This study investigated the potential influences of a marital interaction involving affectionate touch and/or positive relationship-focused conversation on physiological reactivity to a subsequent laboratory stress task, and whether depressive symptoms moderated these relations. It was hypothesized that 1) the stress task would cause cardiac sympathetic activation and cardiac parasympathetic withdrawal; and that physical affection and/or positive conversation would 2) reduce sympathetic activation as indicated by cardiac interbeat interval (IBI), cardiac pre-ejection period (PEP), and finger pulse transit time (FPTT) and 3) reduce parasympathetic withdrawal (as indicated by respiratory sinus arrhythmia; RSA) in response to stress. Further, we expected that, compared to those lower in reported depressive symptoms, those higher in depressive symptoms 4) would show blunted cardiovascular activation in response to stress across experimental conditions; and after engaging in a positive marital exchange, 5) would demonstrate a smaller interaction-related reduction in stress-related sympathetic activation; but 6) show no difference in interaction-related reduction of stress-related parasympathetic withdrawal. Participants were 183 married couples who were at least moderately happy in their marriages and in generally good health. Participants completed a measure of depression (among other questionnaires) in an online survey, then attended a 3-hour laboratory session. After measuring baseline physiology with spouses in separate rooms, couples were then randomly assigned to either touch (while sitting quietly, then hug), talk (positive conversation, but no touch), both (touch while talking, then hug), or neither (sit quietly without touching or talking). Next, participants separately performed a stress-inducing speech task about their spouses’ strengths and weaknesses. Physiological indicators were recorded throughout the stress task. While positive conversation reduced husbands’ stress-related parasympathetic withdrawal, it predicted greater stress-related activation in wives’ PEP response. Stress reactivity (as indicated by FPTT) was reduced in husbands with lower depressive symptoms when the marital exchange included only touch or only talk, whereas for husbands with more depressive symptoms, there were no effects of the marital interaction. For wives, depressive symptoms predicted blunted cardiovascular activation regardless of positive interaction condition, as illustrated by smaller stress-related reduction in FPTT responses. Furthermore, higher self-reported depressive symptoms predicted larger interaction-related decreases in stress-related IBI responses in wives who experienced spousal touch. This study builds on previous work and is the first to explore how depressive symptoms may influence the relations between affectionate touch and stress reactivity.