Delay before synchronization and its role in latency of sensory awareness

Here we show that for coupled-map systems, the length of the transient prior to synchronization is both dependant on the coupling strength and dynamics of connections: systems with fixed connections and with no self-coupling display quasi-instantaneous synchronization. Too strong tendency for synchronization would in terms of brain dynamics be expected to be a pathological case. We relate how the time to synchrony depends on coupling strength and connection dynamics to the latency between neuronal stimulation and conscious awareness. We suggest that this latency can be identified with the delay before a threshold level of synchrony is achieved between distinct regions within the brain, as suggested by recent empirical evidence, in which case the latency can easily be understood as the inevitable delay before such synchrony builds-up. This is demonstrated here through the study of simplistic coupled-map models.

💡 Research Summary

The paper investigates how the transient period preceding synchronization in coupled‑map systems depends on both coupling strength and the dynamics of the connections, and it links this phenomenon to the latency observed between sensory stimulation and conscious awareness. Using two variants of a coupled‑map lattice—one with fixed connections and no self‑coupling, the other with adaptive connection weights that evolve according to the instantaneous state differences between nodes—the authors systematically explore the time required for the network to reach a globally synchronized state. In the fixed‑connection case, once the coupling strength ε exceeds a critical value εc, the system synchronizes almost instantaneously, typically within one or two iterations; this is termed “quasi‑instantaneous synchronization.” By contrast, in the dynamically‑weighted network, even when ε > εc, the continual adjustment of weights creates feedback loops that substantially prolong the transient. Depending on the parameters, the network may need tens to hundreds of iterations before a threshold level of synchrony is achieved. Adding self‑coupling further slows the process because each node’s own past state interferes with the convergence of the ensemble.

The authors argue that these mathematical results map onto brain dynamics. In the cortex, synaptic plasticity and fluctuating functional connectivity make the neural network resemble the dynamic‑connection model rather than a static one. When a sensory stimulus arrives, local neuronal assemblies become active and then interact with distant regions. Conscious perception, according to a growing body of neuroimaging evidence, only emerges after a sufficient degree of inter‑regional synchrony is reached. Consequently, the latency between stimulus onset and awareness can be interpreted as the inevitable time needed for the brain’s distributed network to build up the required synchrony. Weak coupling (e.g., low effective connectivity) or highly variable connections (e.g., rapid plastic changes) lengthen this build‑up period, producing longer perceptual latencies. Conversely, overly strong coupling combined with rigid, fixed connections would yield near‑instantaneous synchrony, a situation that is biologically implausible and reminiscent of pathological hyper‑synchronization seen in epileptic seizures.

Empirical support is drawn from recent EEG and fMRI studies showing a gradual increase in phase‑locking or coherence between sensory and higher‑order cortical areas in the milliseconds to seconds preceding conscious report. This gradual rise mirrors the model’s transient dynamics, reinforcing the proposal that the “delay before synchronization” is a mechanistic substrate of sensory awareness latency. The paper concludes that recognizing this delay provides a parsimonious explanation for normal perceptual timing, offers a framework for interpreting disorders of consciousness, and suggests design principles for artificial neural systems that must balance rapid integration with stability.