Does Akka onReceive method execute concurrently?

Here is my scenario:

I have a master actor, which receives messages from multiple child actors. These messages contain data to be aggregated. In this aggregation logic, do I need to take care of synchronization issues, if I use a shared data structure to collect the aggregation?

else if(arg0 instanceof ReducedMsg){

                           ReducedMsg reduced = (ReducedMsg)arg0;
        counter.decrementAndGet();

        synchronized(finalResult){

            finalResult.add((KeyValue<K, V>) reduced.getReduced());

            if(counter.get() == 0){
                                    if(checkAndReduce(finalResult)){

                    finalResult.clear();
                }
                else{
                    stop();
                    latch.countDown();
                }

            }

        }



    }

So as you can see I have a finalResult, to which each message will be aggregated, and after a processing logic the collection needs to be cleared as well.

Actually what I am trying to implement is a recursive (associative) reduction mapreduce. So I need to keep the synchronized block I assume? Or is it by any chance Akka executes the onReceive one thread at a time?

This logic produces accurate and predictable result on small data set. My problem is when my input data set is a little large, the code hangs. I want to be sure that is because of the context switches for my synchronization block, so that I may dwelve into a different design.