## Abstract ## Many social interactions are characterised by dynamic interplay, such that individuals exert reciprocal influence over each other's behaviours and opinions. The present study investigated how the dynamics of reciprocal influence affect decisions made in a social context, over and above the information communicated in an interaction. To this end, we developed a simple social decision-making paradigm in which two people are asked to make perceptual judgments while receiving information about each other's decisions. In a Static condition, information about the partner was static, only conveying their initial, independent judgment. However, in a Dynamic condition, each individual saw the evolving opinion of their partner as they learnt about and responded to the individual's own judgment. The results indicated that in both conditions, the majority of confidence adjustments followed a step function characterised by an abrupt change followed by smaller adjustments around an equilibrium, and that participants' confidence was used to arbitrate conflict (although deviating from Bayesian norm). Crucially, interaction had systematic effects on opinion change relative to the Static baseline, magnifying confidence change when partners agreed and reducing confidence change when they disagreed. These findings indicate that dynamic interactions are not only affected by task-relevant information but often follow a non-linear self-reinforcing pattern. ## Experiments ## Experiment 1 (m0) is the main experiment reported in the manuscript. Experiment 2 (m1) replicates the same results, using a confidence scale rather than a post-decision wagering scale, thus addressing possible confounding due to economic traits like loss and risk aversion. Experiment 2 (m2) replicates the results of Experiment 1 by adding reminders for self and partner's initial confidence, thus addressing the possible alternative explanation that results might be due to participants forgetting their initial judgment. ## Analysis ## Figures and analyses reported in the paper are in the "across_experiments_multivariate" folder. The pipeline.m script will call all the relevant functions.