---
title: Cyclical Contraction Mapping Theory
url: https://www.emergentmind.com/topics/cyclical-contraction-mapping
type: topic
---

# Cyclical Contraction Mapping Theory

A cyclical contraction mapping is a generalization of Banach’s contraction principle to mappings that cycle through a family of subsets of a metric or generalized metric space, often under nonstandard contractivity or pseudocontractivity conditions. These mappings and their fixed point or best-proximity point theory form a rapidly developing branch of nonlinear analysis, with deep connections to approximation algorithms, best-approximation theory, and the study of nonlinear oscillatory or cyclic phenomena.

## 1. Definition and Formulations

A **cyclical contraction mapping** typically acts on a union of closed subsets $\{A_1, \ldots, A_p\}$ of a metric space $(X,d)$, with a mapping $T:\bigcup_{i=1}^pA_i \to \bigcup_{i=1}^pA_i$ satisfying $T(A_i) \subset A_{i+1}$ (indices mod $p$). The contraction condition is imposed not globally but *between* elements of adjacent cyclical blocks. In generalizations, $T$ can “linger” in a given block for finitely many steps or skip more than one block, as in $r$-cyclic operators and $p$-cyclic self-mappings [1208.1239, 1208.0758, 2602.16226, 2208.02347].

Contractivity is imposed via various possible conditions:
- **Classical cyclic contraction:** $d(Tx,Ty) \leq k\, d(x,y)$, $k\in(0,1)$ for $x\in A_i$, $y \in A_{i+1}$
- **Kannan-type cyclic contraction:** $d(Tx,Ty) \leq \frac{\lambda}{2}\big[d(x,Tx) + d(y,Ty)\big]$
- **Intermediate-sense (pseudo)contractions:** Quadratic forms bounding $d^2(T^n x, T^n y)$ with iteration-dependent parameters and error terms
- **$p$–$\varphi$–cyclic contraction:** $d_p(Tx_1, \ldots, Tx_m; Ty_1, \ldots, Ty_m) \leq d_p(x_1, \ldots, x_m; y_1, \ldots, y_m) - \varphi\big(d_p(x_1, \ldots, x_m; y_1, \ldots, y_m)\big) + \varphi\big(d_p(A_1, \ldots, A_m)\big)$, with $d_p$ an $\ell^p$-type metric and $\varphi$ strictly increasing [2602.16226].

Cyclical contraction notions also generalize to partial metric spaces [1112.5891] and G-metric spaces [2006.15076], where the contraction is expressed via the partial or G-metric.

## 2. Main Fixed Point and Best-Proximity Theorems

### Standard Results
If the intersection $\bigcap_{i=1}^pA_i$ is nonempty, cyclical contractions often admit a unique fixed point in the intersection, with Picard iterates $T^n x$ converging to it for all $x$. If the intersection is empty, “best-proximity” points—elements $z_i\in A_i$ such that $d(z_i, Tz_i) = \mathrm{dist}(A_i, A_{i+1})$—replace classical fixed points [1208.1239, 1208.0758, 2602.16226].

#### Representative Theorems

| Context              | Existence/Uniqueness Result              | Reference     |
|----------------------|------------------------------------------|---------------|
| Complete metric spaces with cyclical contraction | Unique fixed point in $\bigcap A_i$; Picard convergence | [2602.16226, 1301.5050, 2208.02347] |
| Uniformly convex Banach spaces | Convergence to unique finite limiting cycle of best-proximity points; collapse to fixed point if all $A_i$ intersect | [1208.1239, 1208.0758] |
| Partial metric/G-metric spaces | Existence/theory of approximate fixed points, or unique solution in intersection if contractivity is strict | [1112.5891, 2006.15076] |

Proof schemes typically combine boundedness and Cauchy properties of the (possibly interleaved) Picard sequences, closedness/convexity of the subsets, and analysis of asymptotic regularity or nonexpansiveness from the contractive inequalities, to ensure convergence.

## 3. Advanced Cyclical Contractive and Pseudocontractive Conditions

### Intermediate-Sense Cyclic Pseudocontraction
The inequalities studied in [1208.1239, 1208.0758] involve iteration-dependent parameters and error terms:
\[
d^2(T^n x, T^n y) \leq a_n(x,y)\,d^2(x,y) + B_n(x,y)\big[d^2(x,y) + d^2(T^n x, T^n y)\big] + 2u_n(x,y)B_n(x,y)d(x,y)d(T^n x, T^n y)+ E_n(x,y) + \gamma_n(x,y) D^2
\]
with conditions on limit behaviors of $a_n$, $B_n$, $u_n$, $E_n$, $\gamma_n$. This encompasses asymptotically strictly pseudocontractive, strictly contractive, and Meir–Keeler-type conditions.

### Cyclic $p$–$\varphi$–Contraction
This unifies $\ell^p$-averaged distance control and non-linear contractivity (through function $\varphi$). For example, with $\varphi(t) = \lambda t$, $0<\lambda<1$, 
\[
d_p(Tx_1,\dots,Tx_m; Ty_1,\dots,Ty_m)\leq \lambda\,d_p(x_1,\dots,x_m; y_1,\dots,y_m) + (1-\lambda)\,d_p(A_1,\dots, A_m)
\]
This encompasses the Eldred–Veeramani best-proximity cyclic contraction as a special case [2602.16226].

### Synchronous and Asynchronous $r$-Cyclic Contractions
$r$-cyclic operators generalize the phase-jumping phenomenon (e.g., $T(A_i) \subset A_{i+r}$), with results for both synchronous (skipped-phase contraction) and asynchronous (contraction imposed between adjacent phases, mapping skips $r$ sets) formulations [2208.02347].

## 4. Extensions to Generalized Settings

### Partial Metric Spaces
Cyclical Banach contraction principles are extended to partial metric spaces, with suitable completeness assumptions (“0-complete” spaces) and the partial metric inequalities replacing metric ones. The fixed point lies in the intersection of the sets, with “0-self-distance” [1112.5891].

### G-Metric Spaces
Cyclical contraction mappings generalize via inequalities such as, for a G-$a$-cyclical contraction,
\[
G(Tx,Ty,Ty) + G(Ty,Tx,Tx) \leq a [G(x,y,y) + G(y,x,x)]
\]
where $G$ is the ternary metric. Main results concern the approximate fixed point property: for all $\varepsilon>0$ there exists $x$ with $G(x,Tx,Tx)+G(Tx,x,x)<\varepsilon$, even if exact fixed points do not exist [2006.15076].

## 5. Structure of Orbits, Invariant Circuits, and Convergence Phenomena

Cyclical contraction fixed-point problems reduce, via combinatorial properties of the covering and the operator (e.g., via the value $\gcd(m,r)$ in $r$-cyclic cases), to collections of classical contraction problems on invariant “circuits”; convergence behavior is dictated by this decomposition [2208.02347]. In uniformly convex or strictly convex Banach spaces, uniqueness of the limiting sequence or point is guaranteed; in more general settings one may obtain only approximate solutions or multiple fixed points corresponding to the decomposition.

Block-by-block analysis and reduction to (asymptotic) nonexpansiveness are central in the proof architectures. The general principle is that the orbit (or interleaved orbits) generated by a cyclical contraction mapping is bounded, Cauchy under suitable conditions, and converges—either to a unique point in the intersection, a best-proximity point, or, in the generalized-metric or partial metric context, to an approximate fixed point [1208.1239, 1208.0758, 1301.5050, 1112.5891, 2006.15076].

## 6. Practical Applications and Theoretical Significance

Cyclical contraction theory models and underpins convergence in applications with periodicity, alternation, or cycling constraints—including alternating projections, best-approximation algorithms, periodic boundary value problems, and cyclic phenomena in physical, economic, and engineering systems [2208.02347, 1208.0758]. The theory extends the Banach–Picard paradigm to a wide class of settings, including partial and G-metric spaces, with relaxed or non-linear control conditions.

Error-controlling terms, as in Kannan–type and Pata–type generalizations, increase the theory’s flexibility, covering mappings with small, controlled violations of strict contractivity, and unifying numerous earlier fixed point theorems [1301.5050].

## 7. Further Generalizations and Open Directions

Recent work expands the scope of cyclical contraction mapping:
- Nonlinear (e.g., $\varphi$-controlled) contraction conditions [2602.16226].
- More general covering structures (e.g., $r$-cyclic, or with possible “stalls” in blocks).
- Use in spaces lacking standard completeness or convexity (partial metric, G-metric, modular etc.).
- Approximate fixed point theory in G-metric and more exotic frameworks [2006.15076].

Open research areas include best-proximity versions in non-classical settings, coupled fixed point or multivalued operator generalizations, and the analysis of stability and convergence rates for complex cyclic dynamical systems [2208.02347].

Source: https://www.emergentmind.com/topics/cyclical-contraction-mapping