Novel stimuli evoke excess activity in the mouse primary visual cortex

Abstract

To explore how neural circuits represent novel versus familiar inputs, we presented mice with repeated sets of images with novel images sparsely substituted. Using two-photon calcium imaging to record from layer 2/3 neurons in the mouse primary visual cortex, we found that novel images evoked excess activity in the majority of neurons. This novelty response rapidly emerged, arising with a time constant of 2.6 ± 0.9 s. When a new image set was repeatedly presented, a majority of neurons had similarly elevated activity for the first few presentations, which decayed to steady state with a time constant of 1.4 ± 0.4 s. When we increased the number of images in the set, the novelty response’s amplitude decreased, defining a capacity to store ∼15 familiar images under our conditions. These results could be explained quantitatively using an adaptive subunit model in which presynaptic neurons have individual tuning and gain control. This result shows that local neural circuits can create different representations for novel versus familiar inputs using generic, widely available mechanisms.

Efficient Temporal Coding in the Early Visual System: Existing Evidence and Future Directions

“…Adaptation may therefore provide a mechanistic implementation of efficient coding for certain stimulus distributions, without the need for any form of long-term plasticity to encode temporal relationships (see Figure 3 ). This may be demonstrated by a recent paper showing that novel stimuli presented within repeatable sequences evoke excess activity, as predicted by efficient coding ( Homann et al, 2022 ). The proposed mechanism, however, was consistent with a straightforward adaptation model.…”

“…Many related studies have shown evidence for cortical novelty responses across brain regions, with wide variation in the effort to control for adaptation ( Kato et al, 2015 ; Makino and Komiyama, 2015 ; Garrett et al, 2020 ; Poort et al, 2021 ; Schulz et al, 2021 ; Homann et al, 2022 ; Montgomery et al, 2022 ). In most cases, consistent with efficient coding, novel stimuli evoke more spiking activity than familiar.…”

Fast Adaptation to Rule Switching using Neuronal Surprise

“…1 C where different rules correspond to transition matrices of different apartments. Inspired by experimental observations for passive learning in humans and animals 10,24,85,91,92 , we assume that the (potentially unconscious) goal of observers is to predict possible next observations, i.e., estimate transition probabilities T i, j that are as close as possible to the real probabilities . Our spiking neural network model (introduced in the next paragraph) implicitly encods expectations about possible next stimuli in the set of synaptic weights.…”

“…An event is surprising if it does not match our expectations 1,2,3,4 . The unexpected punchline of a joke 3 , the unexpected continuation of a sequence of tones 5 , harmonies 6,7 or images 8,9,10 , as well as rule switching such as shift of escape platform in the Morris watermaze 11 or meaning of cues 12,13,14,15 induce measurable physiological and behavioral reactions in humans and animals. Without expectations arising from previous experiences, an event such as the observation of a new image may be perceived as ‘novel’ but cannot be ‘surprising’ 16,17 .…”

Bayesian surprise shapes neural responses in somatosensory cortical circuits

“…Our demonstration of an increased surprise sensitivity in L2/3 complements previous literature that has identified signatures of stimulus mismatch and deviant stimulus responses in the supragranular cortical layers. L2/3 responses in multiple cortical regions were governed by stimulus predictability 38 and in the primary visual cortex by mismatched sensorimotor input, 39 responses to rarely presented stimuli were elevated above commonly presented stimuli, 40,41 and activity evoked by unexpected stimuli was elevated above that of both expected and random stimuli. 42 Intracellular recordings in L2/3 of the auditory cortex exhibited repression of response to repeated, expected stimuli as well as a late-phase subthreshold response indicative of detection of true deviating stimuli.…”