Abstract for: Oscillations in Organizational Attention: Resource Allocation Dynamics Under Multiple Objectives

Organizations pursuing multiple objectives simultaneously tend to oscillate, repeatedly shifting resources toward whichever goal is currently most below aspiration, never sustainably satisfying either. This paper explains why such cycling is not a symptom of poor management but an emergent structural property of aspiration-based allocation under delay. The paper shows how goal interdependence determines whether oscillation stabilizes, intensifies, or collapses, providing a formal feedback-based explanation of this phenomenon. A system dynamics model is developed in Vensim and executed via PySD, representing an organization allocating a shared resource inflow between two goal domains through a softmax aspiration-gap rule. A two-stage resource accumulation structure introduces realistic investment-to-performance delays. Seven simulation scenarios systematically vary the sign and magnitude of cross-goal interdependence parameters, spanning mutual antagonism, mutual synergy, and asymmetric configurations, to map the resulting performance trajectories. Oscillations in resource allocation and performance arise endogenously under goal independence, driven entirely by the interaction between the aspiration-gap rule and accumulation delays, requiring no external shocks or shifting priorities. The sign of interdependence fundamentally alters this dynamic: negative interdependence sustains and amplifies cycling, driving mutual collapse under full antagonism, while positive interdependence eliminates oscillation entirely, producing smooth convergence to above-aspiration performance across both domains simultaneously. Goal interdependency structure is one of the primary determinants of multi-objective performance trajectories. Organizations experiencing chronic underperformance despite sustained resource investment should diagnose structural cross-domain effects rather than assume insufficient allocation. Future work should extend to three goals, continuous interdependence parameter sweeps to identify oscillation thresholds, endogenous aspiration dynamics, and protective goal structures where performance is driven by rare stochastic events rather than continuous accumulation. Claude was used for editorial assistance in drafting and revising the manuscript.