We study non-gaussianity effects, using the $\delta N$ formalism, in a multi-field inflationary model consisting of K\"ahler moduli derived from type IIB string compactification in the large volume limit. The analytical work in this paper mostly follows the separable potential method developed by Vernizzi and Wands. The numerical analysis is then used in computing non-gaussianity beyond slow-roll regime. The possibility of the curvaton scenario is also discussed. We give the condition for the existence of the curvaton and calculate the non-guassianity generated by the curvaton decay in the large volume limit.

We show that cosmological observables can constrain the topology of the compact additional dimensions predicted by string theory. To do this, we develop a general strategy for relating cosmological observables to the microscopic parameters of the potentials and field-dependent kinetic terms of the multiple scalar fields that arise in the low-energy limit of string theory. We apply this formalism to the Large Volume Scenarios in Type IIB flux compactifications where analytical calculations are possible. Our methods generalize to other settings.

A four-dimensional universe, arising from a flux compactification of Type IIB string theory, contains scalar fields with a potential determined by topological and geometric parameters of the internal -hidden- dimensions. We show that inflation can be realized via rolling towards the large internal volume minima that are generic in these scenarios, and we give explicit formulae relating the microscopic parameters (e.g., the Euler number of the internal space) to the cosmological observables (e.g., the spectral index). We find that the tensor-to-scalar ratio, the running of the spectral index, and the potential energy density at the minimum are related by consistency relations and are exponentially small in the number of e-foldings. Further, requiring that these models arise as low-energy limits of string theory eliminates most of them, even if they are phenomenologically valid. In this context, this approach provides a strategy for systematically falsifying stringy inflation models.