Background: Bariatric surgery has been widely recognized as the most efficient long-term treatment method in severe obesity, yet therapy success shows considerable interindividual variability. Postoperative metabolic adaptations, including improved gut hormone secretion (GLP-1, PYY and ghrelin), and restored executive function may play an explanatory role in weight loss, yet causes for poor success in individual patients remain unknown. This study investigates gut-hormonal and cognitive characteristics in extreme weight loss responders to bariatric surgery. Methods: Patients (n=47) with high or low excessive weight loss (EWL) at least 2 years after Roux-en-Y-gastric bypass or sleeve gastrectomy were allocated into good responders (GR, EWL 82.4 +/- 11.6%) and poor responders (PR, EWL 24.0 +/- SD 12.8%) to study differences in postprandial secretion of GLP-1, PYY, ghrelin and in working memory (WM). Results: Mean BMI was 47.1 +/- 6.2 kg/m^2 in poor responders (n=21) and 28.9 +/- 3.1 kg/m^2 in good responders (n=26, p < 0.001). Fasted GLP-1 and PYY were comparable for GR and PR (p > 0.2) and increased strongly after a standardized test meal (300 kcal liquid meal) with a peak at 15 to 30 minutes. The increase was stronger in GR compared to PR (GLP-1, PYY: Time x Group p < 0.05). Plasma ghrelin levels already differed between groups at fasted state, showing significantly higher levels for GR (p < 0.05). Postprandially, ghrelin secretion was suppressed in both groups, but suppression was higher in GR (Time x Group p < 0.05). GR showed significantly higher WM scores than PR (p < 0.05). Postprandial ghrelin (iAUC), but not GLP-1 or PYY release, significantly mediated the relationship between EWL and a WM subscore (IS score, CI = 0.07 - 1.68), but not WM main score (MIS score, CI = -0.07 - 1.54). Conclusion: Excess weight loss success after bariatric surgical procedures is associated with distinct profiles of gut-hormones at fasted and postprandial state, and differences in working memory. Working memory performance was partly mediated by postprandial incremental drops in ghrelin. Future studies need to integrate longitudinal data, larger samples and more sensitive cognitive tests.