O. H. Rundell

Alcohol and sleep in young adults

In two separate experiments, sleep patterns of 17 young male adults were examined following single and repeated doses (0.9 g/kg body weight) of alcohol. A third study of 10 additional Ss measured the rate of elimination of alcohol from blood during sleep and waking. With a single dose of alcohol, onset of sleep was brisk, the latency of SW sleep (stages 3+4) was reduced and the first episode of stage REM was shortened. These transient alterations were accompanied by loss of high-frequency beta rhythms in the EEG and a gain in abundance and synchrony of activity in the alpha-rhythm range.Experiment 3 found no significant difference between sleep and waking for rate of elimination of alcohol from blood. Starting with peak blood alcohol concentrations (BACs) of about 75 mg percent, average BACs after 4 h of bedtime had decreased to about 30 mg percent. The effects noted above, associated with higher BACs, were confined to the first half of the night. Thus, these results are consistent with conclusions of previous investigators that the depressant effects on EEG sleep patterns of a moderate dose of alcohol are due to its direct action on the brain.There are, however, longer range compensation and adaptation effects associated with single and repeated doses of alcohol which cannot be directly related to its presence in the brain. For example, in the single-dose study reported here, and in most previous studies, rebound of stage REM occurred during the second half of the alcohol night. Further, in the repeated-dose study, most of the effects of alcohol on sleep stages and EEG frequencies observed during the first dose session, disappeared on the second and third alcohol sessions. Finally, heart and respiration rates increased whereas eye movements during stage REM, sigma spindles in stage 2 and non-specific GSR responses in SW sleep tended to be suppressed throughout the night in each alcohol session. Several mechanisms are discussed which might account for these more persistent alterations.

Alcohol and information processing

Three experiments are reported which investigate the effects of acute alcohol intoxication (average blood alcohol concentration 100 mg-%) on some aspects of human information processing. The results are interpreted within the framework of a general information processing model (Smith, 1968), using the Sternberg (1969b) additive-factor method of analysis. Alcohol consistently impaired information outputting operations (i.e., response selection-organization), rather than information inputting operations (i.e., stimulus preprocessing and encoding).

Chronic Alcoholism, Alcohol and Sleep

Since the demonstration of Dement and Kleitman (1957) that the periodic occurrence of rapid eye movements (REMs) in the presence of electroencephalographic (EEG) desynchrony during sleep was associated with visual dreams, and the immediate confirmation by Dement (1958) of similar bioelectric patterns in the cat, a wealth of information has accumulated concerning many behavioral and biological aspects of sleep. Recently, increasing efforts have been made to relate the significance of this variety of data to clinical areas, including the problems of chronic alcoholism and other drug abuse.

Alcohol and Speed-Accuracy Tradeoff

Performance on two auditory choice reaction time (RT) tasks was studied in a group of 12 subjects under the influence of graded doses of ethyl alcohol ranging from placebo to 1 g/kg body weight. Deadline procedures were employed in a side discrimination and a pitch discrimination task to permit the calculation of speed-accuracy tradeoff functions (accuracy versus RT). Accuracy declined as a function of dose, but alcohol did not significantly influence RT. Conversely, accuracy was not affected by task; but the pitch discrimination task required an average of 88 ms more time than the side task. Alcohol dose and task produced independent effects on the speed-accuracy tradeoff function. As dose increased, the slope of the tradeoff function declined; but slopes were equivalent for the two tasks. On the other hand, the x-intercept (where accuracy equals chance levels) was 90 ms greater for the pitch task than for the side task.

Sleep in alcoholic patients: longitudinal findings.

Alterations of sleep associated with both acute and chronic ingestion of alcohol have received a good deal of attention in the past few years. The picture that has emerged from this work suggests an initial sedative effect of alcohol with brisk sleep onset, decreased stage REM, and enhanced slow-wave sleep (Rundell, Lester, Griffiths, and Williams, 1972; Yules, Freedman, and Chandler, 1966; Gross, Goodenough, Nagarajan, and Hastey, 1973; Lester, Rundell, Cowden, and Williams, 1976). As blood levels decline following a moderate dose of alcohol, the sedative effects disappear, and the second half of the night’s sleep usually shows enhanced stage REM and increased sleep disturbance (Yules et al., 1966; Knowles, Laverty, and Kuechler, 1968; Rundell et al., 1972). If drinking continues, the sedative effects lessen after a few nights, and increased doses are required to obtain sedation (Gross, et al., 1972; Lester, et al., 1976; Rundell, et al., 1972).

Alcohol and Sleep in the Chronic Alcoholic

The purposes of this study were to compare physiological sleep profiles of sober chronic alcoholics with those of age-matched normal control subjects, and to examine the effects of two days of drinking on the sleep of the alcoholic patients.1

Secobarbital and information processing.

The Sternberg fixed-set memory-search paradigm was used to assess the relative vulnerability of hypothetical stages of information processing to an oral dose of secobarbital (2.9 mg/kg). D-amphetamine (15 mg, oral dose) was intended to serve as an active placebo. However, since the amphetamine produced a slight, non-significant reduction in choice reaction time (RT), the principal analysis of secobarbital effects was conducted between drug and baseline conditions. Secobarbital slowed choice RT by 60 msec. and did not increase errors significantly. The results, as interpreted within Sternbergs model, suggest that input processes, e.g., stimulus preprocessing-encoding, are particularly sensitive to the effects of the barbiturate. There was no evidence of a drug effect on cognitive processes associated with serial comparison, binary decision, or translation-response organization (response selection). In contrast, earlier studies have indicated that another CNS depressant, alcohol, interferes with both speed and accuracy of output processes, viz., the response selection stage.