Similarity and Strength of Glomerular Odor Representations Define a Neural Metric of Sniff-Invariant Discrimination Time

Abstract

Summary The olfactory environment is first represented by glomerular activity patterns in the olfactory bulb. It remains unclear how these representations intersect with sampling behavior to account for the time required to discriminate odors. Using different chemical classes, we investigate glomerular representations and sniffing behavior during olfactory decision-making. Mice rapidly discriminate odorants and learn to increase sniffing frequency at a fixed latency after trial initiation, independent of odor identity. Relative to the increase in sniffing frequency, monomolecular odorants are discriminated within 10-40 ms, while binary mixtures require an additional 60-70 ms. Intrinsic imaging of glomerular activity in anesthetized and awake mice reveals that Euclidean distance between activity patterns and the time needed for discriminations are anti-correlated. Therefore, the similarity of glomerular patterns and their activation strengths, rather than sampling behavior, define the extent of neuronal processing required for odor discrimination, establishing a neural metric to predict olfactory discrimination time.

Multimodal learning of pheromone locations

“…To assess if such an activation of olfactory bulb (OB) circuitry can modulate other olfactory driven behaviors, we carried out a non-volatile olfactory discrimination learning and memory tasks in another groups of female mice ( Figure 3B ). To this end, we carried out go/no-go odor discrimination paradigm 21 to investigate modulation of discrimination learning and memory of non-pheromonal volatiles in Whitten effect induced female mice ( Figure 5B,C ). CI vs EU odor pair training was carried out for both the groups before Whitten effect was induced in experimental group to detect any possible bias due to the learning efficacy differences that might be existing between two groups ( Figure 5D : CI vs EU ordinary two-way ANOVA, Bonferroni’s multiple comparison test, P > .05 for all data points in learning curve).…”

“…To investigate the sniffing strategies of female mice toward male urine and female urine volatiles under whisker trimmed vs intact conditions, a head-restraining method was used to precisely deliver the volatiles on the snout of the mice. 21 These were subsets of mice utilized for multi-modal pheromonal learning assay and their sniffing strategies were tested 2 months after their 30th day memory was investigated. They were implanted with a head-post (custom-built head fixation set-up) on their head.…”

Local Postsynaptic Signaling on Slow Time Scales in Reciprocal Olfactory Bulb Granule Cell Spines Matches Asynchronous Release

“…With regard to further functions of slow signals, they are at first glance unlikely to play a direct role during odor discrimination or background segregation, since these discriminations usually occur within considerably less than 500 ms, even for difficult mixtures and/or many components ( Abraham et al, 2004 , 2010 ; Kepecs et al, 2005 ; Rokni et al, 2014 ; Bhattacharjee et al, 2019 ). Rather, slow signals may be involved in learning and plasticity, also during learning of the mixture discrimination task ( Abraham et al, 2004 , 2010 ; Gschwend et al, 2015 ).…”

Olfactory bulb granule cells: specialized to link coactive glomerular columns for percept generation and discrimination of odors

“…In addition to the learning curve and discrimination accuracy, the time required to discriminate two odors is a parameter tightly linked to the mechanisms of computations performed by the OB network and, hence, changes in synaptic processing and network function will have a high chance to affect odor discrimination time. The amount of time required by the OB network to discriminate two odors is governed by the similarity and strength of glomerular activity patterns and is independent of sampling behavior (Bhattacharjee et al 2019 ). Hence, discrimination of two monomolecular odorants with different glomerular representations requires less time than discrimination of a binary mixture of these odors (Abraham et al 2004 ).…”