Changes in the frequencies with which items elicited claimed or stated associates were studied before and after the items had participated in mediation and in control paradigms. The interpolated mediation-control task yielded reliable evidence of mediated facilitation. More associates were claimed on the second than on the first association test but the increase did not appear to be influenced by the particular conditions of the interpolated task.Mediated facilitation has been shown to increase when the meaningfulness of all of the learning materials was increased (Peterson & Blattner, 1963) and when the meaningfulness of the common-mediating-term was increased (Horton, 1964). It is possible that the items themselves became more' 'meaningful' , in these experimental sessions. The present experiment was designed to assess changes in the frequency with which associates were claimed for items before and after these items had been presented in an interpolated task. The interpolated task presented the items in two mediation paradigms and in two control paradigms.The mediation paradigms were a chaining model, A-B, B-C, A-CDE. and a response equivalence model, B-C, B-A, A-CDE.
MethodThe basic design was to present a 96-trigram association test before (ATl) and after (AT2) the trigrams had been presented as part of an interpolated task. The association tests were prepared by typing 96 consonant trigrams (CCCs) of from 0 to 42% association value (Witmer, 1935) on memory drum tapes. Two random sequences of the 96 CCCs were prepared.The interpolated task consisted of 24 "sets." A "set" was defined as the consecutive presentation of five cards representing either a mediation or a control format for one of the two paradigms. The 24 sets assigned to an S consisted of a randomized ordering of six sets representing the mediation form of the chaining paradigm (Paradigm I2)j six sets representing the control form of Paradigm Ij six, representing the mediation formj and six, the control form of the response equivalence paradigm (Paradigm VIII 2 ). Four different interpolated tasks were established so that the trigrams could be used in all conditions over the entire group of Ss.To balance the use of the trigrams, sets representing the mediation form, Paradigm I, were constructed and then the trigrams of these sets were systematically re-positioned to correspond to the other conditions as noted below. Three trigrams with minimum letteroverlap were chosen for the A, B, and C positions of each of the 24 mediation sets. The remaining 24 trigrams were used as the X-trigrams ofthe control sets. Thus, for each S, 24 trigrams were presented on ATI Psychon. Sci., 1966, Vol. 5 (2)
MARGARET JEAN PETERSON
INDIANA UNIVERSITYand A T2 but were not included in the particular interpolated task assigned to S. These trigrams were designated the "neutral X" items.The mediation form for Paradigm IwasA-B,A, B-C, B, A-CDE. The immediate retention of an acquisition pair was tested by presenting the stimulus trigram alone, e.g., A was shown immediately aft...