STUDYING CONTROLLED WORD ASSOCIATIONS 419 



of the ten sessions, and preserving an exact balance in the order 

 of presentation of the various groups. In both parts of the 

 experiment the associations were recorded, also the reaction tunes, 

 which were measured with the Johns Hopkins chronoscope. 

 Averages and mean variations were calculated on the Burroughs 

 adding machine, using the Dunlap adding machine formula 

 for the mean variations. From this data it was purposed to 

 observe whether there is any definite tune relation between the 

 adjective-noun and the noun-adjective associations, or between 

 the verb-object and verb-subject association and whether the 

 reaction time has any fixed relation to the number of syllables 

 in the stimulus word, the position of the accent or the logical 

 category to which the stimulus word belongs. Some data was 

 secured on double associations and associations to repeated 

 stimulus words, and the emotional adaptation of the subject to 

 the experiment is discussed. 

 The following conclusions were drawn from the results : 



1. The normal order for adjectives and nouns in the English 

 language gives on the average a shorter association time than 

 the inverse order, when the number of syllables in the compared 

 groups is the same. That is, the reaction time for the adjective- 

 noun association is shorter than for the the noun-adjective as- 

 sociation. This holds when the groups are made up of carefully 

 selected words, or when the stimulus words are all difficult, i.e., 

 the words rejected from the revised lists, or when the stimulus 

 words are still unevaluated and therefore include both difficult 

 and easy words. 



2. The reaction time for both the adjective-noun and noun- 

 adjective associations increases on the average directly with 

 the number of syllables in the stimulus word, and as in (1) holds 

 for carefully selected stimulus words, or difficult words, or mixed 

 groups of words containing both selected and difficult words. 



3. The position of the accent in the stimulus word has no 

 systematic effect on the reaction time for either the adjective- 

 noun or noun-adjective association. 



4. There is no interpretable variation in reaction time on 

 the average according to the grouping of stimulus nouns into 





