We consider an extension of Higgs inflation in which the Higgs field is coupled to the Gauss-Bonnet term.Working solely in the Jordan frame, we firstly recover the standard predictions of Higgs inflation without a Gauss-Bonnet term. We then calculate the power spectra for scalar and tensor perturbations in the presence of a coupling to a Gauss-Bonnet term. We show that generically the predictions of Higgs inflation are robust and the contributions to the power spectra coming from the Gauss-Bonnet term are negligible. We find, however, that the end of inflation can be strongly modified and that we hence expect the details of (p)reheating to be significantly altered, leading to some concerns over the feasibility of the model which require further investigations.

We investigate the feasibility of models of inflation with a large Gauss-Bonnet coupling at late times, which have been shown to modify and prevent the end of inflation. Despite the potential of Gauss-Bonnet models in predicting favorable power spectra, capable of greatly lowering the tensor-to-scalar ratio compared to now-disfavored models of standard chaotic inflation, it is important to also understand in what context it is possible for postinflationary (p)reheating to proceed and hence recover an acceptable late-time cosmology. We argue that in the previously studied inverse power law coupling case, reheating cannot happen due to a lack of oscillatory solutions for the inflaton, and that neither instant preheating nor gravitational particle production would avoid this problem due to the persistence of the inflaton's energy density, even if it were to partially decay. Hence we proceed to define a minimal generalization of the model which can permit perturbative reheating and study the consequences of this, including heavily modified dynamics during reheating and predictions of the power spectra.

In single field slow-roll inflation, one expects that the spectral index n s − 1 is first order in slow-roll parameters. Similarly, its running α s ¼ dn s =d log k and the running of the running β s ¼ dα s =d log k are second and third order and therefore expected to be progressively smaller, and usually negative. Hence, such models of inflation are in considerable tension with a recent analysis hinting that β s may actually be positive, and larger than α s . Motivated by this, in this work we ask the question of what kinds of inflationary models may be useful in achieving such a hierarchy of runnings, particularly focusing on twofield models of inflation in which the late-time transfer of power from isocurvature to curvature modes allows for a much more diverse range of phenomenology. We calculate the runnings due to this effect and briefly apply our results to assess the feasibility of finding jβ s j ≳ jα s j in some specific models. DOI: 10.1103/PhysRevD.94.021301 Constraining models of inflation is one of the most important goals of cosmology. By constraining or even ruling out models of inflation, cosmologists learn a great deal about model building in theories beyond the standard model. Even with the latest cosmological observations [1,2], there is still a plethora of inflationary models compatible with data [3]. It was recently pointed out that observations of the cosmic microwave background (CMB) radiation are consistent with a rather large running of the running of the spectral index, 1 [4,5]. The constraints on α s and β s given in [5] are α s ¼ 0.011 AE 0.010 and β s ¼ 0.027 AE 0.013 (fixing the pivot scale at k ¼ 0.05 Mpc −1 ), which implies β s > 0 at the 2σ confidence level and hints that the running of the running may be larger than the running itself. The running α s also appears to be positive (although less significantly), which leads to a slight tension with a wide range of inflationary models that predict a negative running [6]. Future CMB experiments are needed to determine these parameters more precisely, but given this first hint of what is potentially a powerful piece of evidence in early universe cosmology, it is interesting and worthwhile to consider the theoretical viability for such a hierarchy of runnings to be realized in inflation.In standard single field slow-roll models, the running is of second order in slow-roll parameters and the running of the running is third order [see Eqs. (5)- (7) below]. Thus, in such models of inflation, one would quite generally expect β s to be smaller than α s . These observational hints motivate the current work, in which we study predictions of α s and β s in single and two-field inflationary scenarios with the intention of understanding what kinds of inflationary models could be consistent with such a hierarchy of runnings.While almost all investigations of inflationary models make predictions for the spectral index, relatively few study the running [6][7][8][9], and almost none discuss the running of the running [4,10]. This is largely underst...

Abstract. A disformal coupling between two scalar fields is considered in the context of cosmological inflation. The coupling introduces novel derivative interactions mixing the kinetic terms of the fields but without introducing superluminal or unstable propagation of the two scalar fluctuation modes. Though the typical effect of the disformal coupling is to inhibit one of the fields from inflating the universe, the energy density of the other field can drive viable near Sitter -inflation in the presence of nontrivial disformal dynamics, in particular when one assumes exponential instead of power-law form for the couplings. The linear perturbation equations are written for the two-field system, its canonical degrees of freedom are quantised, their spectra are derived and the inflationary predictions are reported for numerically solved exponential models. A generic prediction is low tensor-to-scalar ratio.

We consider a theory of modified gravity possessing d extra spatial dimensions with a maximally symmetric metric and a scale factor, whose (4 + d)-dimensional gravitational action contains terms proportional to quadratic curvature scalars. Constructing the 4D effective field theory by dimensional reduction, we find that a special case of our action where the additional terms appear in the well-known Gauss-Bonnet combination is of special interest as it uniquely produces a Horndeski scalar-tensor theory in the 4D effective action. We further consider the possibility of achieving stabilised extra dimensions in this scenario, as a function of the number and curvature of extra dimensions, as well as the strength of the Gauss-Bonnet coupling. Further questions that remain to be answered such as the influence of matter-coupling are briefly discussed. arXiv:1809.00920v1 [gr-qc]