Episodic memory refers to the storage and retrieval of information about events in our past. According to dual process models, episodic memory is supported by familiarity which refers to the rapid and automatic sense of oldness about a previously encoded stimulus, and recollection which refers to the retrieval of contextual information, such as spatial, temporal or other contextual details that bring a specific item to mind. To be clear, familiarity is traditionally assumed to support recognition of item information, whereas recollection supports the recognition of associative information. Event Related Potential (ERP) studies provide support for dual process models, by demonstrating qualitatively distinct patterns of neural activity associated with familiarity (Mid-Frontal old/new effect) and recollection (Left-Parietal old/new effect). In the current thesis, ERPs were used to address two important questions regarding associative recognition – namely, the function of the neural signal supporting recollection and whether familiarity can contribute to the retrieval of novel associative information. The first series of experiments was aimed at addressing how recollection operates by employing a recently developed continuous source task designed to directly measure the accuracy of retrieval success. To date, the function of recollection has been fiercely debated, with some arguing that recollection reflects the operation of a continuous retrieval process, whereby test cues always elicit some information from memory. Alternatively, recollection may reflect the operation of a thresholded process that allows for retrieval failure, whereby test cues sometimes elicit no information from memory at all. In the current thesis, the Left Parietal effect was found to be sensitive to the precision of memory responses when recollection succeeded, but was entirely absent when recollection failed. The result clarifies the nature of the neural mechanism underlying successful retrieval whilst also providing novel evidence in support of threshold models of recollection. The second series of experiments addressed whether familiarity could contribute to the retrieval of novel associative information. Recent associative recognition studies have suggested that unitization (whereby multi-component stimuli are encoded as a single item rather than as a set of associated parts) can improve episodic memory by increasing the availability of familiarity during retrieval. To date, however, ERP studies have failed to provide any evidence of unitization for novel associations, whereas behavioural support for unitization is heavily reliant on model specific measures such as ROC analysis. Over three separate associative recognition studies employing unrelated word pairs, the magnitude of the Mid-Frontal old/new effect was found to be modulated by encoding instructions designed to manipulate the level of unitization. Importantly, the results also suggest that different encoding strategies designed to manipulate the level of unitization may be more successful than others. Finally, the results also revealed that differences in behavioural performance and modulation of the Mid-Frontal old/new effect between unitized and non-unitized instructions is greater for unrelated compared to related word pairs. In essence, the results suggest that unitization is better suited to learning completely novel associations as opposed to word pairs sharing a pre-existing conceptual relationship. Overall, the data presented in this thesis supports dual process accounts of episodic memory, suggesting that at a neural level of analysis, recollection is both thresholded and variable, whilst also supporting the assumption that familiarity can contribute to successful retrieval of novel associative information. The results have important implications for our current understanding of cognitive decline and the development of behavioural interventions aimed at alleviating associative deficits.

Dual-process models of recognition memory propose that recognition memory can be supported by either a general sense of familiarity or the recollection of the encoding context. One source of evidence supporting dual-process models comes from event-related potential (ERP) studies of recognition memory, which have identified distinct patterns of neural activity associated with familiarity and recollection (the mid frontal and left parietal old/new effects, respectively). In this thesis, dual-process accounts of recognition memory were investigated in a series of ERP studies using three categories of stimulus: previously unknown faces, famous faces, and names. For previously unknown faces, familiarity was associated with activity over posterior scalp electrodes while recollection was associated with topographically dissociable activity over anterior electrodes. These dissociable patterns of activity support dual-process models. However, the typical pattern of old/new effects was only observed for stimuli associated with pre-existing representations (i.e., names and famous faces), suggesting that the presence/absence of pre-existing representations may determine the particular retrieval processes that support recognition memory. Furthermore, recollection was associated with two different patterns of activity (anterior and left parietal effects), suggesting that recollection is not a homogenous process. Dual-process theories may represent an important starting point for investigating recognition memory, but neither familiarity nor recollection appear to be functionally homogenous processes when theorizing is constrained by the analysis of scalp recorded electrophysiological activity.

Avec l’avancée en âge, l’efficience de l’ensemble des fonctions cognitives, en particulier la mémoire épisodique, tend à diminuer. L’étude des effets Dm et old/new et des résultats de la méthode sLORETA a permis de caractériser sur un plan neurofonctionnel deux types de vieillissement non pathologique chez les seniors. D’une part, le vieillissement cognitif stable se traduit par des processus d’encodage identiques à ceux observés chez des sujets jeunes alors que les processus de récupération nécessitent un recrutement bilatéral des lobes temporaux médians, cette activation étant unilatérale chez les sujets jeunes. Ces résultats vont dans le sens du modèle HAROLD qui suggère une réduction de l’asymétrie hémisphérique avec l’âge. D’autre part, le vieillissement cognitif altéré proviendrait d’une altération des réseaux cérébraux dès la phase d’encodage. La baisse d'activité au niveau du cortex préfrontal gauche, siège des fonctions exécutives, suggèrerait une difficulté de mise en œuvre de stratégies d'encodage efficaces et aurait un impact direct sur l'activité observée au niveau du lobe temporal médian de façon bilatérale ce qui pourrait donc expliquer une difficulté de stockage de l'information. / With advancing age, the efficiency of all cognitive functions, especially episodic memory, tends to decrease. The study of Dm and old/new effects and of the results of the sLORETA method allowed to characterize, on a neurofunctionnal level, two types of non-pathological aging among seniors. On the one hand, stable cognitive aging results in encoding processes similar to those observed in young subjects while the retrieval processes require a bilateral recruitment of the medial temporal lobes, this activation being unilateral in young subjects. These results are in line with the HAROLD model, suggesting a reduction of hemispheric asymmetry with age. On the other hand, the altered cognitive aging might come from an alteration of brain networks at the early stage of the encoding phase. The decline of activity in the left prefrontal cortex, seat of executive functions, would suggest a difficult implementation of effective encoding strategies and have a direct impact on the activity observed in the medial temporal lobe bilaterally which could therefore explain information storage difficulties.