Limits...
Dual roles for spike signaling in cortical neural populations.

Ballard DH, Jehee JF - Front Comput Neurosci (2011)

Bottom Line: Our simulations show that this combination is possible if deterministic signals using γ latency coding are probabilistically routed through different members of a cortical cell population at different times.This model exhibits standard features characteristic of Poisson models such as orientation tuning and exponential interval histograms.In addition, it makes testable predictions that follow from the γ latency coding.

View Article: PubMed Central - PubMed

Affiliation: Department of Computer Science, University of Texas at Austin Austin, TX, USA.

ABSTRACT
A prominent feature of signaling in cortical neurons is that of randomness in the action potential. The output of a typical pyramidal cell can be well fit with a Poisson model, and variations in the Poisson rate repeatedly have been shown to be correlated with stimuli. However while the rate provides a very useful characterization of neural spike data, it may not be the most fundamental description of the signaling code. Recent data showing γ frequency range multi-cell action potential correlations, together with spike timing dependent plasticity, are spurring a re-examination of the classical model, since precise timing codes imply that the generation of spikes is essentially deterministic. Could the observed Poisson randomness and timing determinism reflect two separate modes of communication, or do they somehow derive from a single process? We investigate in a timing-based model whether the apparent incompatibility between these probabilistic and deterministic observations may be resolved by examining how spikes could be used in the underlying neural circuits. The crucial component of this model draws on dual roles for spike signaling. In learning receptive fields from ensembles of inputs, spikes need to behave probabilistically, whereas for fast signaling of individual stimuli, the spikes need to behave deterministically. Our simulations show that this combination is possible if deterministic signals using γ latency coding are probabilistically routed through different members of a cortical cell population at different times. This model exhibits standard features characteristic of Poisson models such as orientation tuning and exponential interval histograms. In addition, it makes testable predictions that follow from the γ latency coding.

No MeSH data available.


Related in: MedlinePlus

Interval histograms. (A) Spikes chosen randomly with a probability that equalizes the total number of spikes used in the γ latency coding. (B) 256 × 3 basis functions used to code 18 image patches (see text). (C) 256 × 2 basis functions used to code 12 image patches. (D) 256 × 2 basis functions. When the γ signal is restricted to a small range (50 ± 1 Hz), it can be clearly seen. For each data set spike interval histograms are computed at 1 ms resolution. The color is used to indicate the SE of the mean value for five runs.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3108387&req=5

Figure 7: Interval histograms. (A) Spikes chosen randomly with a probability that equalizes the total number of spikes used in the γ latency coding. (B) 256 × 3 basis functions used to code 18 image patches (see text). (C) 256 × 2 basis functions used to code 12 image patches. (D) 256 × 2 basis functions. When the γ signal is restricted to a small range (50 ± 1 Hz), it can be clearly seen. For each data set spike interval histograms are computed at 1 ms resolution. The color is used to indicate the SE of the mean value for five runs.

Mentions: Figure 7 shows the comparison of γ latency encodings with one set of random spike trains (Figure 7A) whose level of randomness has been adjusted to make the total spikes similar to those in the γ latency spike trains. All the comparisons use 1 ms resolution. The simulations lasted for 4 s, or approximately 200 γ intervals. We assume that the different patch encodings are not confused, since they use different phases and offsets and the spike trains are very sparse. For each graph, five different runs are made and the data combined in a interval histogram plot with the standard error in the mean at each sample denoted by color. Figure 7B shows that for the overcompleteness ratio of 7.68, the interval histogram is very similar to that observed in purely random spike data. Similarly for the overcompleteness ratio of 5.12, the histogram is very similar to the random plot. However, when the ratio is reduced to 5.12, and the γ range is also restricted to 50 ± 1 Hz, the effects of γ timing clearly can be seen; however, one must appreciate that this test is severe, as the encodings are designed to be similar, 512 cells are used, and the measurement interval is 4 s. Furthermore, experimental measurements of larger numbers of cells using local field potentials regularly show a γ signal.


Dual roles for spike signaling in cortical neural populations.

Ballard DH, Jehee JF - Front Comput Neurosci (2011)

Interval histograms. (A) Spikes chosen randomly with a probability that equalizes the total number of spikes used in the γ latency coding. (B) 256 × 3 basis functions used to code 18 image patches (see text). (C) 256 × 2 basis functions used to code 12 image patches. (D) 256 × 2 basis functions. When the γ signal is restricted to a small range (50 ± 1 Hz), it can be clearly seen. For each data set spike interval histograms are computed at 1 ms resolution. The color is used to indicate the SE of the mean value for five runs.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3108387&req=5

Figure 7: Interval histograms. (A) Spikes chosen randomly with a probability that equalizes the total number of spikes used in the γ latency coding. (B) 256 × 3 basis functions used to code 18 image patches (see text). (C) 256 × 2 basis functions used to code 12 image patches. (D) 256 × 2 basis functions. When the γ signal is restricted to a small range (50 ± 1 Hz), it can be clearly seen. For each data set spike interval histograms are computed at 1 ms resolution. The color is used to indicate the SE of the mean value for five runs.
Mentions: Figure 7 shows the comparison of γ latency encodings with one set of random spike trains (Figure 7A) whose level of randomness has been adjusted to make the total spikes similar to those in the γ latency spike trains. All the comparisons use 1 ms resolution. The simulations lasted for 4 s, or approximately 200 γ intervals. We assume that the different patch encodings are not confused, since they use different phases and offsets and the spike trains are very sparse. For each graph, five different runs are made and the data combined in a interval histogram plot with the standard error in the mean at each sample denoted by color. Figure 7B shows that for the overcompleteness ratio of 7.68, the interval histogram is very similar to that observed in purely random spike data. Similarly for the overcompleteness ratio of 5.12, the histogram is very similar to the random plot. However, when the ratio is reduced to 5.12, and the γ range is also restricted to 50 ± 1 Hz, the effects of γ timing clearly can be seen; however, one must appreciate that this test is severe, as the encodings are designed to be similar, 512 cells are used, and the measurement interval is 4 s. Furthermore, experimental measurements of larger numbers of cells using local field potentials regularly show a γ signal.

Bottom Line: Our simulations show that this combination is possible if deterministic signals using γ latency coding are probabilistically routed through different members of a cortical cell population at different times.This model exhibits standard features characteristic of Poisson models such as orientation tuning and exponential interval histograms.In addition, it makes testable predictions that follow from the γ latency coding.

View Article: PubMed Central - PubMed

Affiliation: Department of Computer Science, University of Texas at Austin Austin, TX, USA.

ABSTRACT
A prominent feature of signaling in cortical neurons is that of randomness in the action potential. The output of a typical pyramidal cell can be well fit with a Poisson model, and variations in the Poisson rate repeatedly have been shown to be correlated with stimuli. However while the rate provides a very useful characterization of neural spike data, it may not be the most fundamental description of the signaling code. Recent data showing γ frequency range multi-cell action potential correlations, together with spike timing dependent plasticity, are spurring a re-examination of the classical model, since precise timing codes imply that the generation of spikes is essentially deterministic. Could the observed Poisson randomness and timing determinism reflect two separate modes of communication, or do they somehow derive from a single process? We investigate in a timing-based model whether the apparent incompatibility between these probabilistic and deterministic observations may be resolved by examining how spikes could be used in the underlying neural circuits. The crucial component of this model draws on dual roles for spike signaling. In learning receptive fields from ensembles of inputs, spikes need to behave probabilistically, whereas for fast signaling of individual stimuli, the spikes need to behave deterministically. Our simulations show that this combination is possible if deterministic signals using γ latency coding are probabilistically routed through different members of a cortical cell population at different times. This model exhibits standard features characteristic of Poisson models such as orientation tuning and exponential interval histograms. In addition, it makes testable predictions that follow from the γ latency coding.

No MeSH data available.


Related in: MedlinePlus