Part 2 of two companion papers is a continuation of the discussion presented in Part 1, and is concerned with the effects of the noise level and type of noise source on activity disturbance as obtained by a simulated-environment study. The design of the experiment, with account taken of the factors which are important in simulation of noise, environment and community, have been explained in detail in Part 1. Tests each lasting 30 min consisted of two consecutive parts under continuous and similar noise conditions, enabling the subjects to perform two different activities, namely, reading and listening without being interrupted. The indoor levels of the road, railway and aircraft noise samples (only side flights were considered) varied in the range of 30-55 dB(A) in L-eq (30 min) with increments of 5 dB(A). 64 subjects participated and each subject attended three different tests. The statistical results indicated that the correlation coefficients between the activity disturbance and noise level were high (r = 0.951 and 0.970) for the GROUP DATA, and the comparison of the dose and annoyance relationships obtained for reading and listening situations, revealed a shift at 45 dB(A) after which the listening annoyance suddenly increased with the noise level. The source-type effect was found to be significant for the listening annoyance and for Summindex (P < 0.05). The reading annoyance did not significantly depend on the source type, probably because of the deeper concentration of the subjects. This implies that the source type may not be a very important factor in daily life activities when transportation noises intrude from the facade. The comparison of (C) 1999 Academic Press.