Both elevated cardiovascular mortality and low cardiovagal (parasympathetic) heart rate variability (HRV)—a risk factor for postmyocardial infarction—are reported in schizophrenia (SZ). Since a number of medications have strong effects of cardiac conductivity, we thought it important to examine if typical neuroleptic medications might also affect HRV. We examined cardiac vagal activity during both neuroleptic treatment and a drug-free condition in seven SZ patients who were participating in a pilot double-blind, crossover study of placebo and haloperidol treatment. Twenty-four–hour Holter electrocardiograms were analyzed for high frequency HRV, quantitated as the percent of successive normal interbeat intervals greater than 50 milliseconds (PNN50), which is a good index of parasympathetic cardiac modulation. The patients showed unchanged PNN50 (8.4±9.5 versus 8.3±10.5; t=.22, df=6, P=.5) between the haloperidol treatment and drug-free conditions. Despite the elapsed time, change in medication, and altered clinical state, the PNN50s were highly correlated (Spearman r=.98, P=.000). SAPS positive symptom scores declined with treatment from 12.8±6.5 to 8.5±3.5; paired t=3.26: df=6; P=.01. PNN50s were significantly associated with positive (r=-.86, df=6, P=.012) and negative symptom scores (r=-.87, df=6, P=.01). We found low cardiovagal modulation in medication-free SZ patients that was associated with core SZ symptoms and was unchanged by haloperidol and benztropine treatment. The reduced HRV in SZ patients at baseline may render them at greater cardiovascular risk than healthy subjects when treated with medications having strong cardiovascular effects.