The development of an automated system for activity-weighted size distribution measurements now permits more complete exposure and dose assessment for indoor radon decay products. Exposures characterized by the semi-continuous measurement of activity-weighted size distributions of radon decay products were obtained over four test periods in three normally occupied houses, two of which were occupied by cigarette smokers. These measured activity size distributions were used to calculate exposure-dose conversion coefficients and the annual effective doses. These doses were found to be approximately two-fold higher than the values derived conventionally from the measured radon concentration, on the assumption that indoor exposure to 20 Bq m-3 radon gas concentration corresponds to an annual effective dose 1 mSv y-1. The degree to which aerosol-measurement-based dose estimates were higher than radon-gas-based estimates was found to be influenced if the study house occupant was a cigarette smoker. Reassessment of the measured PAEC-weighted radon progeny particle size distribution in terms of the classical "unattached" and "attached" modes yielded lower estimates of the exposure-effective dose conversion coefficient that were similar to the reference values derived from a recent study by the National Research Council. Thus, by not resolving the measured radon progeny PAEC that is associated with particles intermediate in size between the two classical radon progeny modes, the estimated annual effective doses were also found to be similar to the values derived conventionally from the measured radon gas concentration. It is concluded that the observed presence of 4 to 13% of the radon progeny PAEC in the size-range 1.5 to 15 nm diameter is a dosimetrically significant factor that appears to be commonplace in various home environments.