Hair cortisol concentrations (hairF) are considered to be relatively robust to various confounding influences. However, a potentially important covariate factor that has received little attention in this context is hair exposure to ultraviolet/sunlight radiation. We conducted a detailed experimental investigation to examine the effects of light exposure on hair cortisol. In study I, a hydrocortisone-containing solution was subjected to short-term artificial light irradiation for 1, 3, 5, 10, 15, or 30min to evaluate the stability of cortisol molecules due to radiant energy. In study II, hair samples (N=12) were subjected to single short-term artificial light irradiation for 0, 1, or 5h to examine light-induced effects in the hair matrix. In study III, hair samples (N=25) were subjected to long-term naturalistic sunlight radiation over a period of two months (during summer) with daily exposure times of 0, 1, 3, or 6h, respectively. Besides cortisol, studies II & III also examined concentrations of cortisone (hairE), dehydroepiandrosterone (hairDHEA) and progesterone (hairP) in hair, quantified using LC-MS/MS technology. Results across the three studies consistently revealed effects of light irradiation on hair steroid concentrations: Longer light exposure resulted in a decrease of dissolved hydrocortisone (study I) as well as of hairF and hairE (studies II and III). Conversely, hairDHEA and hairP increased with longer natural sunlight exposure times (study III), while this effect was not observed for short-term artificial light irradiation (study II). Combined, our findings imply sunlight exposure as a potential confound in hair steroid research. Given the experimental character of this investigation, the magnitude of this effect under real-life testing conditions is difficult to estimate. To support future investigation into this, we designed a ‘sunlight-exposure’ questionnaire to share with the research community. The assessment and statistical accounting for sunlight exposure-related effects in future hair steroid research (using this or a similar questionnaire) may help to reduce the potential influence of this unwanted error source and could thus lead to more valid and reliable results. In addition, our data strongly suggest that hair samples for steroid analyses need to be stored in a dark environment.