a judgment is conditioned by two universal factors : (a ) sensibility, or the degree in which our sensations correspond with the stimuli, either in respect of range (Umfangs-empfindUchkeit) or degree of difference (Unf- wtpJutdKekJceit} ', (b) subjective credibility, that is the credibility of a judgment in respect of the correct apprehension of sensations as such. Dr. Sturnpf insists on this last source of error even in cases where attention is complete. " Where lies the guarantee that through gradual increase of attention all and every difference of the sensations must become recognisable by us? " In respect of sensibility to difference this is, he thinks, demonstrable. Thus, in the case of sensations of pitch, by making the differences in the number of vibrations small enough, we obtain a series of sensations which can no longer be discriminated by the most observant and practised observer. And if the limits of observation were also those of sensibility, it would follow that all members of the series were alike, that is to say, there would be only one sensation of tone. It follows that there are differences among sensations smaller than the least perceptible. Not only are there unnoted differences of sensation, there are unnoticed and altogether unnoticeable sensations, e.g., some of Fechner's " negative sensa- tions " ; elements in a composite clang-sensation which cannot be detected. Subjective credibility is affected not only by this circumstance of the limits of observation at its best but also by other circumstances, as the degree of attention, the accuracy o the reproductive process in comparing successive sensations. It is in these two factors mainly that the effect of practice shows itself. All errors in sense-judgments must according to this view have their source in one of two regions. " Either we see or hear falsely, or we interpret falsely what is seen or heard." The general errors in judgments of Class I. are due to defective sub-
