The hemispherical power asymmetry is regarded as one of (perhaps three) unsolved anomalies of CMB detections to date. Many explanations have been proposed, with most, from an inflationary setting, involving a single high amplitude large-scale mode. For a sufficiently large primordial bispectrum this exceptional mode can explain the asymmetry. Or so the story goes…
In this talk I will discuss an extension of previous calculations clarifying that in the multi-source case, that the key quantity is not the reduced bispectrum but rather a ‘response function’ (only agreeing with the reduced bispectrum in the single source limit). Our extension will include the effect of scale dependency which has often been ignored, but which is observed in the actual power asymmetry.
Having gone through this ‘clarification’ I will then describe why it is so difficult to construct a successful realisation, which has small enough bispectrum amplitude, but large enough power asymmetry. This will be in the form of a no-go theorem with several assumptions.
But don’t worry all is not lost. I will then describe a working example (where we break one of the above assumptions!). This model, which is not at all physical (or motivated or in any way endorsed) will be shown to satisfy all observational constraints.