How does the relativistic extension of MOND avoid the ad hoc interpolating function problem?

# Relativistic MOND and the Interpolating Function Problem

The "ad hoc interpolating function" problem is this: in the original MOND, you need an arbitrary mathematical function μ(a/a₀) that smoothly connects Newtonian gravity (high acceleration) to the deep-MOND regime (low acceleration). Where does this function come from? It looks invented to fit observations, not derived from first principles.

Recent relativistic extensions tackle this differently by grounding the interpolation in deeper physics rather than phenomenological guessing.

The Gauge-Theory Approach

The most direct answer comes from the E₆ × E₆ framework [AG-2026.01-159]. Rather than postulating an interpolating function, gravity itself emerges from a Yang–Mills gauge structure via the Plebanski/MacDowell–Mansouri mechanism. MOND is then implemented through an infrared (IR) metric deformation that is UV-vanishing (so general relativity is automatically recovered at high accelerations) while the deep-MOND regime is *fixed by a symmetry principle*: conformal invariance with the de Sitter group SO(4,1) in three spatial dimensions. The acceleration scale a₀ itself is selected dynamically from the IR regime, not plugged in by hand. This avoids the ad hocness because the form of the transition emerges from gauge-theoretic structure, not from curve-fitting.

Nonlocal Gravity

An alternative avenue uses nonlocal modifications of Einstein's equations derived from quantum gravitational corrections. A single nonlocal model can interpolate between cosmology and bound systems [AG-2025.12-255], avoiding separate ad hoc functions for each regime. The physics comes from persistent quantum stress-tensor effects from primordial inflation.

Entropic Gravity Route

A third path derives relativistic MOND from entropic gravity with thermal corrections [AG-2025.11-149]. Here the interpolation emerges naturally from temperature-dependent modifications to the equipartition law on a holographic screen, combined with the Unruh relation. Again, no separate ad hoc function is needed; the MOND-like behavior flows from the underlying thermodynamic structure.

The Remaining Tension

That said, not all relativistic MOND proposals succeed equally. Versions using Lorentz-violating vector fields struggle: their dynamic nature makes the transition to MOND difficult outside the stationary limit, so cosmological structure growth—a key test—doesn't match MOND's prediction [AG-2025.03-808]. This suggests that the *choice of relativistic completion matters*; avoiding ad hocness in one place can introduce it elsewhere if the underlying framework is not sufficiently constrained.

The unifying theme is that the most promising approaches derive the interpolating behavior from symmetry or from deeper quantum/thermodynamic principles, rather than imposing it phenomenologically.