Synchronous Spike Patterns in Macaque Motor Cortex during an Instructed-Delay Reach-to-Grasp Task - PubMed Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 Aug 10;36(32):8329-40.
doi: 10.1523/JNEUROSCI.4375-15.2016.

Synchronous Spike Patterns in Macaque Motor Cortex during an Instructed-Delay Reach-to-Grasp Task

Emiliano Torre  1 , Pietro Quaglio  1 , Michael Denker  1 , Thomas Brochier  2 , Alexa Riehle  3 , Sonja Grün  4
Affiliations

Synchronous Spike Patterns in Macaque Motor Cortex during an Instructed-Delay Reach-to-Grasp Task

Emiliano Torre et al. J Neurosci. .

Abstract

The computational role of spike time synchronization at millisecond precision among neurons in the cerebral cortex is hotly debated. Studies performed on data of limited size provided experimental evidence that low-order correlations occur in relation to behavior. Advances in electrophysiological technology to record from hundreds of neurons simultaneously provide the opportunity to observe coordinated spiking activity of larger populations of cells. We recently published a method that combines data mining and statistical evaluation to search for significant patterns of synchronous spikes in massively parallel spike trains (Torre et al., 2013). The method solves the computational and multiple testing problems raised by the high dimensionality of the data. In the current study, we used our method on simultaneous recordings from two macaque monkeys engaged in an instructed-delay reach-to-grasp task to determine the emergence of spike synchronization in relation to behavior. We found a multitude of synchronous spike patterns aligned in both monkeys along a preferential mediolateral orientation in brain space. The occurrence of the patterns is highly specific to behavior, indicating that different behaviors are associated with the synchronization of different groups of neurons ("cell assemblies"). However, pooled patterns that overlap in neuronal composition exhibit no specificity, suggesting that exclusive cell assemblies become active during different behaviors, but can recruit partly identical neurons. These findings are consistent across multiple recording sessions analyzed across the two monkeys.

Significance statement: Neurons in the brain communicate via electrical impulses called spikes. How spikes are coordinated to process information is still largely unknown. Synchronous spikes are effective in triggering a spike emission in receiving neurons and have been shown to occur in relation to behavior in a number of studies on simultaneous recordings of few neurons. We recently published a method to extend this type of investigation to larger data. Here, we apply it to simultaneous recordings of hundreds of neurons from the motor cortex of macaque monkeys performing a motor task. Our analysis reveals groups of neurons selectively synchronizing their activity in relation to behavior, which sheds new light on the role of synchrony in information processing in the cerebral cortex.

Keywords: cell assemblies; massively parallel spike trains; motor cortex; spike synchrony; temporal coding.

PubMed Disclaimer

Figures

Figure 1.
Figure 1.
Reach-to-grasp experimental protocol. The trial start (TS) was self-initiated by the monkey by closing a home switch. A WS prepared the monkey for the visual cue (C+ until C−) providing instruction about the grip type to use: SG or PG. One second later, a second visual cue turned on (GO), specifying the force needed to pull the object (HF or LF) and requesting movement initiation. The movement onset is marked by the switch release (SR). After object touch (OT), the monkey pulled the object and held it for 500 ms in a narrow position window until it was rewarded (RW). The timing of the behavioral events SR, OT, and RW, which follow the GO signal, varied depending on the monkey's reaction time and movement speed.
Figure 2.
Figure 2.
Trial types and epochs. Each panel shows the simultaneous spiking activity of all neurons (vertical axis) over time (horizontal axis) for four example trials (one per panel) of different types from a representative session in Monkey N. Each dot indicates one spike. The trials are aligned to trial start (TS). The six colored windows represent the position of the six epochs in the trials. The trigger associated to each epoch and the corresponding epoch name are shown at the bottom (for details, see Table 1). The movement (green) and hold (yellow) epochs are centered around triggers with occurrence times that changed from trial to trial depending on the reaction of the monkey to the GO signal and the movement time.
Figure 3.
Figure 3.
Significant patterns in one representative session. A, Composition of neuron identities (horizontal axis) of each significant pattern in the session (vertical axis, pattern IDs 0–5). B, Each panel shows one color map associated to a specific significant pattern. The color map shows the uncorrected p-value of the pattern's signature in each trial type and epoch (color progression in logarithmic scale). Squares with the black dot indicate the session's trial type and epoch where the pattern was classified as statistically significant.
Figure 4.
Figure 4.
Pattern statistics across sessions. Various statistics of the significant patterns detected across the 10 selected sessions separately for each monkey (columns). A–C, Total number of significant patterns in each epoch (colors mark the contribution of each trial type; A), their average size (B), and their average number of occurrences (C). D, Average distance between electrodes on which neurons are recorded that are involved in the same pattern. Whiskers in panels B–D indicate the SD when it is >0.
Figure 5.
Figure 5.
Dependence of synchrony on electrode distance. A, Distance of the electrodes on the recording array (white dots) from one reference electrode (cross at the bottom left corner). Dark to light shaded areas cover regions at progressively larger distances from the reference electrode from 0.4 mm up to 5.4 mm in steps of 1 mm. B, Histogram showing, for each distance range, the ratio between the number of neuron pairs involved in the same pattern and placed within that range from each other (summed across patterns found in any session, epoch and trial type) and the total number of recorded neuron pairs placed at that distance (vertical axis). The numbers inside each bar represent the relative height of each bar with respect to the sum of all bars; that is, the relative fraction of neuron pairs involved in the same patterns placed at a given distance.
Figure 6.
Figure 6.
Spatial arrangement of neurons participating in significant patterns. A, Locations of the electrode arrays in the motor cortex in the right hemisphere of the two monkeys. CS, Central sulcus; AS, arcuate sulcus; PD, precentral dimple. B, Color maps of the total number of SUAs recorded on each electrode of the recording array summed across all selected sessions. Red squares indicate the four unconnected electrodes. C, Color maps showing, for each electrode, the total number of patterns found that involved neurons recorded on that electrode. The number is obtained separately for each session (summing across all epochs and trial types) divided by the total number of single neurons recorded on the electrode in the session and finally summed across sessions. The final value can exceed the number of sessions (10) if, for instance, in each session there were more significant patterns involving the electrode than neurons recorded by that electrode. Gray squares correspond to a value of 0 and indicate electrodes that never recorded single neurons involved in significant patterns. Black lines connect electrode pairs extracted from each significant pattern found. D, Number of all possible pairs of recorded neurons placed along each orientation. Each panel shows a bar chart in polar coordinates. The direction of each bar corresponds to one orientation on the recording array, whereas the length of the bar represents the total number of neuron pairs placed along that direction computed for each session and summed across sessions. The circles have a diameter of 5000 for Monkey L and of 18,000 for Monkey N. E, Spatial orientation of all pairs of neurons involved in significant patterns found during movement as a fraction over the total number of recorded neuron pairs with that orientation shown in D.
Figure 7.
Figure 7.
Pattern specificity for different behavioral contexts. The four panels show the pattern specificity at each instance of the four behavioral contexts: epochs (A), trial types (B), grip modalities (C), and force levels (D). The values were calculated for each monkey across all sessions. The bars extend between the minimum and maximum value across sessions. The numbers above each bar indicate the number of sessions that actually contained significant patterns in that instance (e.g., in that epoch). When this number is 2, then the values taken by the 2 corresponding sessions are the ends of the range. The other sessions did not contain significant patterns, so a pattern specificity value could not be calculated.
Figure 8.
Figure 8.
Outcome of clustering for one representative session. A, Crosses indicate the neuron IDs (horizontal axis) of each pattern detected in the session (vertical axis; as in Fig. 3A). Red, blue, and green colors identify the membership of each pattern to one of k* = 3 clusters. B, Cost associated with each number k of clusters, k = 1, 2, …, 6.
Figure 9.
Figure 9.
Cluster specificity for different behavioral contexts. The four panels show the cluster specificity value for epochs (A), trial types (B), grip modalities (C), and force levels (D) analogous to Figure 7 for pattern specificity.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Abeles M. Role of cortical neuron: integrator or coincidence detector? Isr J Med Sci. 1982;18:83–92. - PubMed
    1. Abeles M. Corticonics: neural circuits of the cerebral cortex. Cambridge: Cambridge University; 1991.
    1. Azouz R, Gray CM. Dynamic spike threshold reveals a mechanism for synaptic coincidence detection in cortical neurons in vivo. Proc Natl Acad Sci U S A. 2000;97:8110–8115. doi: 10.1073/pnas.130200797. - DOI - PMC - PubMed
    1. Azouz R, Gray CM. Adaptive coincidence detection and dynamic gain control in visual cortical neurons in vivo. Neuron. 2003;37:513–523. doi: 10.1016/S0896-6273(02)01186-8. - DOI - PubMed
    1. Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Ser B. 1995;57:289–300.

Publication types

LinkOut - more resources