Speech as an Objective Marker of Depression and Associated Cognitive Difficulties

Psychiatry lacks objective biomarkers for assessing depression. Instead, symptoms are measured using self- and clinician-reported measures, with the Hamilton Depression Rating Scale (HAMD-17) widely regarded as the gold standard. However, traditional assessments like the HAMD-17 often fail to capture all aspects of depression, particularly cognitive difficulties. This study examined whether speech features could serve as objective markers of depressive symptoms, including cognitive difficulties. Sixty-six individuals with depression and 54 non-depressed control participants completed a speech assessment, responding to the prompt: “Please tell me how you are feeling today.” Linguistic (valence, emotional intensity, agency) and acoustic (pitch, pitch variance, speech rate, pause duration) features were derived from natural language processing. Features were analyzed both individually and as a composite score. A subset of participants (N = 59) also completed an eye-tracking task assessing executive function. ANCOVAs compared speech features between groups, and linear regression models examined the associations between speech features, depression severity (HAMD-17), and executive function performance. Compared to control participants, individuals with depression used language that was significantly more negatively valenced, emotionally intense, and lower in agency. They also exhibited significantly lower pitch, slower speech rate, more pauses, and differed in the composite score. However, speech features were not significantly associated with depression severity. While depression severity was not associated with executive function performance, several speech features were, including negative valence, emotional intensity, agency, pause duration, and the composite score. These findings suggest that speech features may offer an accessible, objective method for assessing both depression and associated cognitive difficulties.