Method

The study protocol was approved by the Ethical Committee of the Medical Faculty of the University of Heidelberg.

Subjects

A total of 25 healthy subjects (15 men) ages 19 to 58 years (average: 33.7 years) participated in the study. Three subjects were excluded because their alcohol absorption was insufficient.

Beverages

Each subject was offered three drinks containing identical amounts of alcohol but different amounts of thujone (100 mg/l, 10 mg/l, no α-thujone; Fluka, München, Germany). Thujone is available as a chemical reference material (purity 99%) for laboratory use only. In Germany its use is allowed only as an additive to food or beverages. For these reasons a thujone-alone subject group was not tested.

Because we did not want the subjects to know which drink they consumed, all drinks were matched to each other in color and taste by adding red food color and an infusion of fennelseed. The alcohol content was adjusted to 16 g/l in all beverages. The amount of liquid to be consumed depended on the weight of the subject. We tried to attain a maximum blood alcohol concentration (BAC) of 0.05% for each subject. The calculated total amount of thujone consumed by each subject in each experiment was 0.28 mg/kg weight (100 mg/l) and 0.028 mg/kg weight (10 mg/l) for men and 0.24 mg/kg weight (100 mg/l) and 0.024 mg/kg weight (10 mg/l) for women.

Tests

The subjects were tested using an attention performance test (APG; Mueller, 1980). This test was developed for aptitude diagnostics in the area of performance and is applied in the diagnostics of alcohol and drug-induced effects on visual orientation performance (Mueller, 1980; Strohbeck-Kuehner, 1998; Urban, 1992; Zuschlag and Jacobshagen, 1982). The attention performance test was chosen because it allows differentiation between the peripheral and central field of selective attention, a distinction that also has important implications in regard to driving. This differentiation allows the assessment of alterations in the field of attention (Werth, 1990).

The testing device is a signal-presenting tool consisting of three boards. The central board is equipped with nine lamps and is located in the subjects' central field of vision. The two adjacent boards, which are located in the subjects' peripheral fields of vision, are equipped with six lamps each. Using this experimental arrangement, an angle of vision of 130° is achieved. In intervals of 1.2 seconds the lamps flash irregularly and thus produce different patterns. Subjects were asked to press a button as soon as four neighboring lamps formed a square. In the present study, the 4 minute testing time included 50 of these signals (32 in the central field, 8 in the left peripheral field and 10 in the right). The number of correct reactions (differentiated according to central and peripheral field of attention), the unnoticed signals and the reactions that occurred without stimulation ("false alarm" reactions) were recorded. The reaction times were also recorded.

Mood was assessed using two questionnaires that test different mood dimensions. The Basel Mood Test questionnaire (BBF-Basler Befindlichkeitsskala; Hobi, 1985) records the factors "vitality," "intrapsychic equilibrium," "social extraversion" and "attentiveness." The general activation-high activation state scale (GA-HA-state; Wieland-Eckelmann, 1992) records state anxiety and current subjective activation.

Experimental design

Every subject had to undergo all three treatments (no thujone, 10 mg and 100 mg thujone/l). Each treatment was performed on a different day. The sequence of the individual treatments was randomized. To take daily variations into account, the experiments were always carried out at the same time of day. Several days before the first experiment, the subjects had to undergo a training phase with the attention-performance-testing device to get acquainted with the tool and thus reduce effects usually obtained with repeated measurement. In each session the subjects answered the mood state while testing negative for blood alcohol (T0). The attention performance of the subject was then assessed with the attention performance test. Subsequently, the subjects received a small standard meal (a roll with ham or cheese) and then the beverage, which had to be drunk within 10 minutes. All tests were performed once again 30 (T1) and 90 (T2) minutes after drinking. The mood state questionnaires were administered first and attention performance was tested immediately afterwards. Alcohol concentration was then measured using a breath analyzer (Siemens Alcomat; Siemens AG, Ditzingen, Germany), and a blood sample was taken to determine the BAC. The concentration of alcohol was determined by both flame-ionization gas chromatography/headspace analysis (Perkin Elmer, Konstanz, Germany) and an enzymatic method involving alcohol dehydrogenase (ethyl alcohol; Microgenics, Passau, Germany).

Each method was run in duplicate, and values given are the means (n = 4). The measurement points T1 and T2 were chosen to catch the BAC at its maximum and during the elimination phase, respectively. On completion of the study, the subjects were asked which beverage they believed they had consumed on each of the 3 days.

Statistical analysis

The three treatment conditions (no thujone, 10 mg and 100 mg thujone/l) represent the independent variables and the five attention parameters (attention in the central and peripheral field of vision, the reaction time in the central and peripheral field of vision and the number of "false alarm" reactions) as well as the mood dimensions—the dependent variables.

Results were analyzed by ANOVA t test for conjoint measurement. Statistical calculations were carried out using SPSS (Version 11.5; SPSS GmbH Software, München, Germany). A p value <.05 was considered to be statistically significant.

Results

The objective of this study was to examine whether the impacts of absinthe on attention performance and mood were different from those of beverages that contain only alcohol.

We first tested whether the subjects were able to classify the consumed drinks correctly. Only three of the 22 subjects were able to do so; hence, we can assume that subjects' expectations did not affect performance or mood, and expectation effects did not have to be taken into consideration.

Because the individual BACs (Table 1) differed, we tested the dependence of BAC differences, and alterations in attention performance were measured by the applied test; we followed the same procedure if changes in the mood state depended on the BAC. To do this, the BAC values were correlated with the changes in attention performance and mood state between T0 and T1, as well as between T0 and T2. To exclude accidental results obtained because the number of tested correlations was large, the data were alpha adjusted according to Bonferroni.

Pearson's correlations between the change in attention performance from T0 to T1 and the BAC were, in most cases, close to zero. Significant correlations were not observed. With regard to mood alterations (T0-T1), there were no indications that they could be primarily explained by differences in BAC (Spearman correlations). The correlation of the differences of the results obtained from the assessment of attention performance at T0 and T2 with the BAC also did not show significant differences in the mood state or attention performance values.

As the influence of blood alcohol concentration seems to have no impact on the alteration of attention performance, the changes in attention performance from T0 to T1 and from T0 to T2 could be analyzed for the three treatments independent of BAC. The design was analyzed separately for the three treatments referring to changes from T0 to T1 and from T0 to T2. The results (see Tables 2-4, Figures 15) revealed no significant alterations in attention performance after the consumption of alcohol and low thujone concentration.

The number of correct reactions in the peripheral field of attention, however, decreased significantly (p < .01) when the subjects were under the influence of the high thujone concentration (Figure 2). Reaction time increased significantly in both the peripheral and central fields of attention (p < .05) (Figures 4 and 5). The number of "false alarm" reactions also increased (p < .01) (Figure 1). No differences could be observed regarding the number of correct reactions in the central field of attention (Figure 3).

An analysis of the changes of performance from T0 to T2 by the procedure described above (see Tables 2-4) revealed results that show a pattern similar to, but less pronounced than, that observed for alterations in attention performance at T0, in contrast to T1, under the high thujone condition. Although we found no significant differences, a tendency (p < .10) could be observed for the reaction in the peripheral field of attention to deteriorate at T2 in comparison with T0 under the high thujone condition (Table 4). We also found a tendency towards longer reaction time at T2 in the central and peripheral field of attention (p < .10), whereas the number of correct reactions in the central field of attention and "false alarm" reactions did not differ. When the subjects were exposed only to alcohol or only to low thujone concentrations, no differences in attention performance were observed (Tables 2 and 3).

We next analyzed differences in attention performance between the three treatments. ANOVA that tested significant differences in the alteration of attention performance between the three treatments revealed no significant differences in attention performance either from T0 to T1 or from T0 to T2 (all F values <2.426).

When the tested mood dimensions obtained at T0 were compared with those of T1 (Tables 5-7), a significant decrease of "attentiveness" was observed under all three experimental conditions. Simultaneously, an increase in "social extraversion" was observed. If only alcohol was consumed, the "state anxiety" decreased significantly. No differences were observed with regard to "subjective activation," "vitality" and "intrapsychic equilibrium."

FIGURE 1. Comparison of the differences of false alarm reaction at T0-T1 (T0-T2) for the three treatments: Testing negative for alcohol (T0), 30 minutes after alcohol intake (T1), 90 minutes after alcohol intake (T2) (*indicates significant differences)

FIGURE 2. Comparison of the differences of peripheral correct reactions at T0-T1 (T0-T2) for the three treatments: Testing negative for alcohol (T0), 30 minutes after alcohol intake (T1), 90 minutes after alcohol intake (T2) (*indicates significant differences)

FIGURE 3. Comparison of the differences of central correct reactions at T0-T1 (T0-T2) for the three treatments: Testing negative for alcohol (T0), 30 minutes after alcohol intake (T1), 90 minutes after alcohol intake (T2)

FIGURE 4. Comparison of the differences of peripheral reaction time at T0-T1 (T0-T2) for the three treatments: Testing negative for alcohol (T0), 30 minutes after alcohol intake (T1), 90 minutes after alcohol intake (T2) (*indicates significant differences)

FIGURE 5. Comparison of the differences of central reaction time at T0-T1 (T0-T2) for the three treatments: Testing negative for alcohol (T0), 30 minutes after alcohol intake (T1), 90 minutes after alcohol intake (T2) (*indicates significant differences)

In contrast to the T1 values obtained for attention performance, which were very similar to the values measured at T0, the mood state changes at T2 seemed to persist or to be partly more prominent (see Tables 5-7). At T2, the condition "alcohol alone" (Table 5) reveals a consistently low value of "state anxiety" and "attentiveness." Lower values of "subjective activation" and "vitality" were also measured. On the other hand, the condition "social extraversion" demonstrated elevated values at T1 but decreased to the initial level at T2. The decreases in "subjective activation" and "vitality" and the decrease in "social extraversion" to the initial level could also be measured for the respective thujone concentrations (Tables 6 and 7). A lower degree of "attentiveness" was measured at both points in time. A significant reduction in "state anxiety" at T2 compared with T1 was observed at the high thujone concentration (Table 7).

As in the procedure for depicting the alterations in attention performance, the alterations in the mood state were compared with each other for all three treatment conditions. Friedman rank variance analyses provided no significant differences in the alternation of the tested mood dimensions (all chi-square values <3.04).

Discussion

Because absinthe was banned for nearly a century, only few case studies or anecdotal observations are available for the beverage and must be interpreted with caution (Albert-Puelo, 1981; Arnold, 1988, 1989; Bonkovsky et al., 1992; Huckenbeck, 2001; Loftus and Arnold, 1991; Monroe, 1991; Strang et al., 1999; Vogt and Montagne, 1982; Weisbord, 1997). Most of the observations ascribed to the effect of absinthe were not scientifically sound. Because the number of such studies is small, it is not possible in retrospect to clearly differentiate the effect of absinthe from that of alcohol (Hein et al., 2001; Olsen, 2000).

In the present study, the effects of a merely alcoholic beverage on attention performance and mood state were compared with beverages containing both alcohol and thujone. The difference of low and high thujone concentration was also tested. A thujone-alone group was not tested because thujone is only permitted to be used as an additive to food and beverages. Thujone is present in some alcoholic beverages, the most prominent being absinthe. The maximum level in the final product ready for consumption is 100 mg/l.

No significant changes were observed in attention performance when beverages with alcohol and low thujone concentration were consumed. Only high thujone concentrations led to significant effects. These were most prominent in test phase T0-T1. In test phase T0-T2, the observed changes were similar to those in the first test phase in all performance parameters but were less pronounced. In test phase T0-T1, the missing alteration in attention performance at low thujone concentrations can be explained by alcohol antagonizing the effect of thujone (Höld et al., 2000).

Consumption of alcohol or alcohol containing the low thujone concentration led to no significant decrease in attention performance. The number of correct reactions decreased significantly, however, at high thujone concentrations. Other studies have shown that people under the influence of alcohol were unable to direct their attention to central and peripheral stimuli simultaneously but tended to concentrate on stimuli in the central field of attention while neglecting signals in the peripheral field of perception (Strohbeck-Kuehner, 1998). These results coincide with those noted by Gustafson (1986), Lamb and Robertson (1987), Marks and Macavoy (1989), Patel (1988) and Strohbeck-Kuehner and Thieme (1998). However, these studies were carried out using higher alcohol concentrations than were used in the present study. The experimental arrangement of this study's attention performance test, however, permits even slight disturbances in attention performance at low alcohol concentrations to be observed. Performance was not reduced at the relatively low BAC values when alcohol alone was consumed. Since we found no significant effects between the treatments when comparing the performance differences from T0 to T1, we can assume that the effects of the high thujone condition are quantitative but not qualitative. Our observation that the performance was reduced under high thujone dosage is similar to the observations proposed in other studies. It can be concluded, therefore, that thujone influences attention performance. One possible explanation of this finding is that the effects of alcohol on attentional processing may be an inverted U-shaped function and that thujone simply shifts the dose-effect function to the left.

Although the effect of high thujone concentration had only a short-term effect on attention performance, the changes in the mood state persisted for a longer period or came to fruition only at a much later stage. The changes in the tested mood dimension were similar under all three experimental treatments, with the exception only of state anxiety. At T1 the subjects already judged themselves to be less anxious than at T0. This effect could not be observed at low thujone concentrations; only high thujone concentrations led to a decrease in state anxiety at T2. It is possible to explain the different experience of fear by an interaction of thujone and alcohol at the gammaaminobutyirc acid (GABA) receptor (Höld et al., 2001; Olsen, 2000) because α-thujone has been shown to act as a GABA type-A receptor antagonistic (Höld et al., 2000). GABA antagonists lead to an increase in fear sensations and also have a stimulating and rousing effect. The effects of α-thujone are opposed to those caused by GABA-en-hancing drugs such as ethanol, diazepam or phenobarbital, which are anxiolytic, sedative and amnesic.

The effect at the GABA receptor might also explain an elevated number of "false alarm" reactions when the subjects acted under higher thujone concentrations. In general, these kinds of reaction are signs of exaggerated motivation that can be expected mainly at a higher level of anxiety or fear. Thujone enhances the motivation to provide a good performance. Increased motivation and a simultaneous experience of anxiety or fear are not linked to better attention and thus better performance. The combination of motivation and fear leads to undirected action and thus to a higher number of "false alarm" reactions. The increase in state anxiety and intrapsychic equilibrium can also be attributed to the psychoactivating effect of thujone as observed by Ott (1993) and Pendell (1995) in cases of smoking dried wormwood. Huckenbeck (2001) and Arnold (1988) also reported elevated stimulation of the vegetative nervous system caused by thujone.

References

ALBERT-PULEO, M. Van Gogh's vision: Thujone intoxication. JAMA 246:

42, 1981. ARNOLD, W.N. Vincent van Gogh and the Thujone connection. JAMA

260: 3042-3044, 1988.

ARNOLD, W.N. Absinthe. Sci. Amer. 260 (6): 112-117, 1989.

BONKOVSKY, H.L., CABLE, E.E., CABLE, J.W., DONOHUE, S.E., WHITE, E.C., GREENE, Y.J., LAMBRECHT, R.W., SRIVASTAVA, K.K. AND ARNOLD, W.N. Porphyrogenic properties of the terpenes camphor, pinene, and thujone. Biochem. Pharmacol. 43: 2359-2368, 1992.

DEHAL, S.S. AND CROTEAU, R. Metabolism of monoterpenes: Specificity of the dehydrogenases responsible for the biosynthesis of camphor, 3 thujone, and 3-isothujone. Arch. Biochem. Biophys. 258: 287-291, 1987.

DEL CASTILLO, J., ANDERSON, M. AND RUBOTTOM, G.M. Marijuana, absinthe and the central nervous system. Nature 253: 365-366, 1975.

GUSTAFSON, R. Visual attentional span as a function of a small dose of alcohol. Percept. Motor Skills 63 (2 Pt 1): 367-370, 1986.

HEIN, J., LOBBEDEY, L. AND NEUMÄRKER, K.J. Absinth: Neue Mode, alte Probleme. (Absinthe: New fashion, old problem.) Dtsch. Ärztebl. 98 (B): 2334-2339, 2001.

HOBI, V. Basler Befindlichkeitsskala, Weinheim, Germany: Beltz Test Gesellschaft, 1985.

HÖLD, K.M., SIRISOMA, N.S. AND CASIDA, J.E. Detoxification of alpha- and beta-Thujones (the active ingredients of absinthe) Site specificity and species differences in cytochrome P450 oxidation in vitro and in vivo. Chem. Res. Toxicol. 14: 589-595, 2001.

HÖLD, K.M., SIRISOMA, N.S., IKEDA, T., NARAHASHI, T. AND CASIDA, J.E. Alpha-Thujone (the active component of absinthe): Gamma aminobutyric acid type A receptor modulation and metabolic detoxifi cation. Proc. Natl. Acad. Sci. 97: 3826-3831, 2000.

HUCKENBECK, W. Absinth: Ein neues Spielzeug für die Spassgesellschaft? (Absinthe: A new toy for the fun generation?) SeroNews: 84-92, 2001.

LAMB, M.R. AND ROBERTSON, L.C. Effect of acute alcohol on attention and the processing of hierarchical patterns. Alcsm Clin. Exp. Res. 11: 243 248, 1987.

LOFTUS, L.S. AND ARNOLD, W.N. Vincent van Gogh's illness: Acute inter mittent porphyria? Brit. Med. J. 303: 1589-1591, 1991.

MARKS, D.F. AND MACAVOY, M.G. Divided attention performance in can nabis users and non-users following alcohol and cannabis separately and in combination. Psychopharmacology 99: 397-401, 1989.

MONROE, R. Another diagnosis for Vincent van Gogh? (letter to the edi tor). J. Nerv. Ment. Dis. 179: 241, 1991.

MUELLER, A. Aufmerksamkeits-Prüf-Gerät (Attention-Performance-Test Handbook), Handanweisung, Göttingen, Germany, 1980.

OLSEN, R.W. Absinthe and gamma-aminobutyric acid receptors. Proc. Natl Acad. Sci. 97: 4417-4418, 2000.

OTT, J. Pharmacotheon: Entheogenic Drugs, Their Plant Sources and History, Occidental, CA: Natural Products, 1993.

PATEL, R.M. Ethanol's effect on human vigilance during a simple task in the presence of an auditory stressor. Psychol. Rep. 63: 363-366, 1988.

PENDELL, D. Pharmako/Poeia: Plant Powers, Poisons, and Herbcraft, San Francisco, CA: Mercury House, 1995.

STRANG, J., ARNOLD, W.N. AND PETERS, T. Absinthe: What's your poison? Though absinthe is intriguing, it is alcohol in general we should worry about. Brit. Med. J. 319: 1590-1592, 1999.

STROHBECK-KUEHNER, P. Alkoholinduzierte Aufmerksamkeitsstörungen: Auswirkungen von Alkoholgewöhnung und Aktivierung. (Alcohol in duced impairment of attention: The impact of high frequency of alco hol consumption and activation.) Blutalkohol 35: 434-446, 1998.

STROHBECK-KUEHNER, P. AND THIEME, C.M. Alkoholinduzierte Wahrneh mengsstörungen im Gesichtsfeld: Effekte und Artefakte. (Alcohol in duced impairments of perception in the visual field: Effects and artifacts.) Blutalkohol 35: 183-194, 1998.

URBAN, M. Voraussetzungen der Fahrtüchtigkeit bei depressiven Patienten (Conditions for Driving Ability of Depressive Patients), Bergisch Gladbach, Germany: Bundesanstalt für Strassenwesen, 1992.

VOGT, D.D. AND MONTAGNE, M. Ansinthe: Behind the emerald mask. Int.

J. Addict. 17: 1015-1029, 1982.

WEISBORD, S.D., SOULE, J.B. AND KIMMEL, P.L. Poison on line: Acute renal failure caused by oil of wormwood purchased through the internet. New Eng. J. Med. 337: 825-827, 1997.

WERTH, R. Funktionsbedingungen des Psychischen. In: PÖPPEL, E., BULLINGER, M. AND HÄRTEL, U. (Eds.) Medizinische Psychologie und Soziologie, New York: Chapman & Hall, 1990, pp. 128-147.

WIELAND-ECKELMANN, R. Kognition: Emotion und psychische Beanspruchung, Göttingen, Germany: Hogrefe, 1992.

ZUSCHLAG, B. AND JACOBSHAGEN, W. Einsatz des Aufmerksamkeits-prüf- gerätes (APG) bei Fahreignungsuntersuchungen. (Use of the attention- performance test in driving ability examinations.) Zeitschr. Verkehrssicherheit 28: 63-73, 1982.