Cadavid, S. & Beato, M. S.
Departamento de Psicología Básica, Psicobiología y Metodología. Universidad de Salamanca. Salamanca, España.
A normative study was conducted using the Deese-Roediger-McDermott paradigm (DRM) to obtain false recognition for twenty-four 6-word lists in Spanish, designed with an original methodology. Lists included words (e.g., regiment, legion, battalion, officer, march, navy) simultaneously associated with 3 non-presented items, called critical words (e.g., ARMY, MILITARY, SOLDIER). Since previous research has shown that Backward Association Strength (BAS) seems to be closely related to the production of false memories, the lists were drawn up based on the BAS values between the words on the list and the critical words. In the present study, the associative strength of the lists were considered as the sum of BAS values for the three critical words. In particular, BAS values ranged between 0.561 and 1.595. Results showed, first, that all lists produced false recognition. And second, the 24 lists presented a high variability in producing the phenomenon. Specifically, the high false recognition rate found in some lists is an aspect of special interest for DRM experiments that record brain electrical activity. This type of lists will enable to raise the Signal-to-Noise ratio (S/N) in false recognition event-related potential (ERP) studies as they increase the number of critical trials per list, and will be especially useful for the design of future research. The results are discussed in terms of the activation/monitoring framework, which derives from spreading activation models. According to this framework, on the one hand, presentation of the word lists may arouse an implicit associative response of critical words. Thus, spreading activation through associative-semantic networks may be responsible, in part, for the false memory effect. On the other hand, monitoring processes allow us to determine whether each word that comes to mind was actually studied. When monitoring effectiveness is poor and a source-monitoring error occurs, there is a greater probability of false memories increasing.