False memory in a list learning paradigm : a maturational test of a putative relationship to frontal lobe function

The principal aim of this thesis was to test the conjecture of Schacter et al. (1995) that the false memory effect in the Deese-Roediger-McDermott (DRM) paradigm is generated by errors in source monitoring and is mediated by frontal lobe circuits. In support of their conjecture, Schacter et al. cited evidence from elderly populations with presumed frontal dysfunction, however this thesis studied children over an age range over which the frontal lobes are believed to be maturing. This thesis represents the first attempt to specifically test Schacter's hypothesis based on the developmental trajectory of 'frontal' functioning. Moreover, the use of a developmental trajectory in such a manner is relatively novel, and a key issue within this thesis is how to operationalise 'frontal' functioning in the sense intended by Schacter et al. Extrapolating from Schacter's conjecture, it was predicted that the strength of the false memory effect is directly proportional to the degree of frontal maturation, as estimated by both chronological age and by performance on a range of neuropsychological tests of 'frontal' executive functioning. False memory and executive functioning were compared in children aged 8 to 12 years and in adults using a modified DRM paradigm with a source monitoring extension after Payne et al. (1996). The modified DRM elicited false memory effects comparable to those reported in adolescents by Newstead and Newstead (1998) and in adults by Roediger and McDermott (1995) and by Payne et al. (1996). Three experiments that demonstrate a strong developmental trajectory for false memory effect are reported, but the pattern of results is not consistent with all of the premises of the conjecture. Although false recognition rates increased significantly with age, source monitoring accuracy did not change across the age groups. Consistent with the principal assertion of the conjecture by Schacter et al., age and a general 'frontal' factor were found to be significantly related to the size of the false memory effect. In addition, veridical memory performance was always a predictor of false memory performance. The IAR (Underwood, 1965) theory can best account for the significant covariation between veridical and false memory development, as well as being able to account for the dissociation between false memory and source monitoring