Bilateral and unilateral odor processing and odor perception

Abstract

Imagine smelling a novel perfume with only one nostril and then smelling it again with the other nostril. Clearly, you can tell that it is the same perfume both times. This simple experiment demonstrates that odor information is shared across both hemispheres to enable perceptual unity. In many sensory systems, perceptual unity is believed to be mediated by inter-hemispheric connections between iso-functional cortical regions. However, in the olfactory system, the underlying neural mechanisms that enable this coordination are unclear because the two olfactory cortices are not topographically organized and do not seem to have homotypic inter-hemispheric mapping. This review presents recent advances in determining which aspects of odor information are processed unilaterally or bilaterally, and how odor information is shared across the two hemispheres. We argue that understanding the mechanisms of inter-hemispheric coordination can provide valuable insights that are hard to achieve when focusing on one hemisphere alone.

Fig. 1. Summary of inter-hemispheric connections in the olfactory system.

a Schematic representation of the inter-hemispheric connections in the olfactory cortex. The unilateral projections are not shown. Circles represent cell bodies, arrow-heads show the projection target. Solid lines represent dense projections and dashed lines represent minor or weak projections. The green arrow originates from all subdivisions (D,V,L,M) of the ipsi-AON, and targets the dorsal contralateral AON. Black arrows mark the information flow from the OB to the OC and vice versa. Yellow box represents the anterior commissure (AC). aPC, anterior piriform cortex; pPC, posterior piriform cortex; AON, anterior olfactory nucleus; OB, olfactory bulb; GL, glomerular layer; GCL, granule cell layer; MCL, mitral cell layer;VP, AON pars ventroposterior; D, dorsal; V, ventral; M, medial; L, lateral; A, anterior; P, posterior. b Inset of (A), focusing on the interbulbar connections and the bilateral connections between the AON, AONpE and OB. The AONpE preserves the glomerulus topography and projects topographically to the contralateral OB beneath the iso-functional glomeruli. Double-headed arrow denotes reciprocal connections between the AONpE and the AON pars dorsalis. AONpE, AON pars externa.

Fig. 2. The direct and the indirect pathways underlying functional bilateral connectivity.

a OC neurons receive inputs from several ipsilateral M/T cells and different OC neurons integrate from a varied number of different M/T cells. OC neurons can receive input directly from the contralateral OC neurons (black arrow). The direct pathway: An odorant inhaled through the contra-nostril activates a glomerulus pattern in the contra-bulb (four colored circles in the glomerulus layer). This pattern then activates several M/T cells in the contra-OB which converge to a single contra-OC neuron (black pyramid). This cortical neuron in turn projects to a corresponding ipsilateral OC neuron via the AC. The big X represents an occluded nostril. b Summary of the indirect pathway. Odorants inhaled through the contra-nostril activate a glomerular pattern in the contra-OB. This pattern activates several M/T cells in the contra-OB, which results in activating mirror-symmetric M/T cells in the ipsi-OB by first activating the contra-AONpE neurons. These mirror-symmetric M/T cells in the ipsi-OB converge to an ipsi-OC neuron (black pyramid), leading to sharing the odor code across the two hemispheres. In addition, the odorant-activated contra-M/T cells converge to a contra-OC neuron (not shown). Thick gray arrow depicts the course of the indirect pathway. c Two examples of ipsi- and contralateral light activation maps of M/T cells. Each pixel represents the average firing rate caused by ~50 optogenetic stimulations of each spot on the grid. Mice expressing ChR2 in M/T cells were used. The values were obtained using a 200-ms window after light onset. The gray bar marks the estimated boundary between the ipsi- and contralateral bulbs. The example in the upper panel was constructed by scanning both bulbs in the same experiment and the lower panel example was constructed from two independent light-scan experiments of the ipsi- and contralateral bulbs. Panels C-E were taken from. d Raster plots of all 108 recorded neurons’ ipsi- and contralateral hotspot responses. Each bin is the average firing ring rate in a 25 ms window. In four neurons the responses to ipsi-light stimulation exceeded 100 spikes/s and were truncated to 100 for better visibility of low firing rates. Double-headed arrow marks the neurons that received significant excitatory input from contralateral light stimulations. e The median response for each 25 ms bin of all ipsilateral and contralateral significantly responding neurons. The duration of the stimuli is indicated by the cyan bar. Responses to stimulation from the contra-bulb tended to have a lower peak and lasted longer.

Fig. 3. Inter-OC connections may enable sharing of odor representation across the two hemispheres.

An odorant inhaled through the left nostril activates several left-OC neurons. These neurons in turn activate several right-OC neurons through non-reciprocal connections, but lead to an odor representation in the right-OC that is similar to the odor representation when the odorant is delivered directly through the right nostril. Black pyramids represent neurons activated regardless of whether the odorant was delivered to the left or the right nostril. Blue and red pyramids represent neurons activated by left-only or right-only delivered odorants, respectively.

Fig. 4. Unilateral and bilateral memory accesses.

a Rat pups were conditioned to a cedar odorant by pairing it with a milk reward. 18-day old pups that were trained with one nostril plugged (marked with a black X) showed a preference for the cedar odorant with either the trained or the untrained nostril. Green V, preference to cedar odorant; Red X, no preference. b 6-day old pups did not show preference for cedar with the untrained nostril but did show preference when reaching the age of 18 days. This suggests that a unilateral odor memory is either bilaterally accessible at the age of 18 days or that it was transmitted to the contralateral side. c 6-day old pups trained with one nostril plugged could not retrieve the memory at 18 days if the AC is cut at that age. Thus, the memory is stored unilaterally and is not transmitted to the contralateral side although it can be accessed through the AC from the contralateral side.

Fig. 5. Models of inter-bulb connections and their putative relevance for odorant localization.

A high concentration of an odorant is delivered to one side (ipsi) and a lower concentration of the same odorant is delivered to the other (contra) side. a Assuming no inter-bulb connections, a higher odorant concentration at the ipsi-nostril leads to higher firing rate of the ipsi-M/T cell than the contra-M/T cell. A large difference enables the animal to localize the odorant on the ipsi-side. b Excitatory inter-bulb connections lead to an increase in the M/T cell firing rates on each side, proportional to the excitatory input from the M/T cell on the contra-side via the AONpE. Assuming the ipsi-M/T cell contributes 50% of its firing rate to the contraM/T cell and vice versa, the difference between firing rates will decrease, making odorant localization more difficult. c With inhibitory inter-bulb connections, the higher firing rate of the ipsi-M/T cell leads to strong inhibition on the already weakly-active contra-M/T cell, while the ipsi-M/T cell experiences negligible inhibition in return. As a result, the contrast between the two sides is maximized, making odorant localization easier.