Several studies have shown a relationship between blood pressure (BP) and cognitive function. Yet very few studies have addressed circadian BP patterns in this context, perhaps due to poor availability of suitable methods to detect slight changes in the cognitive state. Today, brain event-related potentials (ERPs) allow us to detect subclinical changes in cognitive function. We enrolled 30 elderly patients with recently diagnosed hypertension (<2 years) that had never been treated: 18 dippers and 12 nondippers. Patients underwent 24-h ambulatory blood pressure monitoring (ABPM). Careful assessment of their cognitive state was carried out using the mini mental state examination (MMSE), and the recording of P300 and N2 ERPs. No significant differences between the two groups were found. MMSE scores in dippers and nondippers were similar (29.5+/-0.71 vs. 29.3+/-1.07, respectively; p =0.611), as were P300 latency values (377.78+/-33.28 vs. 364.67+/-35.12 in the central (Cz) position, p =0.310; 379.22+/-32.94 vs. 365.25+/-35.07 in the occipital (Pz) position, p =0.277) and N2 wave latency values (253.83+/-24.9 vs. 249.17+/-24.47 in the Cz position, p =0.617; 251.56+/-25.86 vs. 246.58+/-25.46 in the Pz position, p =0.608). These data show no association between the nondipping pattern and lower cognitive function in elderly subjects with recent hypertension.