OBJECTIVE: Disadvantageous decision-making (mainly measured by the Iowa Gambling Task) has been demonstrated in patients with suicidal behavior compared to controls. We, therefore, aimed at clarifying the qualitative and quantitative relationship between decision-making and the risk of suicidal behavior in unipolar and bipolar disorders respectively, as well as establishing the strength of this relationship. METHODS: (1) We conducted a cross-sectional study comparing IGT performances between 141 unipolar suicide attempters and 57 bipolar suicide attempters. (2) We conducted a systematic review and a meta-analysis of studies comparing IGT performances in patients with vs. without a history of suicidal acts in bipolar and unipolar disorder, together and separately. RESULTS: (1) Among suicide attempters, bipolar and unipolar groups performed similarly (t(195)=-0.7; p=0.48). Unipolar non-attempters performed better IGT than unipolar suicide attempters (t(221)=3.1; p=0.002), only in female gender, whereas performances were similar in bipolar patients whatever the history of suicide attempt (t(77)=-0.3; p=0.7). (2) A meta-analysis of 10 studies confirmed significantly impaired decision-making with a moderate effect-size (-0.38 (95% CI[-0.61--0.16]; z=-3.3; p=0.001) in unipolar disorder and (g=-0.4 (95% CI[-0.75 to -0.05]; z=-2.2; p<0.026) in bipolar disorder suicide attempters compared to unipolar and bipolar non-attempters, respectively. LIMITATIONS: It was not possible to analyse according to the level of lethality attempt. CONCLUSION: Overall, a strong significant association was found between decision-making and the risk of suicidal behavior in unipolar disorder and bipolar disorder. However, further neuropsychological studies need to analyse separately unipolar and bipolar disorder and to study gender differences.