From resisting digital distractions to regulating emotional outbursts, response inhibition is critical for daily life, yet this vital cognitive mechanism is surprisingly unstable across moments. This variability is increasingly attributed to spontaneous changes in internal states, particularly tonic arousal. While classic models like the Yerkes-Dodson law and Adaptive Gain Theory predict a static inverted U-shaped relationship between arousal and performance, recent frameworks propose that this relationship is adaptive and scales relative to current environmental baselines. The present study investigated whether spontaneous fluctuations in pupil-indexed tonic arousal exert a linear or non-linear influence on both response execution and response withholding. Forty participants between 18 and 35 completed an auditory Go/No-Go task while baseline pupil size was continuously recorded as a proxy for locus coeruleus-mediated tonic arousal. 

Mixed-effects models were utilized to compare linear versus quadratic fits for response time (Go-RT), commission errors (false alarms), and combined Inverse Efficiency Score (IES). Contrary to the predicted inverted U-shaped curve, the data demonstrated a significant linear relationship. Higher prestimulus tonic arousal was associated with faster response execution and improved overall efficiency. Likelihood ratio tests confirmed that quadratic models did not provide a significantly better fit than simpler linear models for any metric. The absence of a hyper-arousal decline in response inhibition challenges the relative arousal framework in executive functioning. Future research should investigate the existence of the relative arousal framework in response inhibition and research to what extent maximum performance can be reached in global hypo- and hyper arousal states.