Adolescence is characterized by profound change, including increases in negative emotions. Approximately 84% of American adolescents own a smartphone, which can continuously and unobtrusively track variables potentially predictive of heightened negative emotions (e.g. activity levels, location, pattern of phone usage). The extent to which built-in smartphone sensors can reliably predict states of elevated negative affect in adolescents is an open question.

Adolescent participants (n = 22; ages 13–18) with low to high levels of depressive symptoms were followed for 15 weeks using a combination of ecological momentary assessments (EMAs) and continuously collected passive smartphone sensor data. EMAs probed negative emotional states (i.e. anger, sadness and anxiety) 2–3 times per day every other week throughout the study (total: 1145 EMA measurements). Smartphone accelerometer, location and device state data were collected to derive 14 discrete estimates of behavior, including activity level, percentage of time spent at home, sleep onset and duration, and phone usage.

A personalized ensemble machine learning model derived from smartphone sensor data outperformed other statistical approaches (e.g. linear mixed model) and predicted states of elevated anger and anxiety with acceptable discrimination ability (area under the curve (AUC) = 74% and 71%, respectively), but demonstrated more modest discrimination ability for predicting states of high sadness (AUC = 66%).

To the extent that smartphone data could provide reasonably accurate real-time predictions of states of high negative affect in teens, brief ‘just-in-time’ interventions could be immediately deployed via smartphone notifications or mental health apps to alleviate these states.

Major depressive disorder (MDD) is a highly prevalent psychiatric condition, yet many patients do not receive adequate treatment. Novel and highly scalable interventions such as internet-based cognitive-behavioral-therapy (iCBT) may help to address this treatment gap. Anhedonia, a hallmark symptom of MDD that refers to diminished interest and ability to experience pleasure, has been associated with reduced reactivity in a neural reward circuit that includes medial prefrontal and striatal brain regions. Whether iCBT can reduce anhedonia severity in MDD patients, and whether these therapeutic effects are accompanied by enhanced reward circuit reactivity has yet to be examined.

Fifty-two MDD patients were randomly assigned to either 10-week iCBT (n = 26) or monitored attention control (MAC, n = 26) programs. All patients completed pre- and post-treatment assessments of anhedonia (Snaith–Hamilton Pleasure Scale; SHAPS) and reward circuit reactivity [monetary incentive delay (MID) task during functional magnetic resonance imaging (fMRI)]. Healthy control participants (n = 42) also underwent two fMRI scans while completing the MID task 10 weeks apart.

Both iCBT and MAC groups exhibited a reduction in anhedonia severity post-treatment. Nevertheless, only the iCBT group exhibited enhanced nucleus accumbens (Nacc) and subgenual anterior cingulate cortex (sgACC) activation and functional connectivity from pre- to post-treatment in response to reward feedback. Enhanced Nacc and sgACC activations were associated with reduced anhedonia severity following iCBT treatment, with enhanced Nacc activation also mediating the reduction in anhedonia severity post-treatment.

These findings suggest that increased reward circuit reactivity may contribute to a reduction in anhedonia severity following iCBT treatment for depression.

Anhedonia is a core symptom of depression that predicts worse treatment outcomes. Dysfunction in neural reward circuits is thought to contribute to anhedonia. However, whether laboratory-based assessments of anhedonia and reward-related neural function translate to adolescents' subjective affective experiences in real-world contexts remains unclear.

We recruited a sample of adolescents (n = 82; ages 12–18; mean = 15.83) who varied in anhedonia and measured the relationships among clinician-rated and self-reported anhedonia, behaviorally assessed reward learning ability, neural response to monetary reward and loss (as assessed with functional magnetic resonance imaging), and repeated ecological momentary assessment (EMA) of positive affect (PA) and negative affect (NA) in daily life.

Anhedonia was associated with lower mean PA and higher mean NA across the 5-day EMA period. Anhedonia was not related to impaired behavioral reward learning, but low PA was associated with reduced nucleus accumbens response during reward anticipation and reduced medial prefrontal cortex (mPFC) response during reward outcome. Greater mean NA was associated with increased mPFC response to loss outcome.

Traditional laboratory-based measures of anhedonia were associated with lower subjective PA and higher subjective NA in youths' daily lives. Lower subjective PA and higher subjective NA were associated with decreased reward-related striatal functioning. Higher NA was also related to increased mPFC activity to loss. Collectively, these findings demonstrate that laboratory-based measures of anhedonia translate to real-world contexts and that subjective ratings of PA and NA may be associated with neural response to reward and loss.

Treatment for major depressive disorder (MDD) is imprecise and often involves trial-and-error to determine the most effective approach. To facilitate optimal treatment selection and inform timely adjustment, the current study investigated whether neurocognitive variables could predict an antidepressant response in a treatment-specific manner.

In the two-stage Establishing Moderators and Biosignatures of Antidepressant Response for Clinical Care (EMBARC) trial, outpatients with non-psychotic recurrent MDD were first randomized to an 8-week course of sertraline selective serotonin reuptake inhibitor or placebo. Behavioral measures of reward responsiveness, cognitive control, verbal fluency, psychomotor, and cognitive processing speeds were collected at baseline and week 1. Treatment responders then continued on another 8-week course of the same medication, whereas non-responders to sertraline or placebo were crossed-over under double-blinded conditions to bupropion noradrenaline/dopamine reuptake inhibitor or sertraline, respectively. Hamilton Rating for Depression scores were also assessed at baseline, weeks 8, and 16.

Greater improvements in psychomotor and cognitive processing speeds within the first week, as well as better pretreatment performance in these domains, were specifically associated with higher likelihood of response to placebo. Moreover, better reward responsiveness, poorer cognitive control and greater verbal fluency were associated with greater likelihood of response to bupropion in patients who previously failed to respond to sertraline.

These exploratory results warrant further scrutiny, but demonstrate that quick and non-invasive behavioral tests may have substantial clinical value in predicting antidepressant treatment response.

Cognitive deficits in depressed adults may reflect impaired decision-making. To investigate this possibility, we analyzed data from unmedicated adults with Major Depressive Disorder (MDD) and healthy controls as they performed a probabilistic reward task. The Hierarchical Drift Diffusion Model (HDDM) was used to quantify decision-making mechanisms recruited by the task, to determine if any such mechanism was disrupted by depression.

Data came from two samples (Study 1: 258 MDD, 36 controls; Study 2: 23 MDD, 25 controls). On each trial, participants indicated which of two similar stimuli was presented; correct identifications were rewarded. Quantile-probability plots and the HDDM quantified the impact of MDD on response times (RT), speed of evidence accumulation (drift rate), and the width of decision thresholds, among other parameters.

RTs were more positively skewed in depressed v. healthy adults, and the HDDM revealed that drift rates were reduced—and decision thresholds were wider—in the MDD groups. This pattern suggests that depressed adults accumulated the evidence needed to make decisions more slowly than controls did.

Depressed adults responded slower than controls in both studies, and poorer performance led the MDD group to receive fewer rewards than controls in Study 1. These results did not reflect a sensorimotor deficit but were instead due to sluggish evidence accumulation. Thus, slowed decision-making—not slowed perception or response execution—caused the performance deficit in MDD. If these results generalize to other tasks, they may help explain the broad cognitive deficits seen in depression.

Major depressive disorder (MDD) is a highly heterogeneous condition in terms of symptom presentation and, likely, underlying pathophysiology. Accordingly, it is possible that only certain individuals with MDD are well-suited to antidepressants. A potentially fruitful approach to parsing this heterogeneity is to focus on promising endophenotypes of depression, such as neuroticism, anhedonia, and cognitive control deficits.

Within an 8-week multisite trial of sertraline v. placebo for depressed adults (n = 216), we examined whether the combination of machine learning with a Personalized Advantage Index (PAI) can generate individualized treatment recommendations on the basis of endophenotype profiles coupled with clinical and demographic characteristics.

Five pre-treatment variables moderated treatment response. Higher depression severity and neuroticism, older age, less impairment in cognitive control, and being employed were each associated with better outcomes to sertraline than placebo. Across 1000 iterations of a 10-fold cross-validation, the PAI model predicted that 31% of the sample would exhibit a clinically meaningful advantage [post-treatment Hamilton Rating Scale for Depression (HRSD) difference ⩾3] with sertraline relative to placebo. Although there were no overall outcome differences between treatment groups (d = 0.15), those identified as optimally suited to sertraline at pre-treatment had better week 8 HRSD scores if randomized to sertraline (10.7) than placebo (14.7) (d = 0.58).

A subset of MDD patients optimally suited to sertraline can be identified on the basis of pre-treatment characteristics. This model must be tested prospectively before it can be used to inform treatment selection. However, findings demonstrate the potential to improve individual outcomes through algorithm-guided treatment recommendations.