A SIGNAL DETECTION ANALYSIS OF SLEEP-WAKE PERCEPTION IN NORMAL SLEEP AND THE ALTERATION OF THAT PERCEPTION BY A PRIORI SLEEP AND A PRIORI WAKE INSTRUCTION SETS.
Item
-
Title
-
A SIGNAL DETECTION ANALYSIS OF SLEEP-WAKE PERCEPTION IN NORMAL SLEEP AND THE ALTERATION OF THAT PERCEPTION BY A PRIORI SLEEP AND A PRIORI WAKE INSTRUCTION SETS.
-
Identifier
-
AAI8302544
-
identifier
-
8302544
-
Creator
-
SEWITCH, DEBORAH E.
-
Contributor
-
John S. Antrobus
-
Date
-
1982
-
Language
-
English
-
Publisher
-
City University of New York.
-
Subject
-
Psychology, Experimental
-
Abstract
-
Previous work in the field has alluded to discrepancies between the polygraphic criteria used to define sleep as currently standardized and subjective reports of sleep and wakefulness for normal sleepers. In two experiments, a Signal Detection Theory framework was used to systematically assess the extent of this discrepancy in normal sleepers. The main hypothesis was that polygraphic criteria are an external, recordable index of an internal "state" that is detected by subjects and used by them to decide whether they have been awake or asleep.;Eleven normal sleepers (20-35 years old) spent six nonconsecutive nights, one adaptation night and five experimental nights, in the laboratory. An average of twelve trials were studied per experimental night: eight during polygraphically defined sleep (Stage 2 or REM) and four during wakefulness. For each sleep or wake trial, the subject was alerted by a telephone ring and asked to state whether he was "awake" or "asleep" just before hearing the ring. The subject was then asked to rate how confident he was of that answer on a three-point scale: (1) positive, (2) pretty sure, or (3) not so sure.;Six of the original eleven subjects were asked to participate in a second condition (an additional five experimental nights) of study. They were assigned, in a counter-balanced order, to one of two a priori instruction set and limited performance feedback conditions. Sleep-wake discrimination out of REM sleep was affected significantly more by a priori instruction set and limited performance feedback than sleep-wake discrimination out of stage 2 sleep, F(1,4) = 14.9, p < .025. A priori instruction set and limited feedback did significantly affect the response bias associated with discrimination of EEG wakefulness from EEG stage 2 or REM sleep, F(1,4) = 6.89, p < .10.;The hypothesis that polygraphic data are closely correlated with internal signals that are detected by subjects and used by them to decide whether they have been awake or asleep was supported only in part. Normal sleepers do not detect polygraphic stage 2 sleep as being sleep fifty-five percent of the time and they do not detect REM sleep as being sleep twenty-seven percent of the time. The significant bias associated with all subject responses toward reporting 'awake' regardless of polygraphic criteria demonstrates that the sleep-wake decision process of normal sleepers is strongly dependent on factors independent of the conventional polygraphic criteria. These results seriously question the adequacy of conventional polygraphic criteria to decide whether or not someone has been asleep.
-
Type
-
dissertation
-
Source
-
PQT Legacy CUNY.xlsx
-
degree
-
Ph.D.
-
Program
-
Psychology
