Trends in Cognitive Sciences
OpinionImaging recollection and familiarity in the medial temporal lobe: a three-component model
Introduction
Converging results from studies of individuals with amnesia, animal models and imaging research have shown that regions in the medial temporal lobe (MTL) have an essential role in supporting episodic memory (e.g. Refs 1, 2). Although there is broad support for the role of the MTL in memory, there is debate about the extent to which the hippocampus, parahippocampal cortex (PHc) and perirhinal cortex (PRc) support recollection and familiarity-based recognition. Recollection is the process of recognizing an item on the basis of the retrieval of specific contextual details, whereas familiarity is the process of recognizing an item on the basis of its perceived memory strength but without retrieval of any specific details about the study episode (Box 1).
Here we review the results of neuroimaging studies that have examined neural correlates of recollection and familiarity in the MTL, and consider their theoretical implications. We present a three-component model, originally proposed by Eichenbaum et al.[3], which we refer to as the ‘binding of item and context’ (BIC) model. The original model was based on an overview of results from behavioral, neuropsychological and neuroimaging studies of recognition memory in humans and animal models. In this paper, we further specify the BIC model by examining the relevant neuroimaging evidence in more detail, by using the model to generate specific predictions for future neuroimaging studies, and by relating the model to other models of MTL function.
Section snippets
fMRI studies of recollection and familiarity
Event-related functional magnetic resonance imaging (fMRI) studies have used several different measures to disentangle recollection and familiarity. These measures include experiments in which subjects study individual items and are then given recognition tests that require remember/know responses, source memory responses or recognition confidence responses (Box 1). In general, results from contrasts that differentiate between remember and know, between source correct and source incorrect, and
The BIC model of MTL function
Anatomical studies of monkeys and rodents provide a foundation for the BIC model [3]. In general, these studies suggest that the PRc receives detailed information about specific items that are to be remembered, whereas the PHc receives detailed information about the spatial context in which each item was encountered (Box 2). The information about ‘what’ and ‘where’ – two key attributes of episodic memories – converges in the hippocampus. Drawing on this framework, we suggest that the PRc and
Additional model predictions
The relatively simple assumptions described above lead to several predictions for more complex tasks, such as associative memory tasks. For example, tests of associative recognition in which pairs of items are studied and memory is then tested for the specific item pairings might lead to activations that are somewhat different from those seen in tests of single-item recognition. As in item recognition, recollection of inter-item associations should elicit hippocampal activation because
The nature of context in the PHc
A novel aspect of the BIC model is that it predicts that context representations, including both spatial and non-spatial information, are encoded by the PHc. Several studies have implicated the PHc in memory for spatial information 49, 50. These findings have led theorists to suggest that the PHc is important for encoding spatial context [51]. Several studies have demonstrated PHc involvement in a range of memory tasks, however, suggesting that it plays a wider role in memory formation. For
Comparison of BIC with other recent models of MTL organization
Although the BIC model is similar to other recent models of MTL organization, it differs in some important aspects. For example, Fernandez and Tendolkar [40] referred to the PRc as the ‘gatekeeper’ of declarative memory, arguing that its role is to direct encoding resources towards less familiar items. This same familiarity signal is then relevant at retrieval for making item memory judgments. This model predicts the decreasing activation that is seen with repetition of items in the PRc;
Concluding remarks
Imaging data have provided significant insights into the functional organization of the MTL and, more specifically, into the way in which different MTL regions contribute to recollection and familiarity. Whereas activity in the hippocampus and PHc is correlated with recollection, activity in the PRc is correlated with familiarity. The results are not consistent with many existing models, but they can be explained by the BIC model. The BIC model is the first model of MTL function to propose a
Acknowledgements
This work was supported by the National Institute of Mental Health (grants MH068721 and MH059352) and the National Institute of Neurological Disorders and Stroke (grant NS40813). We thank the reviewers for their insights.
References (59)
The hippocampus – what does it do?
Behav. Neural Biol.
(1992)- et al.
Encoding activity in anterior medial temporal lobe supports subsequent associative recognition
NeuroImage
(2004) Dissociable correlates of recollection and familiarity within the medial temporal lobes
Neuropsychologia
(2004)Episodic encoding is more than the sum of its parts: an fMRI investigation of multifeatural contextual encoding
Neuron
(2006)Prefrontal and hippocampal contributions to visual associative recognition: interactions between cognitive control and episodic retrieval
Brain Cogn.
(2004)Memory orientation and success: separable neurocognitive components underlying episodic recognition
Neuropsychologia
(2003)- et al.
Functional–anatomic correlates of remembering and knowing
NeuroImage
(2004) Content-specificity of the neural correlates of recollection
Neuropsychologia
(2005)Is organic amnesia caused by a selective deficit in remembering contextual information?
Cortex
(1985)- et al.
Interleaving brain systems for episodic and recognition memory
Trends Cogn. Sci.
(2006)