Effects of the mnemotechnique of loci in the memorization of concrete words

The present paper explores the effects of the use of locimnemonics by expert people when series of three items (triplets) must be memorized. In an initial experiment, 14 subjects were asked to memorize 20 triplets of concrete, high-imagery, high-frequency words, connecting them to 20 well-known mnemonicloci and following ‘progressive elaboration’ instructions. Their performance on serial recall was compared with that of a group of subjects who used ‘progressive elaboration’ instructions but did not know locimnemonics. Both groups revealed a position effect in the triplet (the item in first position was remembered better than the second or third in the triplet). However, the overall performance of the locimnemonic group was far better, especially in the order of recall both within the triplets and between them. In the second experiment following the same procedure, the effects of different modalities of a 7-day delayed test were analyzed in 8 groups of 7 subjects. The mnemonic group performed better both on free recall and recognition, but not on cued recall. When a word strongly associated to the first word of the triplet was given as a cue, both locimnemonic and control subjects gave a low, not significantly different, performance. It is argued that the latter cue is in conflict both with the cue-locus and with the specific encoding of the triplet given also by controls. Further, in all groups, retrieval seems to proceed from the first to the second to the third items. The most general result of the two experiments is that progressive elaboration instructions make locimnemonics not only powerful with series of materials, but also capable of maintaining the order of information within and between the loci. Finally, the effects of locimnemonics appear stronger in subjects with good performance on a memory localization test requiring them to remember where pictures appeared on the pages of a book.