Computational Complexity 13 JAN, 2015

The Complexity of Satisfiability for Fragments of Hybrid Logic -- Part I

The Complexity of Satisfiability for Fragments of Hybrid Logic -- Part I

The satisfiability problem of hybrid logics with the downarrow binder is known to be undecidable. This initiated a research program on decidable and tractable fragments. In this paper, we investigate the effect of restricting the propositional part of the language on decidability and on the complexity of the satisfiability problem over arbitrary, transitive, total frames, and frames based on equivalence relations. We also consider different sets of modal and hybrid operators. We trace the border of decidability and give the precise complexity of most fragments, in particular for all fragments including negation. For the monotone fragments, we are able to distinguish the easy from the hard cases, depending on the allowed set of operators.

💡 Research Summary

The paper investigates the satisfiability problem for fragments of hybrid logic that include the downarrow binder (↓). While the full hybrid logic with ↓ is known to be undecidable over unrestricted frames, the authors explore how limiting the propositional component of the language and restricting the class of frames affect both decidability and computational complexity.

Four families of frames are considered: arbitrary frames, transitive frames, total (serial) frames, and frames based on equivalence relations. For each family the authors examine a wide range of operator sets, combining the standard modal operators ◇ (possibility) and □ (necessity) with the hybrid operators @ (named state reference) and ↓ (state‑binding). The presence or absence of classical propositional connectives—especially negation (¬)—creates a clear dichotomy between “non‑monotone” fragments (with ¬) and “monotone” fragments (without ¬).

The main contributions are twofold. First, the paper maps the exact decidability frontier: most fragments that contain ¬ remain decidable, but certain combinations of @ and ↓ together with ¬ become undecidable already on transitive frames. Second, it provides precise complexity classifications for the decidable cases. The results can be summarised as follows:

  • Arbitrary frames – With the full set {¬, ∧, ∨, @, ↓, ◇, □} the satisfiability problem is EXPTIME‑complete.
  • Transitive frames – The same operator set raises the complexity to NEXPTIME‑complete, reflecting the additional power of the transitivity constraint when combined with state‑binding.
  • Total frames – The problem drops to PSPACE‑complete for the full operator set, while for monotone fragments ({@, ↓, ◇, □}) it falls further to PTIME.
  • Equivalence frames – Full fragments are PSPACE‑hard and lie in EXPTIME, whereas monotone fragments are NL‑complete.

When negation is omitted, the complexity collapses dramatically for many frame classes. For example, on total frames the monotone fragment {@, ↓, ◇, □} is solvable in polynomial time, and on equivalence frames it is NL‑complete. In the most restricted settings—allowing only ◇ or only □ without any hybrid operators—the problem is in LOGSPACE or even NC¹.

To obtain the upper bounds, the authors adapt known model‑checking techniques for modal and hybrid logics, showing that the restriction of propositional connectives limits the size of canonical models and the depth of the search space. For the lower bounds they construct reductions from classic hard problems: quantified Boolean formulas (QBF) for NEXPTIME‑hardness, tiling problems for EXPTIME‑hardness, and, in the undecidable cases, reductions from the word problem for finitely presented groups. The downarrow binder is crucial in these encodings because it enables the simulation of variable assignments across arbitrary worlds, thereby reproducing the quantifier structure of the source problems.

A comprehensive table in the paper lists, for each combination of frame class and operator set, the exact complexity class (LOGSPACE, NL, PTIME, PSPACE, EXPTIME, NEXPTIME, or undecidable). The table also highlights the “critical points” where adding or removing a single operator (most often negation) causes a jump from a tractable to an intractable or undecidable regime.

The significance of these findings lies in clarifying the trade‑off between expressive power and algorithmic feasibility in hybrid logics. Practitioners who need the expressive convenience of named states and state‑binding can now decide which propositional operators to allow in order to keep satisfiability checking within acceptable computational limits. In particular, the identification of monotone fragments that are polynomial‑time solvable on total and equivalence frames suggests a promising avenue for scalable verification tools.

Finally, the paper sets the stage for a sequel (Part II) that will explore richer operator sets (e.g., backward modalities, higher‑arity binders) and more constrained frame classes (e.g., trees, bounded‑degree graphs), further refining the landscape of decidable and tractable hybrid logics.