Binge Eating Disorder (BED) is thought of as a disorder of cognitive control but evidence regarding its neurocognitive mechanisms is inconclusive. A key limitation in prior research is the lack of clear separation between effects of BED and obesity. Moreover, research has largely disregarded self-report evidence that neurocognitive deficits may emerge primarily in contexts focused on avoiding aversive states. Addressing these gaps, this longitudinal study investigated behavioral flexibility and its underlying neuro-computational processes in approach and avoidance contexts in normal weight individuals, obese individuals, and obese individuals with BED. Participants performed a probabilistic reversal learning task during functional imaging, with different blocks focused on obtaining wins or avoiding losses. They were reinvited for a 6-months follow-up. Analyses were informed by computational models of reinforcement learning (RL). Compared to obese individuals without BED, BED participants had relatively more difficulties to avoid losses, while obese participants without BED had relatively more difficulties to obtain rewards. Computationally, this was explained by differential learning sensitivities in the win vs loss conditions between groups. In the brain, this was mirrored in differential neural learning signals in the ventromedial prefrontal cortex (vmPFC) per condition. The condition-specific behavioral and neural differences were subtle, but strikingly scaled with BED symptoms between and within subjects. Compared to normal weight controls, obese participants without BED switched more between choice options. This was reflected in diminished representation of choice certainty in the vmPFC. Hence, RL alterations in obesity with and without BED may be qualitatively different: our finding that BED was associated with relative RL deficits in loss-avoidance contexts aligns with the notion that in BED, neurocognitive alterations may selectively emerge in aversive states. Our study thus highlights the importance of distinguishing between obesity with and without BED to identify unique neuro-computational alterations underlying different styles of maladaptive eating behavior.