Development of Recognition Memory: Process Dissociation of Recollection and Familiarity in Children

Abstract

There is an extensive debate in the adult literature on whether recognition memory can better be explained by a single- or a dual-process account. Single-process accounts assume that a single memory strength signal underlies recognition. Dual-process accounts propose two independent processes, namely recollection (slow and associated with contextual details) and familiarity (fast and automatic). The aim of this dissertation was to advance this debate using a cognitive developmental approach. By investigating age-related changes of recognition memory across childhood as a function of theoretically motivated experimental manipulations, predictions drawn from single- and dual-process models of recognition memory were tested. We adapted the Process Dissociation Paradigm (PDP; Jacoby, 1991) to disentangle processes underlying recognition memory in 5-, 7-, and 11-year-olds and adults using a Dual-Process Signal Detection cognitive modelling approach (DPSD; Yonelinas, 1996). Experiments 1 – 6 demonstrated that 5-year-olds are able to recollect items based on perceptual details. Consistent with dual-process theory, across all age groups a response time limit decreased recollection while leaving familiarity unaffected (Chapter 2). Converging evidence consistent with dissociations during childhood was found after repeated item presentation (Chapter 3). Finally, after a thorough empirical validation of our approach, the new paradigm was used to investigate the developmental perceptual to semantic shift (Chapter 4). These findings, using a double dissociation logic, have advanced the theoretical debate on the nature of recognition memory by showing that one process is insufficient to account for the developmental and experimental findings reported here. Recollection and familiarity follow different developmental trajectories and are affected by encoding and retrieval manipulations (i.e., repetition and time limits). This provides a challenge for existing theories of recognition memory.