The Iowa Gambling Task (IGT; Bechara et al. 1994) is commonly associated with research on decision making under uncertainty. It is a task, developed within the framework of the Somatic Marker Hypothesis (HMS), which simulates decision-making in real life. Initial results show that patients with ventromedial cortex (VM) damage have poor performance on IGT despite retaining their reasoning abilities. This leads to the hypothesis that such patients have a deficiency in the somatic affective signal (somatic marker) that is critical to guide their behavior advantageously, regardless of conscious awareness of the task. Recent research questions the validity and reliability of the IGT on the subsequent analysis of the evidence in healthy participants, arguing that 1) they do not exhibit a predominant choice of advantageous options and 2) their behavior is widely heterogeneous. However, this analysis is performed outside the context of comparison with MV patients. Around this debate, the objective of this work has been to carry out a systematic review on the performance of VM patients in IGT. Our results show a significant difference between the performance of healthy subjects and VM, highlighting the advantageous behavior of healthy subjects compared to the pathological group. In effect, we question the interpretation of the results of healthy participants and the consequent criticisms of HMS.