In the present study, 81 Norwegian students were taught the meaning of words by the Word Generation (WG) method and 51 Norwegian students were taught by an approach inspired by the Thinking Schools (TS) concept. Two sets of words were used: a set of words to be trained and a set of non-trained control words. The two teaching methods yielded no significant differences for the trained words, but the WG participants performed better than the TS participants on the non-trained words. An analysis of the dependencies between the trained words and non-trained words suggested that the participants of the WG program, in contrast to the TS students, profited from a transfer effect that resulted in augmented scores on the non-trained words.