Fixed point indices and fixed words at infinity of selfmaps of graphs

Abstract Indices of fixed point classes play a central role in Nielsen fixed point theory. Jiang-Wang-Zhang proved that for selfmaps of graphs and surfaces, the index of any fixed point class has an upper bound called its characteristic. In this paper, we study the difference between the index and the characteristic for selfmaps of graphs. First, for free groups, we extend the notion of attracting fixed words at infinity of automorphisms into that of injective endomorphisms. Then, by using a relative train track technique, we show that the difference mentioned above is quite likely to be the number of equivalence classes of attracting fixed words of the endomorphism induced on the fundamental group. This gives a new algebraic approach to estimating indices of fixed point classes of graph selfmaps. As a consequence, we obtain an upper bound for attracting fixed words of injective endomorphisms of free groups, generalizing the one for automorphisms due to Gaboriau-Jaeger-Levitt-Lustig. Furthermore, we give a simple approach to roughly detecting whether fixed words exist or not.