Olfactory bulb network: neurogenetic restructuring and odor decorrelation (Chow et al. 2012)

 Download zip file 
Help downloading and running models
Adult neurogenesis in the olfactory bulb has been shown experimentally to contribute to perceptual learning. Using a computational network model we show that fundamental aspects of the adult neurogenesis observed in the olfactory bulb -- the persistent addition of new inhibitory granule cells to the network, their activity-dependent survival, and the reciprocal character of their synapses with the principal mitral cells -- are sufficient to restructure the network and to alter its encoding of odor stimuli adaptively so as to reduce the correlations between the bulbar representations of similar stimuli. The model captures the experimentally observed role of neurogenesis in perceptual learning and the enhanced response of young granule cells to novel stimuli. Moreover, it makes specific predictions for the type of odor enrichment that should be effective in enhancing the ability of animals to discriminate similar odor mixtures. NSF grant DMS-0719944.
1 . Chow SF, Wick SD, Riecke H (2012) Neurogenesis drives stimulus decorrelation in a model of the olfactory bulb. PLoS Comput Biol 8:e1002398 [PubMed]
Citations  Citation Browser
Model Information (Click on a link to find other models with that property)
Model Type: Realistic Network;
Brain Region(s)/Organism: Olfactory bulb;
Cell Type(s): Olfactory bulb main mitral cell; Olfactory bulb main interneuron granule MC cell;
Gap Junctions:
Simulation Environment: MATLAB;
Model Concept(s): Activity Patterns; Rate-coding model neurons; Sensory processing; Apoptosis; Neurogenesis;
Implementer(s): Chow, Siu-Fai ;
Search NeuronDB for information about:  Olfactory bulb main mitral cell; Olfactory bulb main interneuron granule MC cell;
matlab code for
Chow S-F, Wick SD, Riecke H (2012) Neurogenesis Drives Stimulus
Decorrelation in a Model of the Olfactory Bulb. PLoS Comput Biol 8(3):
e1002398. doi:10.1371/journal.pcbi.1002398

run main.m for figures similar to fig.2, but with only 26 channels
    set sim = 10 for 442 channels

run main2.m for figures similar to fig.9
	enrich = [1 2]; for related enrichment
    enrich = [3 4]; for unrelated enrichment


    1: input patterns and correlations
    2: output patterns and correlations, connection
    3: GC information
    4: correlation
        blue: mean correlation of all paris
        red: mean correlation of tracked pairs (see parameter tracking)
    101: 2d representation of input patterns
    102: 2d representation of output patterns


    sim: level of down sampling
        for the default input set, sim 40 -> 26 channels, sim 10 ->
        442 channels
    odor_names: file names of patterns from http://gara.bio.uci.edu/
    choose: odors chosen to be in the training set

    non_lin: 0 for linear network, 1 for rectified nonlinear network
             (much slower)
    conn: number of connection each GC makes (mean # connection for
          prob_conn = 1)
    CS: coupling strength

    ts: threshold of survival function
    gamma: steepness of survival function
    th: minimal GC activity that would count towards survival
    rm, rg: thresholds for rectifer, only works for non_lin == 1

    cont_density: 0 for discrete GC population, 1 for population description
        population description version runs very slow for large network
    exp_time: total experiment time
    step: plotting/output interval
    dt: for equation stepping

    track mean correlation for a subset of the odor pairs

    We gratefully acknowledge the support of NSF grant DMS-0719944

20121127 matlab code in main.m main2.m modified by replacing ~ with
the variable "ignore" for backwards compatibility with matlab versions
before R2009b.