On a Coupled Adoption-Opinion Framework for Competing Innovations This work was supported in part by the Wallenberg AI, Autonomous Systems and Software Program (WASP) funded by the Knut and Alice Wallenberg Foundation, and by the European Union – Next Generation EU, Mission 4, Component 1, under the
Martina Alutto, Fabrizio Dabbene, Angela Fontan, Karl H. Johansson, Chiara Ravazzi · 2026
Read On a Coupled Adoption-Opinion Framework for Competing Innovations This work was supported in part by the Wallenberg AI, Autonomous Systems and Software Program (WASP) funded by the Knut and Alice Wallenberg Foundation, and by the European Union – Next Generation EU, Mission 4, Component 1, under the by Martina Alutto, Fabrizio Dabbene, Angela Fontan, Karl H. Johansson, Chiara Ravazzi (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-16719.txt.
Introduction
Martina Alutto, Fabrizio Dabbene, Angela Fontan, Karl Johansson, and Chiara Ravazzi model a race of two new tech goods. Folks in a crowd sway each other by talk. Use of one good feeds views of that good. Views then feed fresh use in a loop. Their 2026 preprint builds a two-layer frame for this race.
Once folks adopt one good, some grow sour on it. They may switch to the rival good. The frame tracks such switches over time. It can show when a lead is safe or frail. Firms and towns could use it to time launches. It may also flag where ads or rules tip a race.
Catalog record
| CREATOR | Martina Alutto, Fabrizio Dabbene, Angela Fontan, Karl H. Johansson, Chiara Ravazzi |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 35 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2601-16719
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
On a Coupled Adoption-Opinion Framework for Competing Innovations This work was supported in part by the Wallenberg AI, Autonomous Systems and Software Program (WASP) funded by the Knut and Alice Wallenberg Foundation, and by the European Union – Next Generation EU, Mission 4, Component 1, under the PRIN project TECHIE: A control and network-based approach for fostering the adoption of new technologies in the ecological transition , Cod. 2022KPHA24, CUP Master: D53D23001320006, CUP: B53D23002760006. Martina Alutto, Fabrizio Dabbene, Angela Fontan, Karl H. Johansson, Chiara Ravazzi Abstract In this paper, we propose a two-layer adoption-opinion model to study the diffusion of two competing technologies within a population whose opinions evolve under social influence and adoption-driven feedback. After adopting one technology, individuals may become dissatisfied and switch to the alternative. We prove the existence and uniqueness of the adoption-diffused equilibrium, showing that both technologies coexist and that neither partial-adoption nor monopoly can arise. Numerical simulations show that while opinions shape the equilibrium adoption levels, the relative market share between the two technologies depends solely on their user-experience. As a consequence, interventions that symmetrically boost opinions or adoption can disproportionately favor the higher-quality technology, illustrating how symmetric control actions may generate asymmetric outcomes. keywords Social networks and opinion dynamics 1 Introduction Innovation diffusion processes, describing the spread of new ideas, technologies, behaviors or practices within a social system, have been extensively investigated across different disciplines ( 26 ) . Numerous studies examine how individuals adopt innovations under the influence of both external exposure and peer influence ( 8 ; 28 ) . Classical economic approaches, such as the Bass model ( 5 ) , formalize adoption as the combined effect of independent decisions and imitation. Threshold-based models show that widespread adoption can occur once a critical mass of adopters is reached ( 16 ) . More generally, network-based frameworks, such as the linear threshold and independent cascade models ( 18 ) , describe the spread of innovation as a contagion process, in which individuals change their behavior when the influence of their social contacts exceeds a tolerance threshold. Overall, these perspectives highlight that the diffusion of innovation is a form of complex contagion, characterized by nonlinear adoption responses and strongly influenced by both social dynamics and external factors ( 9 ) . Epidemic-inspired compartmental models have emerged as an established paradigm to study diffusion dynamics. Initially developed for infectious diseases ( 19 ) , SIR and SIS models have been adapted to represent the propagation of behaviors, technologies, and information ( 13 ; 6 ; 25 ) . Variants including mechanisms of reversible adoption or dissatisfaction, such as the SIRS framework ( 20 ; 33 ) , provide a richer representation. It is worth noting that adoption is also strongly influenced by beliefs, attitudes, and perceived social norms. Opinion dynamics models provide a complementary perspective to capture these effects. The DeGroot model ( 11 ) explains consensus formation through iterative averaging of neighbors’ views, whereas the Altafini’s model ( 1 ) allows for antagonistic relationships and persistent disagreement. The Friedkin–Johnsen model ( 12 ) further incorporates individual predispositions, highlighting how resistance and heterogeneity naturally emerge in social decision processes. Building on this, a growing body of work integrates epidemic-like adoption dynamics with opinion evolution, acknowledging their mutual reinforcement: opinions shape willingness to adopt, while adoption feeds back into beliefs ( 15 ; 29 ; 21 ; 27 ; 7 ; 31 ) . These coupled frameworks reveal complex outcomes that traditional contagion-only models cannot capture. Recent contributions also address how to effectively intervene in such systems. 3 propose a predictive control framework where different policies, shaping opinions, adoption enhancement, and dissatisfaction reduction, are compared, and show that adaptive strategies outperform the static ones. Then, 2 extend this modeling to data-driven heterogeneous populations, embedding demographic factors such as mobility and age to better reflect real-world variation in influence and propensity to adopt. However, in many realistic scenarios, individuals face multiple competing technologies, making adoption a strategic decision influenced by peer coordination, as in the case of mobile service providers, car-sharing services, or streaming platforms. To capture this strategic dimension, game-theoretic models based on network coordination games have gained increasing attention ( 24 ; 32 ) , formalizing technology adoption as a process in which individuals select the option that maximizes expected utility given their neighbors’ choices. Theoretical studies have shown that a variety of outcomes may arise from such competitive dynamics. While traditional economic theory suggests convergence to a single dominant technology ( 10 ; 4 ; 30 ) , more recent works reveal that monopolistic dominance is not the only possible scenario: heterogeneity, segmentation, switching costs, and asymmetric network effects can enable long-term coexistence. Moreover, allowing dual users in complex contagion processes further broadens the parameter region supporting coexistence ( 23 ) . Similar phenomena are also observed in epidemic models of multi-virus or multi-contagion spreading ( 22 ; 14 ) . Existing models often treat competition and opinion dynamics separately. To address this gap, we propose a novel adoption-opinion model to study the diffusion of two competing technologies within a population structured as a two-layer network. Adoption evolves through epidemic-inspired compartmental dynamics with susceptible, adopter, and dissatisfied compartments for both technologies, while opinions follow a modified Friedkin–Johnsen process, integrating social influence and feedback from adoption levels. Analytical results establish the existence and uniqueness of an adoption-diffused equilibrium point, showing that at equilibrium both technologies coexist and no partial-adoption or monopoly states are possible. The rest of the paper is organized as follows. Section 2 introduces the adoption-opinion model. Stability results are provided in Section 3 , while Section 4 presents numerical simulations. Finally, Section 5 concludes the paper and outlines directions for future research. 1.1 Notation We denote by \mathbb{R} and \mathbb{R}_{+} the sets of real and nonnegative real numbers, respectively, while \mathbb{R}_{+}^{n\times n} indicates the set of real matrices with dimension n\times n and nonnegative entries. The all-1 vector and the all-0 vector are denoted by \bm{1} and \bm{0} , respectively. The identity matrix and the all-0 matrix are denoted by I and \mathbb{O} , respectively. The transpose of a matrix A is denoted by A^{T} . For x in \mathbb{R}^{n} , let ||x||_{1}=\sum_{i}|x_{i}| and ||x||_{\infty}=\max_{i}|x_{i}| be its l_{1} and l_{\infty} norms, while {\rm diag}\,(x) denotes the diagonal matrix whose diagonal coincides with x . For an irreducible matrix A in \mathbb{R}_{+}^{n\times n} , i.e., a matrix whose associated directed graph is strongly connected, we let \rho(A) denote the spectral radius of A . Inequalities between two vectors x and y in \mathbb{R}^{n} are meant to hold true entry-wise, i.e., x\leq y means that x_{i}\leq y_{i} for every i , x<y means that x_{i}<y_{i} for every i , and x\lneq y means that x_{i}\leq y_{i} for every i and x_{j}<y_{j} for some j ; analogous holds for matrices. 2 Model Description We consider a population divided into n communities, where each community represents a subpopulation of indistinguishable individuals that can interact and adopt two competing technologies, denoted by 1 and 2 . Within each community i\in\mathcal{V}=\{1,\dots,n\} , individuals are classified into the following compartments at each time instant t\geq 0 : • s_{i}(t) : the fraction of susceptibles to both technologies,