Skillful, voluntary movements are underpinned by computations performed by networks of interconnected neurons in the primary motor cortex (M1). Computations are reflected by patterns of coactivity between neurons. Using pairwise spike time statistics, coactivity can be summarized as a functional network ( FN ). Here, we show that the structure of FNs constructed from an instructed-delay reach task in nonhuman primates is behaviorally specific: Low-dimensional embedding and graph alignment scores show that FNs constructed from closer target reach directions are also closer in network space. Using short intervals across a trial, we constructed temporal FNs and found that temporal FNs traverse a low-dimensional subspace in a reach-specific trajectory. Alignment scores show that FNs become separable and correspondingly decodable shortly after the Instruction cue. Finally, we observe that reciprocal connections in FNs transiently decrease following the Instruction cue, consistent with the hypothesis that information external to the recorded population temporarily alters the structure of the network at this moment. Author Summary It remains unclear how motor cortical neurons flexibly perform the computations necessary to generate movement. We hypothesized that neuronal coactivity contains movement information, and its dynamics can reveal how the population switches computations during a task. We quantified coactivity as a functional network (FN) with single neurons as nodes and population coactivity as directed weighted edges. We also constructed FNs within short epochs across a trial to determine when coactivity begins to carry information and to investigate the dynamic structure of these interactions. Following the Instruction cue, reciprocal connections in FNs transiently decrease, and shortly after, FNs become maximally decodable for reach direction.

Author Summary Mapping the routing of spikes through local circuitry is crucial for understanding neocortical computation. Under appropriate experimental conditions, these maps can be used to infer likely patterns of synaptic recruitment, linking activity to underlying anatomical connections. Such inferences help to reveal the synaptic implementation of population dynamics and computation. We compare a number of standard functional measures to infer underlying connectivity. We find that regularization impacts measures heterogeneously, and that individual algorithms have unique biases that impact their interpretation. These biases are nonoverlapping, and thus have the potential to mitigate one another. Combining individual algorithms into a single ensemble method results in a stronger inference algorithm than the best individual component measure. Ensemble-based inference can yield higher sensitivity to underlying connections and an improved estimate of the true statistics of synaptic recruitment. Abstract A promising alternative to directly measuring the anatomical connections in a neuronal population is inferring the connections from the activity. We employ simulated spiking neuronal networks to compare and contrast commonly used inference methods that identify likely excitatory synaptic connections using statistical regularities in spike timing. We find that simple adjustments to standard algorithms improve inference accuracy: A signing procedure improves the power of unsigned mutual-information-based approaches and a correction that accounts for differences in mean and variance of background timing relationships, such as those expected to be induced by heterogeneous firing rates, increases the sensitivity of frequency-based methods. We also find that different inference methods reveal distinct subsets of the synaptic network and each method exhibits different biases in the accurate detection of reciprocity and local clustering. To correct for errors and biases specific to single inference algorithms, we combine methods into an ensemble. Ensemble predictions, generated as a linear combination of multiple inference algorithms, are more sensitive than the best individual measures alone, and are more faithful to ground-truth statistics of connectivity, mitigating biases specific to single inference methods. These weightings generalize across simulated datasets, emphasizing the potential for the broad utility of ensemble-based approaches.