HABITUATION OF SLEEP TO ROAD TRAFFIC NOISE AS DETERMINED BY POLYSOMNOGRAPHY AND AN ACCELEROMETER

Abstract

The habituation of human sleep to a noisy environment was investigated by polysomnography (PSG), a wrist activity device (Actiwatch®), subjective evaluation and a performance test on the following morning. Eleven young male students slept for 17 nights in a sleep laboratory. PSG on the first, fourth, fifth, ninth, 14th, and 17th nights was judged visually. Four of the subjects were continuously monitored by the wrist activity device. From the fifth to 14th nights, there was exposure to road traffic noise all-night long, and consecutive experiments were conducted from the fifth to 17th nights. Agreement of sleep/wake assessment for Actiwatch®and PSG was 88·4%, on average, based on the data for 24 nights. Pearson's correlation coefficient of TST for Actiwatch®and sleep PSG was 0·848. Habituation to noise by wrist movement, sleep latency by PSG, and activity of mental muscles was not recognized. The association between wrist activity and mental muscle activity was significant for three subjects out of four (r=0·56, 0·81, 0·71, respectively). Percentages of positive wrist movement in each sleep stage, such as the 3+4 stages, REM stage and stage MT, were compared with those in other stages. Wrist activity in Stage REM was significantly more frequent than that in other stages for the three subjects. Wrist movement in Stage MT was significantly more frequent than in other stages for the three subjects. REM latency, REM cycle, and five factors of subjective sleep, from the Oguri–Shirakawa–Azumi questionnaire (SQ), showed significant differences by analysis of variance for repeated measurements. When change from the 4th night was checked, sleepiness, worry, integrated sleep feeling and sleep initiation by SQ showed habituation of sleep to noise. Namely, sleep quality recovered to the level on a silent night by the fifth noisy night during the experiment. There is thus a habituation of sleep to noise when a subjective evaluation of sleep, such as the SQ, is used.