Objective This study aimed to determine the incremental validity of the Minnesota Multiphasic Personality Inventory, Second Edition (MMPI‐2), relative to the Beck Depression Inventory‐Second Edition (BDI‐II) in the detection of depressive symptoms in neurological patients. An additional aim was to determine whether over‐reporting and under‐reporting as measured by the MMPI‐2 validity scales, predicted BDI‐II scores. Method The sample comprised 406 consecutive participants seen for neuropsychological assessment. The sample was divided into two subsamples for analysis, a heterogeneous neurosciences sample and a seizure disorders sample. Rates of elevated scores were compared and the prevalence of over‐reporting and under‐reporting by subsample was also examined. Results A total of 31.8% participants were classified as depressed by both the BDI‐II and the MMPI‐2 Scale 2 and 38.2% participants were classified as not depressed by both tools. However, the BDI‐II missed a total of 24.4% cases of elevated depressive symptoms detected by MMPI‐2 Scale 2, whereas Scale 2 only missed 5.7% of depressed cases detected by the BDI‐II. Depression rates did not vary significantly by subsample. Over‐reporting and under‐reporting rates were similar for both subsamples. Multiple linear regression showed MMPI‐2 depression (Scale 2) and over‐reporting of symptoms (scale Fb) predicted BDI‐II scores in both subsamples. Under‐reporting of symptoms (scales L) was found to be a significant predictor of BDI‐II scores in the seizure disorders sample. Conclusions If only the BDI‐II is used to screen for depression, there is a risk of misclassification errors as BDI‐II scores are influenced by over‐reporting and under‐reporting of symptoms.