Display Settings:

Items per page

Results: 9

1.
Fig 4

Fig 4. From: An open hypothesis: is epilepsy learned, and can it be unlearned?.

Our computational model. Represented are nodes i and j with respective spontaneous firing probabilities S(i) and S(j). The connection strengths are represented by the conditional firing probabilities P(i,j) and P(j,i).

David Hsu, et al. Epilepsy Behav. ;13(3):511-522.
2.
Figure 9

Figure 9. From: An open hypothesis: is epilepsy learned, and can it be unlearned?.

Top diagram: simple chain diagram with one intermediate node. Bottom diagram: simple branching diagram with two parallel intermediate nodes.

David Hsu, et al. Epilepsy Behav. ;13(3):511-522.
3.
Fig 5

Fig 5. From: An open hypothesis: is epilepsy learned, and can it be unlearned?.

Recovery after simulated seizure. A, the spontaneous firing probability declines with seizure onset at arrow, and gradually recovers after seizure stops. B, Connectivity recovers faster than the spontaneous firing probability, and overshoots for 50 million time steps (or 55 hours) in this example. From Hsu et al [21].

David Hsu, et al. Epilepsy Behav. ;13(3):511-522.
4.
Figure 3

Figure 3. From: An open hypothesis: is epilepsy learned, and can it be unlearned?.

How the branching ratio, σ, influences the spread of network activity. The left column shows the three ways activity could spread in a network, characterized by the three values of σ. The far right column shows avalanche size distributions produced by a network model with these different values of σ. Note that the power law distribution (middle plot; dashed line) only occurs when σ = 1. Adapted from Beggs [27].

David Hsu, et al. Epilepsy Behav. ;13(3):511-522.
5.
Figure 2

Figure 2. From: An open hypothesis: is epilepsy learned, and can it be unlearned?.

Power law of neuronal avalanches. Distribution of sequence sizes taken from acute slice LFPs recorded with a 60 electrode array, plotted in log-log space. Actual data are shown in black, while the output of an independent Poisson model is shown in red. The actual data follow a nearly straight line for sequence sizes from 1– 35; after this point there is a cutoff induced by the electrode array size. The nearly straight line is indicative of a power law. From Beggs [27].

David Hsu, et al. Epilepsy Behav. ;13(3):511-522.
6.
Fig 6

Fig 6. From: An open hypothesis: is epilepsy learned, and can it be unlearned?.

Simulated acute deafferentation. A system of 100 nodes is suddenly reduced (arrow) to 10 nodes. A, Steady state spontaneous firing probability is higher for the smaller system. B, Acute deafferentation is accompanied by an immediate drop in the relative firing rate, which slowly recovers as the spontaneous firing probability attains its new steady state value. In the meantime, connectivity rises to supercritical levels and remains there for 8 million time steps (9 hours) in this example. From Hsu et al [21].

David Hsu, et al. Epilepsy Behav. ;13(3):511-522.
7.
Figure 1

Figure 1. From: An open hypothesis: is epilepsy learned, and can it be unlearned?.

Population spikes recorded at the network level. Upper left, cortical slice on the 60-channel microelectrode array. Electrodes appear as small black circles at the ends of lines. Upper right, local field potential (LFP) signals on electrodes. Large LFPs are caused by the synchronous spiking of many neurons near the electrodes, as seen in interictal spikes. Bottom, suprathreshold LFPs represented by small black squares. A sequence of 3 active frames is shown. The sequence has a size of 10, as this is the number of electrodes driven over threshold. Adapted from Beggs [27].

David Hsu, et al. Epilepsy Behav. ;13(3):511-522.
8.
Fig 7

Fig 7. From: An open hypothesis: is epilepsy learned, and can it be unlearned?.

Cross-correlograms of Markovian and non-Markovian connectivity. A, Network simulation can produce either Markovian (dark line, rapidly falling) or non-Markovian (light line, with bumps) connectivity. The number of bumps can be varied by varying the non-Markovian parameters of the model (see Methods Aim 3). That is, through our simulations we can produce many bumps, one bump or no bumps. B, Actual data from two cortical slices. The dark curve is rapidly falling and indicates Markovian connectivity. There is a broad shoulder with peak at 40–50 ms suggestive of possible non- Markovian connectivity as well. There are also examples where no shoulder is apparent (not shown). The lighter curve with a distinct, separate bump peaking near 120 ms is strong evidence of non- Markovian connectivity. In this sample, there is only one non-Markovian bump. Note different time scales in the two plots.

David Hsu, et al. Epilepsy Behav. ;13(3):511-522.
9.
Figure 8

Figure 8. From: An open hypothesis: is epilepsy learned, and can it be unlearned?.

Plasticity induced by electrical stimulation at 10 Hz. Each plot shows cross-correlograms for all electrodes with respect to one reference electrode. All correlograms are superimposed to give a summary of temporal relationships in the entire network. Plots A and B were taken from the first and second 10 minutes of the baseline period. Note minor differences in correlograms, suggesting relative stability. Plot C is the overall correlogram for the 20 min pre stimulation period. Plot D is the correlogram for the 20 min period that began 10 min after stimulation. Note the appearance of a new bump (at arrow) near −100 ms. This bump is probably not the result of drift, as there was no evidence of it in the pre stimulation baseline plots A or B. Evidence for long term plasticity of this sort has been seen in 3/3 experiments so far.

David Hsu, et al. Epilepsy Behav. ;13(3):511-522.

Display Settings:

Items per page

Supplemental Content

Recent activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...
Write to the Help Desk