The example given in the text is of the relationship between a fire alarm, smoke, and fire. 

Fire causes smoke, which causes the fire alarm to go off. The three links are necessary in the graph because fire does not directly cause the fire alarm to go off. Therefore if we know that there is smoke, the presence of fire does not lift nor lower the probability of hearing the fire alarm go off. 

Example of the Screening Off Effect

The mediator is a variable that functions as the mechanism by way of transferring an effect of one thing to another.  

Consider the chain junction, where $A -> B -> C$.  $B$ is the mediator that transfers the effect of $A$ to $C$.   In other words, the mediator, $B$, is the variable that explains the causal effect of $A$ on $C$.  $A$ can be an indirect cause of $C$, meaning that $A$ affects $B$, which in turn contributes to the change in $C$.  

Mediator

When the junction $$A \rightarrow B \rightarrow C$$ appears in a causal diagram, we note that the presence of $$B$$ "screens-off" the value of $$A$$. This means that $$C$$ and $$A$$ are conditionally independent given the value of $$B$$.

University of Michigan - Ann Arbor

Google

Pearl outlines 3 basic types of junctions in Causal Diagrams which appear commonly and help explain the data. 

The three junctions take the forms:
   1. $A \rightarrow B \rightarrow C$ (Chain)
 2. $A \leftarrow B \rightarrow C$ (Fork)
 3. $A \rightarrow B \leftarrow C$ (Collider)

The 3 Junctions in Causal Diagram

Pearl, J., & Mackenzie, D. (2018). $\mathit{The \ Book \ of \ Why: \ The \ New \ Science \ of \ Cause \ and \ Effect.}$ Basic Books.

The Book of Why

When the junction $A \leftarrow B \rightarrow C$ appears in a Causal Diagram, we say that B is a confounder for both A and C. It might appear that there is a causal relationship between A and C, because there is an observable relationship between A and C when B is varied. However, if we condition on B ("hold B constant") we observe that A and C become conditionally independent. 

$A \leftarrow B \rightarrow C$: The Fork and Confouders

When a single node is caused by two other nodes we encounter something called "collider-bias" where one condition explains-away another. 

Since A and C both effect the outcome of B, by conditioning on B, A and C become **dependent**.

$A \rightarrow B \leftarrow C$: The Collider

A proxy is used as a substitute measure when a variable that is not directly measurable is needed for analysis.  For a variable to be a proxy for another variable, the proxy variable needs to be highly correlated with the variable of interest.  The proxy can be used for any variables such as a mediator, confounder, collider.  

Learn Before

Related

Learn After