Models where dark matter and dark energy interact with each other have been proposed to solve the coincidence problem. We review the motivations underlying the need to introduce such interaction, its influence on the background dynamics and how it modifies the evolution of linear perturbations. We test models using the most recent observational data and we find that the interaction is compatible with the current astronomical and cosmological data. Finally, we describe the forthcoming data sets from current and future facilities that are being constructed or designed that will allow a clearer understanding of the physics of the dark sector.

We determine the range of parameter space of an Interacting Quintessence Model that best fits the recent WMAP measurements of Cosmic Microwave Background temperature anisotropies. We only consider cosmological models with zero spatial curvature. We show that if the quintessence scalar field decays into cold dark matter at a rate that brings the ratio of matter to dark energy constant at late times, the cosmological parameters required to fit the CMB data are: dark energy density ${\Omega }_x=0.43\pm 0.12$, baryon fraction ${\Omega }_b=0.08\pm 0.01$, slope of the matter power spectrum at large scales $n_s=0.98\pm 0.02$ and Hubble constant $H_0=56\pm 4\mathrm{km}/\mathrm{s}/\mathrm{Mpc}$. The data prefers a dark energy component with a dimensionless decay rate parameter $c^2=0.005$ and noninteracting models are consistent with the data only at the $99.9\%$ confidence level. Using the Bayesian Information Criteria we show that this extra parameter fits the data better than models with no interaction. The quintessence equation of state parameter is less constrained; i.e., the data sets an upper limit $w_x\le -0.86$ at the same level of significance. When the WMAP anisotropy data are combined with supernovae data, the density parameter of dark energy increases to ${\Omega }_x\simeq 0.68$ while $c^2$ augments to $6.3\times {10}^{-3}$. Models with quintessence decaying into dark matter provide a clean explanation for the coincidence problem and are a viable cosmological model, compatible with observations of the CMB, with testable predictions. Accurate measurements of baryon fraction and/or of matter density independent of the CMB data, would support/disprove these models.

The coincidence problem of late cosmic acceleration constitutes a serious riddle with regard to our understanding of the evolution of the Universe. Here we argue that this problem may someday be solved—or better understood—by expressing the Hubble expansion rate as a function of the ratio of densities (dark matter/dark energy) and observationally determining the said rate in terms of the redshift.