--------+-----------+--------------+--------------+---------
-----+--------------+--------------+--------------+--------------+--------------+--------------+--------------+---+---+---+---+-------- | | | | 1.2.3.4.5 | | | | | | 1.2.3.4 | 2.3.4.5 | | | | | | | 5.6.7.8 | 2.3.4.5.6 | 6.7.8.9 | 5.6.7 | 1.2.3.4 | 4.5.6 | 2.3.4.5 | 1.2.3 | 5.6.7 | 6.7.8.9 | | | | | | +--------------+--------------+--------------+--
------------+--------------+--------------+--------------+--------------+--------------+--------------+---+---+---+---+-------- 11 | 1-10 | 6.7 | 3.5 | 8 | 6 | 3 | 5 | 3.4 | 3.2 | 7.7.2.6 | 8 | 5 | 5 | 5 | 5 | 1:  1.00 12 | 11-20 | 7 | 3.6.6.2.5 | 8 | 6 | 3 | 4.5 | 4 | 2 | 7.6 | 8 | 7 | 3 | 7 | 3 | 1:  0.43 ========+===========+==============+==============+=========
=====+==============+==============+==============+==============+==============+==============+==============+===+===+===+===+========

[Footnote *:  Aided by experimenter.]

Examination of table 5 indicates that some of the settings proved very easy for Sobke; others, extremely difficult.  Consequently, the number of methods which were tried and rejected for a given setting varies from two to five.  Setting 2 proved a fairly simple one, and after the inhibition of the tendency to choose the first box at the left, the only definite tendency to appear was that of choosing the first box at the right, and then the one next to it.  After one hundred and thirty trials, this method suddenly gave place to direct choice of the right box, and during the following twenty-eight series, no error occurred for this setting.  Setting 4, on the contrary, proved extremely difficult, and a variety of methods is more or less definitely indicated by the records.

It is needless to lengthen the description by analyzing the data for each setting, since the reader by carefully scanning the columns of data in table 5 may observe for himself the various tendencies and their mutual relations.

Sobke’s curve of learning (figure 19) in problem 2, is extremely irregular, as was that of Skirrl.  Similar irregularities appear in the daily ratios of right to wrong first choices presented in the last column of table 5.  Most of these irregularities were due, I have discovered, to unfavorable external conditions.  Thus, dark rainy days and disturbing noises outside the laboratory were obviously conditions of poor work.

On the day following the final and correct series for problem 2, a control series was given.  In this Sobke seemed greatly surprised by the new situations which presented themselves.  Repeatedly he exhibited impulses to enter the box which would have been the correct one in the regular series of settings.  He frequently inhibited such impulses and chose correctly, but at other times he reacted quickly and made mistakes.  It was evident from his behavior that he was not guided by anything like a definite idea of the relation of the right box to the other members of the group.