This paper presents a procedure to combine the top association hypotheses generated in a track-to-track (T2TA) association problem. The standard procedure for such problems consists of keeping only the most likely hypothesis, but the extra information carried by other hypotheses remains unused. The proposed combination method allows for the extraction of this information in an efficient way, improving over a similar method [5], providing system tracks that account for the correlation ambiguity. This method will prove useful when there is track contention (correlation ambiguity), and the information carried by the best hypothesis alone renders optimistic estimates. As a result of using this method, both better estimates (fused system tracks) are obtained and an estimate of the difficulty of the association problem is obtained based on the aggregation of neighboring tracks. In this work we consider two applications, one consisting of a T2T fusion (T2TF) and a dynamic tracking problem where measurement-to-track association (M2TA) hypotheses from a multiple hypothesis tracker (MHT) are combined. The comparison of results from the proposed procedure vs. the standard approach indicate that the latter can be improved upon in scenarios with significant association ambiguities.