Article Figures & Data
Figures
- Figure 1.
STDP rules. At the synaptic level, but to a lesser extent at the systems level, it has been demonstrated that the relative timing of activity between the presynaptic neuron and postsynaptic neuron is crucial for plasticity. When presynaptic activity repetitively occurs within several milliseconds prior to postsynaptic activity, LTP (red) is induced. When the timing is reversed, LTD (blue) is induced. The potential for LTP or LTD decreases as the time window between the presynaptic and postsynaptic activity at the synapse increases (Song et al., 2000).
- Figure 2.
Experimental showcase. A, Rats were chronically implanted with three pairs of subcutaneous stainless steel microwires to stimulate and record from the ECR, trapezius (Trap), and biceps (Bi) muscles contralateral to the cortical array. B, Dorsal view of the rat brain, showing where we inserted the 2 × 2 platinum-iridium electrode array in the CFA of M1. Coordinates are anterior (A) and lateral (L) relative to bregma (B). C, Corticomotor excitability was assessed before and after PAS using closed-loop, EMG-controlled motor cortical stimulation. The top envelope (red) of the EMG signal was calculated in pseudo-real time on the computer controlling data acquisition and stimulation using the MATLAB envelope function. Cortical stimulation was invoked when the envelope rose within 2–12 SDs above the mean signal (green horizontal lines) for at least 50 ms. The minimum time between stimulations was 1 s. D, We know from our and previous studies under anesthesia that it takes ∼9 and 3 ms for signals issued from cortical and peripheral stimulation to arrive at the spinal cord, respectively. Peripheral stimulation of afferent fibers results in a volley of motor cortical activity after 16 ms. The presynaptic and postsynaptic activity offset at the levels of spinal cord and motor cortex for different interstimulus intervals were calculated based on these conduction latencies.
- Figure 3.
Example experiment-level result: MEPs recorded in the right ECR of one rat, obtained from electrical cortical stimulation for each probe. Shaded areas in light blue and red indicate 1 SD about the mean. Inset, Each session began with three 5 min probes in which we performed closed-loop EMG-dependent cortical stimulation to assess baseline MEP amplitudes, each separated by 10 min. The PAS session itself, involving 300 pairs of stimuli to the cortex and the muscle at a rate of 0.5 Hz, took ∼10 min. This was followed by three post-PAS probes so we could assess corticospinal excitability up to 30 min after paired stimulation for each interstimulus interval. After each experiment, we manually verified all MEPs using custom software and excluded traces with movement artifacts or noisy EMG signals.
- Figure 4.
PAS does not significantly potentiate MEP responses in vivo. Grouped bar plot, depicting that for Post 1 (2 min after PAS, light red), Post 2 (17 min after PAS, medium red), and Post 3 (32 min after PAS, dark red) sessions, there were no significant differences between STDP experimental conditions and control conditions. Error bars are 95% confidence intervals about the mean. The horizontal reference line marked in red signifies no change between that post condition and the baseline average. Control conditions are shown to the right of the vertical dotted black line, to separate them from the ISI conditions tested to the left (Cx, cortical stimulation only; Ms, muscle stimulation only; No, no stimulation). ISI refers to the latency between stimulation of the cortex and the muscle. Here, positive numbers refer to muscle stimulation occurring after cortical stimulation. “Spinal Cord” numbers are the estimated latencies between the arrival of the descending volley onto the motoneurons in the spinal cord and the arrival of the antidromic action potentials evoked from muscle stimulation (positive if orthodromic arrives before antidromic); “Motor Cortex” numbers are the estimated latencies between the arrival in M1 of peripheral stimulation-induced afferent activity and the motor cortex stimulation (positive if peripheral afferent signal arrives before cortex stimulation signal). The colors on the horizontal bars at the bottom indicate conditions expected to induce LTP-like effects (red), LTD-like effects (blue), or no significant modulation (green) based on the Hebbian STDP model.
Tables
Figure Type of test Term Data structure df Numerator df Denominator F Value p Value 4 Mixed-effects ANOVA a. Session × Condition Model residuals normal 28.00 250.92 0.53 0.976 b. Condition 14.00 253.60 1.56 0.092 c. Session 2.00 250.92 0.08 0.921 Mixed-effects model analysis, fixed effects (type III) using the REML method. df = Degrees of Freedom.
Figure 2-1
Spinal and cortical evoked potentials from peripheral stimulation. A, The latency of the deepest trough response at 3 ms in the C5 region of the spinal cord, averaged across 200 stimulations. Note that a DC offset in the baseline signal was manually adjusted here. B, Measuring the latency of the deepest trough response at 16 ms in the cortex, averaged across 360 stimulations. Download Figure 2-1, EPS file.
Figure 4-1
Study design for chronic experiments and a priori randomization order (number within each cell). The final sample sizes for each PAS condition are written to the left. Sessions used in the final dataset are shaded in gray. Download Figure 4-1, DOC file.
Figure 4-2
Muscle distribution. Distribution of muscles used as PAS target (ISI Conditions, Latencies are Stimulation Offsets). *PAS muscle stimulation component was between one EMG electrode and reference. Download Figure 4-2, DOC file.
Figure 4-3
No individualized effect of PAS. Analogous plot to Figure 4, but paneled by rat, demonstrating that there was no effect consistent with STDP from our PAS intervention even on the level of individual animals. Download Figure 4-3, EPS file.
Figure 4-4
No cumulative effect of PAS. Baseline MEP amplitude on days immediately following a PAS session was not correlated to MEP changes in that session. The “Within-Session MEP Ratio” is defined as the normalized change in averaged corticomotor excitability across a given PAS intervention WITHIN an experimental day (N), and the “Next-Day Baseline MEP Ratio” is defined in the same way but BETWEEN the baseline average on day N and the baseline average on day N + 1 (consecutive calendar day). While processing the data BETWEEN days, we ensured that there were no changes in cortex stimulation intensity, muscle stimulation intensity, the electrode leads used for both sites, the type of EMG recording (monopolar or bipolar), and the EMG range for closed-loop stimulation. If any of these parameters changed between days, those data points were excluded. This ensured full homogeneity in the stimulation conditions used to calculate the appropriate quotients. All of the above parameters were held constant by design within a particular experimental day. By testing the correlation between these two ratios, we could directly assess whether the change induced by any particular PAS protocol on a given day is related to the change in baseline excitability across days. The two variables were not strongly correlated (r = 0.18) and the relationship was not significant (p = 0.21), even after winsorization to remove outliers (p = 0.19), confirming there was no carryover effect of our PAS intervention. Download Figure 4-4, EPS file.
In this issue
