Severe obesity may be accompanied by cognitive dysfunction and NAFLD, but the associations remain unclear. We describe the prevalence and features of cognitive dysfunction and examine the associations between cognitive dysfunction and the presence and severity of NAFLD, and the associations between cognitive dysfunction and signs of other obesity-related comorbidities and neuronal damage. A cross-sectional study of patients with a body mass index of 35kg/m2 underwent evaluation for bariatric surgery. They were screened for adiposity-related comorbidity and underwent a liver biopsy and basic cognitive testing with the Continuous Reaction Time test, the Portosystemic Encephalopathy Syndrome test, and the Stroop Test. A representative subgroup also underwent the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS). The primary study outcome was "cognitive impairment," defined as ≥2 abnormal basic cognitive tests and/or an abnormal RBANS. The Triggering Receptor Expressed on Myeloid Cells 2 (TREM2) served as a biomarker for neuronal damage. We included 180 patients; 72% were women, age 46 ± 12 years, 78% had NAFLD, and 30% with NASH without cirrhosis. 8% were cognitively impaired by the basic tests and 41% by RBANS results. Most impaired were executive and short-time memory functions. There were no associations between cognitive impairment and BMI, NAFLD presence or severity, or metabolic comorbidities. Male sex (OR: 3.67, 95% CI, 1.32-10.27) and using 2 or more psychoactive medications (5.24, 95% CI, 1.34-20.4) were associated with impairment. TREM2 was not associated with cognitive impairment. Nearly half of this severely obese study cohort exhibited measurable multidomain cognitive impairment. This was not dependent on NAFLD or another adiposity comorbidity.