The evidence for the accelerated expansion of the Universe and the time dependence of the fine-structure constant suggests the existence of at least one scalar field with a mass of order H 0 . If such a field exists, then it is generally assumed that its coupling to matter must be tuned to unnaturally small values in order to satisfy the tests of the equivalence principle ͑EP͒. In this paper, we present an alternative explanation which allows scalar fields to evolve cosmologically while having couplings to matter of order unity. In our scenario, the mass of the fields depends on the local matter density: the interaction range is typically of order 1 mm on Earth ͑where the density is high͒ and of order 10-10 4 AU in the solar system ͑where the density is low͒. All current bounds from tests of general relativity are satisfied. Nevertheless, we predict that near-future experiments that will test gravity in space will measure an effective Newton's constant different by order unity from that on Earth, as well as EP violations stronger than currently allowed by laboratory experiments. Such outcomes would constitute a smoking gun for our scenario.

We present a novel scenario where a scalar field acquires a mass which depends on the local matter density: the field is massive on Earth, where the density is high, but is essentially free in the solar system, where the density is low. All existing tests of gravity are satisfied. We predict that near-future satellite experiments could measure an effective Newton's constant in space different from that on Earth, as well as violations of the equivalence principle stronger than currently allowed by laboratory experiments.

No abstract

After a decade and a half of research motivated by the accelerating universe, theory and experiment have a reached a certain level of maturity. The development of theoretical models beyond Λ or smooth dark energy, often called modified gravity, has led to broader insights into a path forward, and a host of observational and experimental tests have been developed. In this review we present the current state of the field and describe a framework for anticipating developments in the next decade. We identify the guiding principles for rigorous and consistent modifications of the standard model, and discuss the prospects for empirical tests.We begin by reviewing recent attempts to consistently modify Einstein gravity in the infrared, focusing on the notion that additional degrees of freedom introduced by the modification must "screen" themselves from local tests of gravity. We categorize screening mechanisms into three broad classes: mechanisms which become active in regions of high Newtonian potential, those in which first derivatives of the field become important, and those for which second derivatives of the field are important. Examples of the first class, such as f (R) gravity, employ the familiar chameleon or symmetron mechanisms, whereas examples of the last class are galileon and massive gravity theories, employing the Vainshtein mechanism. In each case, we describe the theories as effective theories and discuss prospects for completion in a more fundamental theory. We describe experimental tests of each class of theories, summarizing laboratory and solar system tests and describing in some detail astrophysical and cosmological tests. Finally, we discuss prospects for future tests which will be sensitive to different signatures of new physics in the gravitational sector.The review is structured so that those parts that are more relevant to theorists vs. observers/experimentalists are clearly indicated, in the hope that this will serve as a useful reference for both audiences, as well as helping those interested in bridging the gap between them.

We present a screening mechanism that allows a scalar field to mediate a long range (∼Mpc) force of gravitational strength in the cosmos while satisfying local tests of gravity. The mechanism hinges on local symmetry restoration in the presence of matter. In regions of sufficiently high matter density, the field is drawn towards φ = 0 where its coupling to matter vanishes and the φ → −φ symmetry is restored. In regions of low density, however, the symmetry is spontaneously broken, and the field couples to matter with gravitational strength. We predict deviations from general relativity in the solar system that are within reach of next-generation experiments, as well as astrophysically observable violations of the equivalence principle. The model can be distinguished experimentally from Brans-Dicke gravity, chameleon theories and brane-world modifications of gravity.Scalar fields are the simplest of fields. Light, gravitationally coupled scalars are generically predicted to exist by many theories of high energy physics. These scalars may play a crucial role in dark energy as quintessence fields, and generically arise in infrared-modified gravity theories [1][2][3][4][5][6][7]. Despite their apparent theoretical ubiquity, no sign of such a fundamental scalar field has ever been seen, despite many experimental tests designed to detect solar system effects or fifth forces that would naively be expected if such scalars existed [8,9].Several broad classes of theoretical mechanisms have been developed to explain why such light scalars, if they exist, may not be visible to experiments performed near the Earth. One such class, the chameleon mechanism [5,6], operates whenever the scalars are nonminimally coupled to matter in such a way that their effective mass depends on the local matter density. Deep in space, where the local mass density is low, the scalars would be light and would display their effects, but near the Earth, where experiments are performed, and where the local mass density is high, they would acquire a mass, making their effects short range and unobservable.Another such mechanism, the Vainshtein mechanism [10], operates when the scalar has derivative selfcouplings which become important near matter sources such as the Earth. The strong coupling near sources essentially cranks up the kinetic terms, which means, after canonical normalization, that the couplings to matter are weakened. Thus the scalar screens itself and becomes invisible to experiments. This mechanism is central to the phenomenological viability of brane-world modifications of gravity [1,2] and galileon scalar theories [3].In this Letter, we explore a third class of mechanisms for hiding a scalar. A similar framework was studied in [12,13] with different motivations, and some the results below overlap with these works. In this mechanism, the vacuum expectation value (VEV) of the scalar depends on the local mass density, becoming large in regions of low mass density, and small in regions of high mass density. In addition, the coupling of th...