Late-life depression (LLD) is associated with disability, increased risk for cognitive decline and dementia, elevated suicide risk, and greater all-cause mortality. These outcomes are related to depression being a recurrent disorder, with repeated episodes over a patient's lifetime. Recurrence rates (defined as including both relapse and recurrence) are high in LLD.

The goals of this study are to identify neurobiological factors that predict recurrence risk, and examine how cognitive performance changes are both influenced by these neurobiological factors and also predict recurrence risk.

Inclusion Criteria:

• Currently depressed participants (who will enter the treatment algorithm): 1) Age > 60 years; 2) diagnosis of major depressive disorder, recurrent episode (DSM5); 3) severity: Montgomery Asberg Depression Rating Scale (MADRS) (67) ≥ 15; 4) cognition: Montreal Cognitive Assessment (MoCA) >24 or Montreal Cognitive Assessment (MoCA) - BLIND ≥ 18; 5) fluent in English.
• Remitted depressed participants (who will directly enter the longitudinal study): 1) Age > 60 years; 2) diagnosis of major depressive disorder, recurrent, in partial or full remission (DSM5); 3) severity: MADRS < 10; 4) cognition: MoCA >24 or Montreal Cognitive Assessment (MoCA) - BLIND ≥ 18; 5) fluent in English.
• Never- depressed elders: 1) Age > 60 years; 2) severity: MADRS < 8; 3) cognition: MoCA >24 or MoCA - BLIND ≥ 18; 4) fluent in English.

Exclusion Criteria:

• Currently depressed participants (who will enter the treatment algorithm): 1) Other Axis I psychiatric disorders, except for simple phobia or anxiety disorders present during the depressive episode (e.g., generalized anxiety disorder (GAD) or panic disorder symptoms); 2) History of alcohol or drug dependence or abuse (other than nicotine) in the last year; 3) History of a developmental disorder or history of IQ < 70; 4) Acute suicidality within 3 months of study entry; 5) Acute grief (<1 month); 6) Current or past psychosis; 7) Primary neurological disorder, including dementia, clinical stroke, brain tumor, epilepsy, etc.; 8) Presence of unstable medical illness requiring urgent treatment; 9) Any MRI contraindication; 10) ECT in last 6 months; 11) Vagal Nerve Stimulation within 6 months of study entry; 12) Current use of TMS, ketamine, or esketamine with plans to continue treatment.
• Remitted depressed participants (who will directly enter the longitudinal study): Identical to the currently depressed participants, plus 1) currently receiving brain stimulation treatment (ECT, TMS, VNS), ketamine or esketamine. There are no other exclusions for antidepressant treatment.
• Never- depressed elders: Identical to the currently depressed participants, plus: 1) No diagnosis of current or past depressive disorder or other Axis I disorder except for simple phobia; 2) No history of psychotropic medication use for psychiatric symptoms.

We will not be enrolling vulnerable populations, specifically pregnant women, children, prisoners, or institutionalized individuals. We also will not enroll subjects incapable of providing their own consent. Should a potential subject present where there is a concern (either by a study clinician or staff member) about their ability to understand study procedures and provide meaningful consent, their cognitive ability and understanding will be evaluated by a study doctor. If there is any concern that the individual may be impaired, they will not be enrolled in the study. The rationale will be provided to the individual as well as his or her family members. Referrals for further evaluation, including urgent or emergent evaluation, will be made as needed and clinically warranted.

We will enroll patients from clinical referrals and response to advertisements. For clinical referrals, referring providers may either provide information about the study to the participant, including our contact information. Alternatively, they may obtain permission from the participant for the study team to contact them. With the clinician's agreement, study staff can review their clinic schedules to help identify potentially eligible patients, however data will not be recorded for research purposes. Advertisements may include radio, newspapers, internet, online research registries, public transportation, and flyers / brochures.