Abstract
Social phobics have an increased risk of alcoholism. The mechanism behind this co-morbidity is not well understood. According to the appraisal–disruption model [Sayette, M. A. (1993). An appraisal–disruption model of alcohol's effects on stress responses in social drinkers. Psychological Bulletin, 114, 459–476], alcohol disrupts appraisal of threat stimuli unless the stimuli are easy to process. We investigated whether alcohol alters the judgment of emotional facial expressions in social phobics and controls. We also tested the judgment of emotionally ambiguous faces which should be more difficult to process.
Forty social phobics and controls rated faces depicting five emotional expressions on an animosity rating scale. For two ambiguous facial expressions, angry, respectively, happy faces were blended with neutral faces. Half of the participants consumed alcohol.
Socially phobic participants rated neutral and happy facial expressions as less friendly than controls, irrespective of alcohol consumption. In both groups, consuming alcohol reduced the perceived rejection of angry faces.
In line with current theories of social phobia, patients interpreted neutral facial expressions as more rejecting than controls. The rejection perceived in explicitly angry facial expressions was less after drinking alcohol. This reduction of the adversity of socially threatening stimuli by alcohol might act as negative reinforcement and thus contributes to alcohol problems.
. Results
2.1. Blood–alcohol concentration
Social phobic and control participants as well as male and female subjects who had received alcohol reached comparable levels of blood–alcohol concentration (Fig. 1). Repeated measures analyses with group and sex as between subjects factor and time as repeated measures factor revealed only a significant main effect for time (F(2, 40) = 975.82, p < 0.01). Contrasts indicated a significant increase from measurement 1–2 (F(1, 40) = 1033.4, p < 0.01) and 2–3 (F(1, 40) = 124.77, p < 0.01). Thus, for all groups, BAC increased from the first to the second, and from the second to the third measurement similarly.
Blood–alcohol concentration in participants that received alcohol. Note: Time 1: ...
Fig. 1.
Blood–alcohol concentration in participants that received alcohol. Note: Time 1: blood–alcohol concentration at baseline; Time 2: blood–alcohol concentration five minutes after last drink; Time 3: blood–alcohol concentration after self report task.
Figure options
2.2. Self-report data
Analysis of self-report data with group (social phobics versus controls) and beverage (orange juice versus alcohol) as between subject factor and stimulus (explicitly angry, ambiguously angry, neutral, ambiguously happy, explicitly happy) as within subject factor resulted in two main effects for stimulus (F(4, 80) = 226.00, p < 0.001; View the MathML sourceηp2 = 0.93) and group (F(1, 80) = 4.38, p = 0.04; View the MathML sourceηp2 = 0.06) and an interaction effect for stimulus × beverage (F(4, 80) = 2.57, p = 0.045; View the MathML sourceηp2 = 0.13). Furthermore, a small trend for stimulus × group was detected (F(4, 80) = 2.04, p = 0.097; View the MathML sourceηp2 = 0.101).
A subsequent analysis of the effects for each emotional facial expression with group and beverage as between factor resulted in a main effect for beverage for the explicitly angry faces (F(1, 80) = 6.36, p = 0.014; View the MathML sourceηp2 = 0.08), indicating lower perceived rejection in the alcohol condition, for both groups alike (compare Fig. 2 and Table 2). Thus, alcohol seems to reduce perceived threat of explicitly angry facial expressions irrespective of the clinical status of the participants. For explicitly happy and neutral faces, the analysis resulted in a main effect for group (explicitly happy: (F(1, 80) = 4.62, p = 0.035; View the MathML sourceηp2 = 0.06); neutral: (F(1, 80) = 5.12, p = 0.025; View the MathML sourceηp2 = 0.06)), indicating higher values of perceived threat in social phobics than in controls. Accordingly, patients and controls differed significantly in how they perceived neutral and happy emotional facial expressions, and this bias is not affected by consumption of alcohol.
Perceived friendliness vs. rejection ratings of social phobics and controls for ...
Fig. 2.
Perceived friendliness vs. rejection ratings of social phobics and controls for emotional facial expressions. Note: EA: explicitly angry faces, AA: ambiguously angry faces, N: neutral faces; AH: ambiguously happy faces, explicitly happy faces. Bars are standard errors (S.E.). EFE: emotional facial expression. ×: significant group effects; *significant alcohol effects
Figure options
Table 2.
Self-report ratings for social phobic and controls
UA AA N AH UH
SP-O 7.98 (0.15) 6.24 (0.16) 5.43 (0.19) 3.84 (0.24) 2.98 (0.20)
SP-A 7.39 (0.21) 6.12 (0.24) 5.58 (0.21) 4.18 (0.23) 3.20 (0.21)
C-O 7.65 (0.19) 5.97 (0.19) 5.16 (0.17) 3.93 (0.19) 2.36 (0.13)
C-A 7.22 (0.24) 5.71 (0.17) 5.06 (0.14) 4.13 (0.19) 2.81 (0.12)
Note: SP-O: social phobics orange juice condition; SP-A: social phobics alcohol condition; C-O: controls orange juice condition; C-A: controls alcohol condition; UA: unambiguously angry; AA: ambiguously angry; N: neutral; AH: ambiguously happy; Unambiguously happy. Values enclosed in parentheses represent S.E.
Table options
Analysis of correlations between ratings of emotional facial expressions did yield significant results only for social phobics. Higher measures of general anxiety sensitivity and social anxiety predicted negative ratings for the explicitly angry faces (rSPS = 0.47, p = 0.037; rASI = 0.52, p = 0.019). These associations disappeared in the alcohol condition. No significant correlations for the other facial expressions were found.
