Odor Perception on the Two Sides of the Brain: Consistency Despite Randomness

Abstract

SUMMARY Neurons in piriform cortex receive input from a random collection of glomeruli, resulting in odor representations that lack the stereotypic organization of the olfactory bulb. We have performed in vivo optical imaging and mathematical modeling to demonstrate that correlations are retained in the transformation from bulb to piriform cortex, a feature essential for generalization across odors. Random connectivity also implies that the piriform representation of a given odor will differ among different individuals and across brain hemispheres in a single individual. We show that these different representations can nevertheless support consistent agreement about odor quality across a range of odors. Our model also demonstrates that, whereas odor discrimination and categorization require far fewer neurons than reside in piriform cortex, consistent generalization may require the full complement of piriform neurons.

Bilateral Alignment of Receptive Fields in the Olfactory Cortex

“…While there may be some functional organization in the glomerular layer (Mori, 1999;Murthy, 2011;Strotmann et al, 2000), whatever order present in the OB appears to be largely lost in the OC (Bekkers and Suzuki, 2013;Ghosh et al, 2011;Miyamichi et al, 2011;Sosulski et al, 2011;Wilson and Sullivan, 2011) (Figure 1A). This lack of apparent structure led to the prevalent view that OC projections are structurally random, each OC neuron receiving an arbitrary set of OB inputs (Bekkers and Suzuki, 2013;Choi et al, 2011;Davison and Ehlers, 2011;Ghosh et al, 2011;Kay, 2011;Miyamichi et al, 2011;Schaffer et al, 2018;Sosulski et al, 2011;Wilson and Sullivan, 2011).…”

“…Additionally, since the OB does not send contralateral projections, inter-hemispheric projections originate exclusively from the OC (Boyd et al, 2012;Haberly and Price, 1978;Kikuta et al, 2008;Scott et al, 1980). Consequently, the lack of apparent structure in the OC should propagate across hemispheres (Schaffer et al, 2018) (Figure 1A). In other words, the prevalent view of OC connectivity being random must apply to bilateral projections.…”

“…For example, there is some evidence for responses in olfactory cortical areas to contralateral odor stimulation, but it is limited by the number of neurons reported and a lack of comparison of ipsi-vs contralateral stimulation (Kikuta et al, 2008(Kikuta et al, , 2010Wilson, 1997). More recently, it was shown that despite the apparent randomness of the OB-OC projections, the piriform cortex could partially retain the olfactory map overlap present in the OB, allowing the OC in each hemisphere to perform similar odor generalization operations (Babadi and Sompolinsky, 2014;Schaffer et al, 2018).…”

A neuromimetic approach to the serial acquisition, long-term storage, and selective utilization of overlapping memory engrams

“…We hypothesize that such wiring schemes are implemented in association cortices and other subnetworks in which there is a requirement for breadth of learnable input correlations. Notably, spatial randomness of projections that allows for the self-organized development of input-representation correspondence is a design feature of the piriform cortex (35). Similar randomness is found in the organization of orientation-tuned neurons in mouse visual cortex; in animals with larger visual cortex, more elaborate organization schemes are thought to be driven by minimization of biological wiring length rather than by information processing considerations (36).…”

Interhemispheric connections between olfactory bulbs improve odor detection

“…Moreover, interneurons in the deafferented contralateral OB were also activated, thus confirming our hypothesis (Figure5F). The topographic odor map within the OB is not prominent in the rodent piriform cortex [52,[59][60][61] and in the zebrafish Dp [31,43]: axons from similarly tuned mitral cells project diffusely to these areas without strong spatial preference. Our results add to this by showing that Dp neurons in turn diffusely innervate the contralateral OB granular cell layer (Figure 1).…”