Abstract
Introduction
Until today, subjective sleepiness research has focused mainly on its relationship with objective sleepiness. A different, theoretically interesting question is on the composite nature of subjective sleepiness: "how do different factors that contribute to objective sleepiness combine into subjective sleepiness?"
Functional Measurement (FM) has been proposed as a framework to identify such combination rules using factorial designs (Anderson, 1981). FM combines subjective representations of several sleepiness-inducing stimuli into a single subjective response according to an integration function, whereafter a response function converts it into an overt response (Anderson, 1981; Anderson, 1996).
Method and results
Thirteen participants enrolled in a full-factorial 3x3 FM experiment. The arousal factor was manipulated in three levels (baseline (8h of sleep), deprivation (2.30h of sleep), stress (2.30h of sleep and intermittent speech tasks), along with the circadian factor (time of day: 9.00, 11.00 and 13.00hrs). Subjective judgments of sleepiness were collected using commonly used rating scales (KSS, SSS, VAS, Pictorial Sleepiness Scale (PSS)), all semi-counterbalanced and presented before, during and after each counterbalanced MSLT/MWT session.
For most judgment tasks it is found that the integration function is either one of three combination rules: addition, multiplication or average (Anderson, 1996). In a factorial graph, parallelism supports an additive model. Since non-additivity or a non-linear response scale would undo parallelism in the data, observed parallelism is considered to support both an additive integration rule and linearity of the response scale. Our results support an additive model and linearity for all response scales: parallelism is shown (Figure 1), which means that all main effects are significant and no significant interaction is found (Table 1).
Table 1: ANOVA results for interaction terms
interaction term
(arousal*TOD) F(4,48) p
SSS KSS VAS PSS 1.30 .282
Conclusion
The perceived magnitude of objective stimuli such as arousal level and time of day are integrated by means of a simple additive model into an overall appreciation of sleepiness. In addition, all of the sleepiness scales used in this study were found to be linear response measures.
References
Anderson, N.H. (1981). Foundations of Information Integration Theory. London: Academic Press
Anderson, N.H. (1996). A functional theory of cognition. New Jersey: Lawrence Erlbaum Associates.
Until today, subjective sleepiness research has focused mainly on its relationship with objective sleepiness. A different, theoretically interesting question is on the composite nature of subjective sleepiness: "how do different factors that contribute to objective sleepiness combine into subjective sleepiness?"
Functional Measurement (FM) has been proposed as a framework to identify such combination rules using factorial designs (Anderson, 1981). FM combines subjective representations of several sleepiness-inducing stimuli into a single subjective response according to an integration function, whereafter a response function converts it into an overt response (Anderson, 1981; Anderson, 1996).
Method and results
Thirteen participants enrolled in a full-factorial 3x3 FM experiment. The arousal factor was manipulated in three levels (baseline (8h of sleep), deprivation (2.30h of sleep), stress (2.30h of sleep and intermittent speech tasks), along with the circadian factor (time of day: 9.00, 11.00 and 13.00hrs). Subjective judgments of sleepiness were collected using commonly used rating scales (KSS, SSS, VAS, Pictorial Sleepiness Scale (PSS)), all semi-counterbalanced and presented before, during and after each counterbalanced MSLT/MWT session.
For most judgment tasks it is found that the integration function is either one of three combination rules: addition, multiplication or average (Anderson, 1996). In a factorial graph, parallelism supports an additive model. Since non-additivity or a non-linear response scale would undo parallelism in the data, observed parallelism is considered to support both an additive integration rule and linearity of the response scale. Our results support an additive model and linearity for all response scales: parallelism is shown (Figure 1), which means that all main effects are significant and no significant interaction is found (Table 1).
Table 1: ANOVA results for interaction terms
interaction term
(arousal*TOD) F(4,48) p
SSS KSS VAS PSS 1.30 .282
Conclusion
The perceived magnitude of objective stimuli such as arousal level and time of day are integrated by means of a simple additive model into an overall appreciation of sleepiness. In addition, all of the sleepiness scales used in this study were found to be linear response measures.
References
Anderson, N.H. (1981). Foundations of Information Integration Theory. London: Academic Press
Anderson, N.H. (1996). A functional theory of cognition. New Jersey: Lawrence Erlbaum Associates.
| Original language | English |
|---|---|
| Pages (from-to) | 131-131 |
| Number of pages | 1 |
| Journal | Journal of Sleep Research |
| Volume | 15 |
| Issue number | S1 |
| Publication status | Published - Sep 2006 |
| Event | Finds and Results from the Swedish Cyprus Expedition: A Gender Perspective at the Medelhavsmuseet - Stockholm, Sweden Duration: 21 Sep 2009 → 25 Sep 2009 |
Bibliographical note
Journal of Sleep Research, p.131, Vol 15, 2006Keywords
- functional measurement
- subjective sleepiness
- Karolinska Sleepiness Scale
- Stanford Sleepiness Scale
- Visual Analogue Scale for sleepiness
- Pictorial Sleepiness Scale