A network representation in which the same set of nodes is connected by two or more coexisting types (layers) of ties—each type representing a distinct relational domain, interaction channel or role—so that interlayer structure and coupling influence system‑level dynamics.

Principle

Layered relation types create interdependencies that change diffusion, resilience and centrality: processes that ignore layer structure (aggregate networks) can misestimate reach, bottlenecks and the role of actors who bridge layers.

Demonstration

Illustrative scenario → Situation: In a city, persons are connected through family ties (layer A), workplace ties (layer B) and a local messaging group (layer C). Recognition: Analysts model these as separate layers in a multiplex network. Action: A public‑health message is released via the messaging group and amplified through workplace ties. Consequence: The combined layer dynamics produce faster and broader uptake than any single layer would predict, showing how multiplex coupling alters spread.

Misapplication

Aggregating different relation types into a single undifferentiated network without preserving layer identity; the semantic error is treating heterogeneous channels as equivalent, which can obscure cross‑layer bottlenecks or reinforcing effects.

Consequence

Multiplex structure affects thresholds for contagion, diffusion speed, systemic robustness and the identification of strategic actors (e.g., cross‑layer brokers); policy or intervention design must account for layer coupling to be effective.

Reversal

If layers are highly redundant (strongly correlated), the multiplex model may collapse effectively to a single‑layer representation for certain dynamics; conversely, when layers are independent, multiplex effects are amplified and simplification produces larger modeling errors.

Boundary

Clearly within: a model with separate family, work and online friendship layers connecting the same individuals. Boundary case: two partially overlapping interaction types where one is sparsely observed. Clearly outside: a single‑layer network that represents only one relation type (e.g., only economic transactions).

Semantic Tension

Parsimony versus fidelity: aggregating layers simplifies analysis but sacrifices the fidelity needed to capture interlayer effects; retaining layers preserves explanatory power at the cost of complexity.

Synthesis

A multiplex network is not merely multiple ties per node but a layered analytical object: understanding social processes requires examining how layers interact, correlate and transfer effects rather than treating relations as homogeneous.