Computational models can distinguish the contribution from different mechanisms to familiarity recognition

Abstract

Familiarity is the strange feeling of knowing that something has already been seen in our past. Over the past decades, several attempts have been made to model familiarity using artificial neural networks. Recently, two learning algorithms successfully reproduced the functioning of the perirhinal cortex, a key structure involved during familiarity: Hebbian and anti‐Hebbian learning. However, performance of these learning rules is very different from one to another thus raising the question of their complementarity. In this work, we designed two distinct computational models that combined Deep Learning and a Hebbian learning rule to reproduce familiarity on natural images, the Hebbian model and the anti‐Hebbian model, respectively. We compared the performance of both models during different simulations to highlight the inner functioning of both learning rules. We showed that the anti‐Hebbian model fits human behavioral data whereas the Hebbian model fails to fit the data under large training set sizes. Besides, we observed that only our Hebbian model is highly sensitive to homogeneity between images. Taken together, we interpreted these results considering the distinction between absolute and relative familiarity. With our framework, we proposed a novel way to distinguish the contribution of these familiarity mechanisms to the overall feeling of familiarity. By viewing them as complementary, our two models allow us to make new testable predictions that could be of interest to shed light on the familiarity phenomenon.

Predictive coding model can detect novelty on different levels of representation hierarchy

“…This model achieves high capacity even when patterns are correlated (Androulidakis et al, 2008). Recently, Kazanovich and Borisyuk (2021) and Read et al (2024) have extended the anti-Hebbian model and bridged the gap between testing ND on binary patterns (i.e., each pixel value can be either 0 or 1) and natural images. In their experiments, the input to their anti-Hebbian ND model is not the image itself (as is the case in all of our experiments), but rather the features processed and detected by a deep convolutional network.…”

“…One of these models, the Anti-Hebbian model (Bogacz and Brown, 2003a), has been shown to replicate the capacity seen in human recognition memory, when presented with input patterns with a correlation structure likely in neurons representing visual stimuli (Androulidakis et al, 2008;Kazanovich and Borisyuk, 2021;Read et al, 2024).…”