Navigating the approach-avoidance matrix: A novel foraging task in virtual reality

Adaptive fear responses help individuals avoid danger, but excessive avoidance can hinder learning and contribute to persistent anxiety. To investigate the dynamics of approach–avoidance (AA) conflict under threat, we developed and validated a novel immersive virtual reality (VR) paradigm that integrates gamified reward-punishment contingencies into a naturalistic foraging task. In a multimethod approach a total of 242 participants navigated three distinct VR environments (water, forest, desert), each associated with different probabilities of experiencing an aversive electric stimulation and varying reward potentials. Participants were required to collect sufficient virtual energy to ensure survival, balancing risk and reward over 24 trials. Behavioral results revealed a consistent “better safe than sorry” strategy, with increased time spent and faster transitions toward the safest, low-reward context. Self-reported motivations (e.g., reward-seeking vs. threat-avoidance) aligned with behavioral choices, while standard BIS/BAS trait measures failed to predict AA behavior. Electrodermal activity mirrored behavioral patterns during pre-exploration, but only the water context showed a significant post-task decline in arousal, possibly reflecting habituation or novelty effects. Exploratory post-task behavior suggested increased comfort in moderate-risk environments, supporting potential applications in exposure-based interventions. This paradigm offers a flexible and ecologically valid tool for studying AA conflict, with potential for identifying individual differences in motivation and affective regulation. Future research may leverage this approach for clinical profiling, real-time assessment of avoidance strategies, and testing adaptive interventions in anxiety and related disorders.