artificial neural network evaluating nonlinear evolutionary effect of cognitive and risk complexity in ambidextrous strategy renewal

This research examines the micro-foundations of individual ambidexterity as a managerial dynamic capability for addressing the tension associated with pursuing dual strategies. it empirically investigates the coevolutionary intertwining effects of different levels of integrative complexity (ic) and business complexity to identify the optimal composition of cognition–risk interaction that enhances ambidexterity performance. a multidisciplinary approach was adopted to address dualities and ambidexterity through a managerial cognition lens. the application of artificial neural networks has helped to gain a deeper understanding of the role of the cognitive dimension related to dualities. the results clarify that risk and business complexity can have both negative and positive effects. specifically, in low-risk settings, high levels of ic are not capable of inducing higher ambidexterity performance, suggesting that risk should stimulate the cognitive complexity of ceos. moreover, the findings reveal that a high degree of cognitive complexity cannot be viewed as the sole dominant factor in successful strategy renewal, as it is exogenously fueled. an investigation of different scenarios indicates that, at higher levels of risk, differentiation-dominant ic, encompassing a perceptual decision-making style can enhance ambidexterity despite limited integration. in contrast, in low-risk settings, integration supported by analogical thinking may be more effective.