Inconsistencies in the alertness dimensions within measures of sleepiness

This research investigates whether there is any significant relationship between four commonly used measures of sleepiness; the Multiple Sleep Latency test (MSLT), the Psychomotor Vigilance Task (PVT), the Epworth Sleepiness Scale (ESS) and the Karolinska Sleepiness Scale (KSS) in a group of healthy alert individuals, and to what extent the individual factors of sex, morningness-eveningness disposition and personality variables are related to these measures of sleepiness. Fifty normal sleepers (26 females, 24 males) aged between 21-40 years of age (mean age 25.18 years) volunteered undergo a standard research MSLT. All participants had attained 7-8 hours of sleep prior to participation. Between the sleep opportunities they completed a practice and experimental session of the PVT, the Eysenck Personality Questionnaire (Adult EPQ-R) and the State-Trait Anxiety Inventory version (STAI Form Y). Subjective levels of sleepiness were collected using the KSS before each sleep opportunity of the MSLT, and either side of the PVT sessions. The Horne- Ostberg Morningness-Eveningness disposition questionnaire and ESS scores were collected prior to the testing day. There was no significant correlation between the measures of sleepiness using the whole data set. Relationship between the measures of sleepiness was not improved by time of day. When the data was split by latency on the MSLT, a significant negative correlation was found between the MSLT and ESS, but only in nine participants. Sex of the participant was only significantly related to the ESS. The MSLT was the only measure of sleepiness found to be significantly related to the morningness/eveningness disposition of the participants. Personality characteristics were not significantly related to the ESS or PVT. A significant positive relationship between neuroticism score and latency on the MSLT was found when those participants who did not sleep in the MSLT were excluded from analysis. Scores on the KSS were significantly related to levels of neuroticism, state and trait anxiety and scores on the Lie scale of the EPQ. It is concluded that although these measures of sleepiness are commonly used together, they do not have a significant relationship to each other in a group of healthy alert individuals. This suggests that each measure of sleepiness may be measuring a different component of the alertness-sleepiness spectrum, therefore no one measure can be relied upon to replace another in the measurement of alertness or sleepiness. Some significant relationships between the alertness dimensions of certain measures of sleepiness and the individual differences within the participant group have been revealed, yet as no single individual difference measured in this investigation was significantly related to all of the measures of sleepiness, it is unlikely that these particular individual differences in the participant group are solely responsible for the inconsistencies in the alertness dimensions of these measures of sleepiness. Further investigation is required in order to further establish why there are inconsistencies in the alertness dimensions of these measures of sleepiness.