Projects

The ability to form new associations is the core function of the human memory system. At least when it comes to declarative memory contents, this is fundamentally related to intact hippocampal functioning. However, recent data show that there can be exceptions from the rule. These findings suggest that associative memory independent of the hippocampus is possible unless the to-be associated information is processed in a specific manner. For instance, this is the case when information has to be unitized, i.e. integrated into a coherent whole. Then, extra-hippocampal structures such as the perirhinal cortex can bind the to-be associated information into a rigid integrated configuration. Moreover, recently, patients with selective hippocampal lesions have been found to perform above-chance when tested on arbitrary label-object associations acquired by means of so-called “fast mapping”. Fast mapping paradigms differ from standard associative learning procedures in three important ways: Associative links are not given but are discovered actively by participants, novel associations are created within an existing context such as already known objects, and learning of associations is not deliberate. However, it is an absolutely unresolved issue which of the specifics of this paradigm are crucial to the effectiveness and the potential to bypass hippocampal-based processes. None of them has been systematically investigated so far and the neurocognitive mechanisms behind fast mapping remain completely unclear. The first goal of this grant proposal is to elucidate the neuronal basis and the determinants for the formation of novel object-label-associations by means of fast mapping. We expect the perirhinal cortex to play an important role as patients with lesions including this region could not acquire novel associations by means of fast mapping. The second aim is to shed light on the potential benefit of specific study contexts as it can be assumed that high feature overlap between context and the to-be-learned item encourages perirhinal involvement. A further important objective is the evaluation of the interaction of different study contexts with variable testing conditions as distinct test formats (e.g., forced-choice vs. yes/no-recognition) are supposed to be differentially optimal for retrieving memories represented in the perirhinal cortex. These goals will be achieved by a functional magnetic resonance imaging study with healthy participants and by performance analysis in patients with amnestic mild cognitive impairment, a population with a selectively reduced perirhinal volume in early stages. Having established the factors driving fast mapping, this knowledge will finally be used to set up learning conditions which are beneficial in attenuating the known associative memory deficit in healthy older adults.