Neural Mechanisms of Familiarity

Despite its crucial role in our everyday lives, compared to most other forms of memory, familiarity memory has, over the years, received relatively little targeted attention in the human memory literature. We believe this is because familiarity has predominantly been studied as a comparison condition for recollection, where much more attention has been targeted. The behavioural and neural signatures of familiarity have often been derived from tasks, and under conditions, not optimal for familiarity to be observed or measured in its richest form, or its fullest capacity. We argue here that tests designed specifically to trigger and measure familiarity, reveal a much stronger and more confident form of memory than often assumed. Findings from such methods provide strong confirmation that the hippocampus does not support familiarity memory. Instead, the familiarity memory network includes the medial temporal lobe cortices (MTLc) and amygdala, the thalamus, and prefrontal and parietal cortex. We propose that the material specific MTLc and amygdala regions are responsible for generating familiarity signals, which project via the ventral lateral, and ventral posterior thalamic nuclei, to the mediodorsal thalamic nucleus. This intra-thalamic communication is modulated by familiarity strength. The mediodorsal thalamic nucleus then communicates its familiarity signal to the prefrontal cortex where relative familiarity is computed from independent familiarity and novelty signals. These signals converge in integration regions in both the prefrontal and parietal cortex allowing memory judgements and familiarity decisions to be made. We conclude by highlighting and briefly discussing several key outstanding questions.