Altered Emotional Perception in Alcoholics: Deficits in Affective Prosody Comprehension

Share Embed


Descripción

0145-6008/01/2503-0362$03.00/0 ALCOHOLISM: CLINICAL AND EXPERIMENTAL RESEARCH

Vol. 25, No. 3 March 2001

Altered Emotional Perception in Alcoholics: Deficits in Affective Prosody Comprehension Marilee Monnot, Sara Nixon, William Lovallo, and Elliott Ross

Background: Affective prosody is a nonlinguistic aspect of language that conveys emotion and attitude during discourse. It is a dominant function of the right hemisphere. Because skills associated with the right hemisphere have been found to be impaired in alcoholics, this study explored the possibility that affective prosodic functioning may be sensitive to the effects of alcohol due to heavy persistent drinking or prenatal exposure. Methods: Subjects were aged 25 to 58 years. Twenty-nine men and three women who met DSM-IV criteria for an alcohol use disorder with a median of 39 days of sobriety, 11 men with a probable history of fetal alcohol exposure (FAexp), and 41 age-matched control subjects of both sexes were tested by using the Aprosodia Battery. This instrument assesses affective prosodic comprehension (APC) across a range of verbal articulatory demands. Results: The alcoholic group scored 2 SD below the control mean, and the FAexp group scored ⫺5 SD regardless of whether they had ever been diagnosed with alcohol abuse. Despite their poor performance on APC, alcoholic and FAexp groups performed similarly to the control group on vocabulary, abstract reasoning, and an index of cognitive impairment that used the Shipley Institute of Living Scale. Multiple regression analyses that used nine alcohol use variables to model APC resulted in four significant contributors to the effect. These regressors were related to early exposure to ethanol and chronicity of alcohol abuse. Conclusions: Alcoholics and FAexp subjects were significantly less accurate at APC compared with controls. These alcohol-exposed subjects appear to be deficient in the ability to understand emotional valence in the speech of others, which results in errors of judgment that may impair social interactions. Key Words: Alcoholism, Affective Prosody, Fetal Alcohol Effects, Emotion.

A

N EXTENSIVE LITERATURE describes the neurological and psychological effects of alcohol abuse on the adult individual as well as the physiological, anatomical, and functional damage to offspring that results from alcohol abuse by the mother, called fetal alcohol syndrome (FAS) and fetal alcohol effects (FAE) (American Psychiatric Association, 1994; Jones et al., 1973; Mattson and Riley, 1999). Cognitive deficits and brain shrinkage in chronic heavy drinkers have been documented by neuropsychological assessments and imaging studies (Cala et al., From the Department of Neurology, College of Medicine, and Affective Communication Research Laboratory (MM, ER), and the Department of Psychiatry & Behavioral Sciences and the Behavioral Sciences Laboratories (WL), Veterans Affairs Medical Center, Oklahoma City, Oklahoma; and the Department of Psychiatry & Behavioral Sciences and the Center for Alcohol & Drug Related Studies (SN), University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma. Received for publication September 22, 2000; accepted December 1, 2000. Supported, in part, by a Merit Review Grant from the Medical Research Service of the Department of Veterans Affairs (to ER) and a grant from the Tharp Foundation, Nashville, Tennessee (to MM). Reprint requests: Marilee Monnot, PhD, University of Oklahoma Health Sciences Center, Department of Neurology, College of Medicine, 711 Stanton L. Young Blvd., Suite 215, Oklahoma City, OK 73104; Fax: 405-271-1078; E-mail: [email protected] Copyright © 2001 by the Research Society on Alcoholism. 362

1983; Holloway, 1994; Nixon and Phillips, 1999; OscarBerman, 1987; Parsons, 1998). However, emotional processing abilities in alcohol abusers have been studied less extensively (Philippot et al., 1999), and affective prosodic functioning in alcoholics largely has been overlooked, with a few notable exceptions (Oscar-Berman et al., 1990), despite their importance in social interaction and communication. It is recognized that detoxified alcoholics exhibit deficits in a wide range of neuropsychological domains. Most of the literature to date has focused on more traditional intellectual cognitive skills. However, other skills may be impaired in alcohol abusers. Right hemisphere visual spatial skills have been shown to be affected by alcohol abuse (Parsons and Nixon, 1998). Another “nontraditional” skill found to be altered by alcohol abuse is interpersonal problemsolving (Nixon et al., 1992). Because affective prosodic functioning, an essential element of social interaction, is localized to the right hemisphere (Heilman and Gilmore, 1998; Roberts, 1996; Ross, 2000; Ross and Mesulam, 1979; Ross et al., 1997), we thought that this cognitive skill also might be affected by alcohol abuse. There is evidence that prenatal exposure to alcohol affects not only general intellectual skills but also abilities related to appropriately interpreting social and affective Alcohol Clin Exp Res, Vol 25, No 3, 2001: pp 362–369

363

AFFECTIVE PROSODY FUNCTIONING IN ALCOHOLICS

stimuli (Streissguth, 1997). Children with some prenatal exposure to alcohol, but without the growth retardation, craniofacial anomalies, or microcephaly seen with FAS/ FAE diagnoses, still function similarly to FAS/FAE children on tests of intellectual functioning (Mattson and Riley, 1998; Mattson et al., 1997). In addition, adolescents and adults diagnosed with FAS/FAE show a wide range of physical, behavioral, social, and emotional dysfunction (Streissguth et al., 1991). To our knowledge, studies of adult detoxified alcoholics have not taken into account lower levels of fetal alcohol exposure (FAexp) that may result in more subtle or unusual cognitive impairments, such as altered affective prosodic functioning. These data call attention to a need for studies of affective prosodic comprehension (APC) in relation to alcohol abuse and prenatal exposure to alcohol. Prosody is a nonlinguistic feature of language that is embedded in the articulatory aspects of speech through acoustic features such as pitch, intonation patterns, stress, timing, rhythm, and differential pausing. It imparts information beyond that transmitted by word choice and grammar (Bolinger, 1980; Crystal, 1975; Kent and Read, 1992; Monrad-Krohn, 1963; Ross, 2000). Prosodic systems include affective prosody; linguistic prosody, which helps to clarify semantics and syntax; dialectical prosody, which underlies regional variations; and idiosyncratic prosody, which reveals personal variations (Monrad-Krohn, 1948; Ross, 2000). Affective prosody gives the listener information about the speaker’s emotional and attitudinal state and allows the speaker to convey the same. Many studies show that if the linguistic message is at odds with the affective prosodic content, the affective prosodic message usually takes precedence (Ackerman, 1983; Bolinger, 1972). Thus, the ability to modulate affective prosody is an essential element of face-to-face communication. Research suggests that modulation of affective prosody is a dominant and lateralized function of the right hemisphere (Bowers et al., 1993, 1987; Buchanan et al., 2000; Gorelick and Ross, 1987; Heilman, 1984; Heilman et al., 1975; Ross et al., 1997). When this ability is disrupted by focal brain lesions, social and emotional difficulties may ensue (Ross, 1982; Ross and Mesulam, 1979). Assessment of social/emotional judgment in alcoholics has yielded somewhat equivocal results. Cermak et al. (1989) employed verbal and nonverbal tasks to test right brain damaged patients, alcoholic subjects, and normal elderly controls. Alcoholics performed better than patients with brain damage but worse than elderly controls on two dimensions: (1) making correct inferences by integrating information and (2) humor appreciation. However, the alcoholics were equal to the elderly controls in identifying facial emotion, although both groups were superior to the lesioned patients. In a study that assessed the ability of Korsakoff patients, alcoholics, and normal subjects to identify facial emotion and emotional intonation in the voice (Oscar-Berman et al., 1990), alcoholics and normal subjects

were similar. However, emotional intonation tests were confounded with semantic tasks, which tap left hemisphere abilities. On the other hand, a recent publication showed significantly poorer performance by recently detoxified alcoholics on a facial emotion identification task (Philippot et al., 1999). Therefore, these results about cognitive processing in alcoholics leave questions unanswered about alcoholics’ ability to comprehend important social/emotional cues. The goal of this research project was to determine whether there is a defect in APC among those exposed to high levels of alcohol via prenatal exposure and/or chronic abuse. The second goal was to detect other aspects of behavior and substance abuse associated with APC. We predicted that alcohol abuse and/or prenatal exposure would disrupt affective prosodic functioning and that demographic, psychological, and substance use characteristics of subjects would be associated with this decrement. In particular, we thought that prenatal exposure and earlyonset alcohol abuse would impair this aspect of communication and social interaction.

METHODS Subject Data Subjects gave informed consent to participate in an Institutional Review Board-approved study of emotional expression in speech. Subjects ranged in age from 25 to 58 years, were either African American or white, and were all right-handed (Table 1). Detoxified alcoholics who met DSM-IV criteria for an alcohol use disorder (American Psychiatric Association, 1994), with at least 21 days of sobriety, were scheduled for testing. Exclusion criteria included serious neurological injury, such as a stroke or concussion, or neurological illness such as multiple sclerosis, epilepsy, or Parkinson’s disease. Other exclusions were (1) chronic illnesses such as cirrhosis, (2) diseases that significantly restrict oxygen to the brain such as emphysema or cardiopulmonary obstructive disease, and (3) serious psychiatric disorders such as axis I diagnoses, other than alcohol use disorders. Medical records were examined for evidence of neurological injuries or serious illnesses that are known to affect test results, such as schizophrenia (Ross et al., 2001); therefore, no formal psychiatric assessments were conducted. As a result of this review, one subject was removed from the final study group due to a closed head injury. Because prior studies that used the Aprosodia Battery excluded subjects with major depression, we used a measure of depressive symptoms to determine if this aspect of psychological functioning is correlated with APC. However, the scores from the Beck Depression Inventory (BDI) II were not used to exclude any subjects, which includes the four alcoholics who scored 45 or more on this instrument. Thirty-two alcoholics with no evidence of fetal alcohol exposure and 41 controls were included in the final project group. Also included were nine FAexp subjects who were recovering alcoholics and two FAexp subjects who reported no lifetime alcohol use disorder.

General Health and Alcohol Use Data A health questionnaire was used during telephone screening for inclusion/exclusion criteria. During the testing session, a structured interview instrument determined extent of substance abuse in volunteers and their extended families. Subjects were asked about details of their drinking behaviors within the last 6 months and their lifetime alcohol use. Subjects’ histories of treatment for abuse of 12 illegal substances used frequently in the United States were elicited. By using a standard genogram, we ques-

364

MONNOT ET AL. Table 1. Subject Characteristics

Sex (men/women) Ethnicity (African American/white) Age (range in years) Mean ⫾ SD Education (range of years completed) Mean ⫾ SD Beck Depression Inventory II (range of scores) Median score Alcoholic fathers (n) Percentage of group total Alcohol dependence chronicity (range in years) Median Abuse age onset (range in years) Median Age at first drunken episode (range in years) Median Substance abusers in family (range of percentages) Mean percentage Quantity-frequency index (range) Mean ⫾ SD Drug abuse (range of no. of drugs requiring treatment) Mean ⫾ SD Shipley Vocabulary (mean ⫾ SE) Shipley Abstraction (mean ⫾ SE) Shipley Conceptual Quotient (mean ⫾ SE)

Controls (n ⫽ 41)

Alcoholics (n ⫽ 32)

FAexp (n ⫽ 11)

19/22 6/35 25–63 43.5 ⫾ 9.5 12–18 14.1 ⫾ 1.8 0–24 7 10 50 N/A N/A N/A N/A 14–32 20 0–47 20 0–3.1 0.52 ⫾ 0.87 0–2

29/3 17/15 29–58 47.3 ⫾ 7.2 11–16 13 ⫾ 1.3 1–54 17 17 55 7–38 23.5 16–46 22 10–23 15.6 4–68 27 5.3–56.3 14.1 ⫾ 10.9 0–6

11/0 9/2 34–56 46.9 ⫾ 6.7 10–16 12.7 ⫾ 1.9 6–29 19 7 63 11–39 26 16–26 22 6–23 15 14–58 37 0–15.84 9.39 ⫾ 5.19 0–4

1.0 ⫾ 1.0 17.3 ⫾ 0.6 14.1 ⫾ 0.7 85.7 ⫾ 3.6

1.52 ⫾ 1.46 16.1 ⫾ 0.5 13.7 ⫾ 0.5 85.9 ⫾ 2.6

1.55 ⫾ 1.51 15.9 ⫾ 0.8 13.2 ⫾ 0.8 83.4 ⫾ 4.2

Median values were used when outliers distorted the distribution.

tioned subjects about the alcohol and substance use behaviors of all relatives known to them through the grandparent generation. Alcohol Use Variables Data Several variables were derived from the structured interview, from the Alcohol and Substance Use Questionnaire, and from medical records that detailed substance abuse assessment and treatment at the Oklahoma City Veterans Affairs Medical Center (Table 1). Thirty alcoholic subjects had received an alcohol dependence diagnosis, according to their medical records, before inclusion in this study, and they had at least 10 years of chronicity. Two alcoholic subjects had alcohol abuse diagnoses, with 7 and 9 years of chronicity. Other information included the following: 1. Mother’s alcohol use was derived from subject’s report of mother’s drinking behavior during pregnancy. 2. Father’s alcohol use was derived from subject’s recollection of father’s alcohol consumption and substance use habits. 3. Chronicity was the subject’s self-reported number of years of excessive drinking minus the number of years of abstinence between abuse periods; control subjects were scored as “0” because they were never in alcohol abuse treatment. 4. Abuse age onset was the self-reported earliest age at which control of drinking behavior was lost and regular, excessive intake of alcohol started; controls scored “0.” 5. First drunken episode was the self-reported age at subject’s first drunken episode, for all subjects. 6. Alcoholism in family was the percentage of relatives whom subject reported abusing alcohol and/or drugs, based on the number of relatives listed by subject. 7. QFI was the subject’s self-reported Quantity Frequency Index, the average number of ounces of absolute ethanol consumed per day, during the 6 months just before the current substance abuse treatment regimen. For control subjects, the 6 months before the testing session were assessed; lifetime history of drinking for both groups was estimated from subjects’ reports (Cahalan et al., 1969). 8. Last drink was the number of days since subject’s last drink of alcohol. 9. Drug abuse was the number of drugs for which subject reported receiving abuse treatment, at any time during his or her life.

Subjects with a QFI (Cahalan et al., 1969) that showed consumption of 5 oz or more of absolute alcohol per day were placed in the alcohol abuse group. Those consuming less than 5 oz of absolute alcohol per day were placed in the control group if they reported no history of alcoholism treatment or fetal alcohol exposure. During testing, careful clinical interviewing was applied to extract information about the subject’s mother’s history of alcohol consumption. With one exception, all subjects stated on the screening Health Questionnaire that their mothers did not abuse alcohol when pregnant with them. Thirty-two alcoholics reported data that made it very unlikely that their mothers abused alcohol at any time during the early childbearing years, although several subjects stated that their mothers developed alcoholism later in life. However, 11 subjects revealed strong presumptive evidence that their mothers abused alcohol when pregnant with the subjects. Behavioral outcomes used to justify maternal assignment to the FAexp group included concrete examples of alcohol abuse, which included the following: (1) mother’s loss of custody of infant due to conviction for “public drunkenness,” (2) reports from collateral relatives of mother’s generation that mother drank excessively during pregnancy, and (3) subject’s memory very early in life of mother’s frequent drunkenness. All subjects in this study were born before 1973, when data were first published in the United States about FAS (Jones et al., 1973) and before the general public knew of the dangers of drinking alcohol when pregnant. This major dependent variable, mother’s use of alcohol, was collected by retrospective subject report because all 11 subjects in this group stated that their mothers were deceased, demented, or out of contact with the subjects at the time of data collection.

Affective and Cognitive Status Data A self-report inventory and a cognitive assessment instrument administered during the single data collection session were the BDI II (Beck et al., 1996) and the Shipley Institute of Living Scale (SILS; Zachary, 1986). Conventional methods were used to compute subjects’ scores on the BDI and the SILS. Thus, four dependent variables were derived that included the total BDI score and three scores from the SILS: (1) Vocabulary, (2) Abstraction, and (3) Conceptual Quotient scale.

365

AFFECTIVE PROSODY FUNCTIONING IN ALCOHOLICS

Affective Prosody Data The Aprosodia Battery assesses affective prosodic functioning in adults (Ross et al., 1997). This test was devised to measure functioning in brain-damaged patients with aphasia, who are difficult to assess with language-based instruments. Therefore, decreasing levels of verbal articulatory demands are used to test both repetition of affective prosody and comprehension (identification and discrimination of conveyed emotion in speech). The testing protocol includes listening to compact disk recordings of three sets of sentences. The first comprehension subtest, the Word subtest, uses a completely articulated sentence, “I am going to the other movies,” which is repeated by using random variations of six emotions. These emotions are happy, sad, angry, surprised, disinterested or bored, and neutral. In addition, two stress patterns with emphasis on am and with emphasis on other test the subject’s ability to suppress attention to linguistic prosody cues. Each exemplar is presented twice, which results in 24 sentences in each subtest. Subjects listen to each sentence and then indicate which one of six emotions they hear; exemplars can be repeated at the subject’s request. Thereafter, 24 monosyllabic (“ba ba ba ba ba”) and 24 asyllabic (“aaaaahhhhh”) utterances with the same emotional and stress patterns also are presented to the subject. The fourth subtest, Discrimination, presents 24 pairs of low-pass filtered exemplars from the Word subtest. Subjects must tell if the two sentences use the same or different emotional categories. These sentences also employ different stress patterns, so the task tests the ability to discriminate between affective and linguistic prosody.

Testing Procedures After screening for psychiatric diagnoses and neurological injury or illness, subjects were scheduled for the 1.5 hr testing session. Subjects signed consent forms, filled out the self-report inventories, responded to questions from the structured interview about substance use, and were tested on the Aprosodia Battery. Subjects were compensated for their time and transportation expenses after the data collection session. Some subjects did not contribute all the data requested during the testing session, which resulted in differences in the degrees of freedom for some statistical tests.

Statistical Analyses The numbers of correct responses on each of the four subtests of the Aprosodia Battery were the primary dependent variables used in this study. Z scores were computed based on the control group mean and standard deviation. A fifth comprehension variable, T-Comp, was formed by using the total number of correct responses recorded from the subject from the Word, Monosyllabic, and Asyllabic subtests. For the purposes of the study, this variable is associated conceptually with the subject’s total comprehension performance across decreasing levels of verbal-articulatory demands. Both the JMP version of SAS (Sall and Lehman, 1996) and SPSS 8.0 for Windows (SPSS, Inc., Chicago, IL) statistical programs were used to analyze data. Variables reported in Tables 1 and 2 used a mean for normal distributions and median to more accurately describe sample characteristics when outliers distorted the distribution. Variables were assessed by using skewness figures (Zar, 1984) and the Shapiro-Wilk W normality test (Sall and Lehman, 1996) for their appropriateness in parametric analyses. Nonparametric tests used when variables did not show normal distribution were the Wilcoxon/Kruskal-Wallis tests (rank sums), median test (number of points above median), and Van der Waerden test (normal quantiles). Variance equality was assessed via the Barlett; the conservative Welch ANOVA was used to test means when the variances were unequal. When variance was equal, means comparisons were completed by using TukeyKramer honestly significant difference (HSD) test.

RESULTS

Group Demographics Group comparisons are shown in Table 1. No differences in age across groups were found, but there were significantly more women in the control group (Pearson ␹2 ⫽ 22.2, p ⬍ 0.0001), and controls were more educated [F(2,83) ⫽ 6.02, p ⬍ 0.004]. There were also significant differences between groups in ethnicity, and controls showed a lower proportion of African American subjects than either the alcoholic or FAexp group (Pearson ␹2 ⫽ 21.55, p ⬍ 0.000). However, sex, ethnicity, and years of education were not associated with Aprosodia Battery scores among controls; one-way ANOVA and regression analyses performed on each of these variables with the subtests resulted in probability figures of ⬎0.10 on all tasks. Therefore, inequities in subject groups for sex, ethnicity, and years of education were not likely to distort the results for the major dependent variables. Beck Depression Inventory II A one-way ANOVA that used the Tukey-Kramer HSD post hoc analysis showed that the control group scored significantly lower on the BDI than either the alcoholic or FAexp group [F(2,52) ⫽ 7.6, p ⬍ 0.002; see Table 1]. However, correlations between BDI and APC subtest scores were negligible (range of values for r ⫽ 0.09 to ⫺0.14), and regression analyses indicated no significant relationships between BDI scores and either the Aprosodia Battery subtests or the SILS (all Fs ⬍ 2.0; all ps ⬎ 0.17). Shipley Institute of Living Scales No significant differences between the groups were noted in the SILS scores on vocabulary, abstract reasoning, and conceptual quotient (Table 1). No relationships were found between the SILS and the Aprosodia Battery scores. Aprosodia Battery Scores Overall, controls were 93% accurate, alcoholics were 79% accurate, and FAexp subjects were 62% accurate in their attempts to correctly identify the emotion in all the recorded exemplars. These percentages and other Aprosodia Battery subtest values are summarized for the three groups in Table 2. The first research goal was to determine whether alcohol-exposed subjects have deficits in APC. To address this goal, we analyzed the variable T-Comp by using a one-way ANOVA (Fig. 1). A significant main effect was found [F(2,83) ⫽ 92.9, p ⬍ 0.000]. Means comparisons that used Tukey-Kramer HSD showed that the FAexp group (mean Z score ⫽ ⫺5.03) was significantly more impaired than the alcoholic group (mean Z score ⫽ ⫺1.9), which was more impaired than the control group (mean Z score ⫽ 0.0). Because only three alcoholic and no FAexp subjects were female, another ANOVA was performed that ex-

366

MONNOT ET AL. Table 2. Aprosodia Battery Scores Battery subscales Word (Z score range) Mean ⫾ SD Percentage correct out of 24 One-way ANOVA between groups Monosyllabic (Z score range) Mean ⫾ SD Percentage correct out of 24 One-way ANOVA between groups Asyllabic (Z score range) Mean ⫾ SD Percentage correct out of 24 One-way ANOVA between groups Discrimination (Z score range) Mean ⫾ SD Percentage correct out of 24 One-way ANOVA between groups

Controls (n ⫽ 41)

Alcoholics (n ⫽ 32)

FAexp (n ⫽ 11)

⫺2.3 to 1.1 0.0 ⫾ 1 93.3

⫺4.9 to 1.1 ⫺1.5 ⫾ 1.6 84.1

⫺6.3 to ⫺1.0 ⫺3.8 ⫾ 1.6 69.7

⫺2.9 to 1.6 0.0 ⫾ 1 89.8

⫺5.5 to 1.5 ⫺1.8 ⫾ 1.7 77.9

⫺5.5 to ⫺1.6 ⫺3.8 ⫾ 1.2 65.2

⫺2.3 to 1.3 0.0 ⫾ 1 86.7

⫺3.1 to 0.5 ⫺1.03 ⫾ 1.03 75.8

⫺6.2 to ⫺1.9 ⫺3.5 ⫾ 1.2 49.6

⫺2.7 to 1.4 0.0 ⫾ 1 90.1

⫺2.7 to 1.4 ⫺0.4 ⫾ 1.1 87.5

⫺2.7 to 0.9 ⫺1.4 ⫾ 0.9 79.9

F ratio (df ⫽ 2,83)

34.6, p ⬍ 0.0001

52.2, p ⬍ 0.0001

49.5, p ⬍ 0.0001

10.2, p ⬍ 0.0004

0.001]. Means comparisons showed that the FAexp group (mean Z score ⫽ ⫺1.4) was more impaired than both the alcoholic (mean Z score ⫽ ⫺0.36) and control (mean Z score ⫽ 0) groups, which did not differ significantly from each other. Adding this fourth subtest to T-Comp computations did not significantly reduce the impact of alcohol exposure on APC among the three groups in this study [F(2,83) ⫽ 82.7, p ⬍ 0.000] or change the means comparisons found for the original T-Comp variable. Thus the first research question was answered affirmatively: detoxified alcoholics and FAexp individuals had deficits in their ability to comprehend affective prosody. Alcohol Use Variables Fig. 1. Group differences in T-Comp Z scores (total affective prosodic comprehension); box plots are used to show group median as the heavy black line, and whiskers represent 95% of distribution.

cluded all women in the three groups, which included 22 control subjects. For this analysis, Z scores were computed by using men only. The results were still significant even though the total number of subjects was reduced from 84 to 59 [F(2,58) ⫽ 60.2, p ⬍ 0.000]. Also, as a further assessment of the effect, the group category of FAexp was eliminated. This was done by placing the nine FAexp subjects with alcohol abuse histories in the alcoholic group and the two FAexp subjects without abuse histories in the control group. Despite this change in subject categorization, the analysis was still significant [F(1,83) ⫽ 47.67, p ⬍ 0.000]. The Aprosodia Battery comprehension subtest, Discrimination, appears to assess somewhat different cognitive skills than the other three comprehension subtests. Discrimination taps attention and short-term memory skills as well as APC; subjects must listen to two sentences and judge whether they use the same or different affective prosody. A one-way ANOVA was used to assess differences between groups on the Discrimination subtest, and it showed a significant main effect [F(2,83) ⫽ 7.95, p ⫽

Alcohol use variables for each group are found in Table 1. We used ␹2 and one-way ANOVAs to assess group differences on these nine variables collected from the Alcohol and Substance Use Questionnaire. Significant differences were found between the three groups for (1) QFI [F(2,56) ⫽ 12.2, p ⬍ 0.000], (2) age at first drunken episode [F(2,56) ⫽ 8.0, p ⫽ 0.0009], and (3) percentage of alcoholics in family [F(1,41) ⫽ 5.62, p ⫽ 0.023]. No differences were noted between the two groups of 32 alcoholic and 9 FAexp detoxified subjects on the variables of “alcoholic fathers,” “days since last drink,” “chronicity,” or “abuse age onset” or the proportion of all subjects who reported other drug abuse in addition to alcohol [F(2,83) ⫽ 0.12, p ⬍ 0.88]. Also, drug use did not correlate to any Aprosodia Battery score despite the fact that 74% of alcoholic and FAexp subjects reported being treated for abuse of other drugs in addition to alcohol. Subjects with alcohol abuse histories reported a sobriety median of 39 days at the time of testing; no subject had less than 21 days. Relation of Alcohol Use Variables to Affective Prosodic Comprehension Standard least squares multiple regression analysis that used the substance use variables and one crossed variable (chronicity*abuse onset age) to model T-Comp produced

AFFECTIVE PROSODY FUNCTIONING IN ALCOHOLICS

five significant regressors [F(5,37) ⫽ 27.3, R2 adjusted ⫽ 0.78, p ⬍ 0.000]. These substance use variables were the most powerful combination in predicting overall APC accuracy in 32 alcoholic and 9 FAexp individuals with alcohol abuse histories. The five regressors and their F ratios with p values, plus ␤ weights, are as follows: (1) age at first drunken episode ⫽ 11.9, p ⬍ 0.0016, ␤ ⫽ 0.29; (2) alcohol abuse chronicity ⫽ 15.6, p ⬍ 0.0004, ␤ ⫽ 1.35; (3) abuse onset age ⫽ 18.6, p ⬍ 0.0001, ␤ ⫽ 0.84; (4) chronicity*abuse age ⫽ 20.7, p ⬍ 0.0001, ␤ ⫽ ⫺1.29; and (5) alcohol use by mother ⫽ 63.3, p ⬍ 0.0001, ␤ ⫽ 0.64. Thus, the second research question was answered: Deficits in APC were predicted only by four alcohol use variables plus one crossed variable and by none of the demographic variables. DISCUSSION

Detoxified alcoholics and FAexp individuals demonstrated significant deficits in their ability to identify emotion in the voices of others, a skill that is essential during discourse for appropriate social interaction. The variables that predicted APC with the greatest power all were associated with early intense alcohol exposure as well as sustained abuse. Several demographic and psychological variables appear inconsequential to APC in alcohol-exposed subjects and controls. Age up to 63 years, sex, education, and ethnicity did not predict APC in the control group. Also, BDI scores did not correlate with APC scores. Only four alcoholic subjects scored at a level similar to clinically depressed patients, so this result may have arisen from a restricted range on the BDI. However, in this sample there appears to be no relationship between the ability to accurately identify emotion in the voice and self-reported depressive symptoms. Therefore, APC appears to be a highly conserved skill necessary for survival in a gregarious species and is not affected by standard demographic variables or a range of self-reported depression symptoms. In fact, the ability to perceive this aspect of communication has been documented in infants (Monnot, 1999; Scherer and Oshinsky, 1977) and is one of the earliest substrates for language acquisition activated by verbal stimulation and social interaction. The production and comprehension of affective prosody, as a cognitive and behavioral system, are clearly in place by the age of 10 years (Doherty et al., 1999; Van Der Meulen et al., 1997). Thus, the comprehension abilities assessed in this study are fundamental skills, and the Aprosodia Battery was devised so that even subjects with severe communication disabilities due to aphasia could be tested. In natural discourse, processing of affective prosody must be accomplished continuously while many other aspects of language, communication, and social interaction are assessed. Therefore, the deficits noted here in alcoholexposed subjects suggest a significant functional impairment that may impact not only personal and work relation-

367

ships but the effectiveness of therapeutic interventions such as individual, group, and family therapies. The alcohol use variables that predicted impairment in APC were related to both early exposure and sustained alcohol abuse. These regressors, listed in ascending order according to their impact on the effect, are (1) age at first drunken episode, (2) alcohol abuse chronicity, (3) abuse onset age, (4) chronicity*abuse age, and (5) alcohol use by mother. This suggests that fetal or early exposure to the teratogenic effects of alcohol is the most destructive environmental factor of the alcohol exposure variables tested here. Possible mechanisms of alcohol toxicity and its effects on the brain are suggested by work that used animal models. Intake of alcohol by 7-day old rat pups that produced a blood ethanol level of ⬎200 mg/dl for four consecutive hours significantly increased neuronal apoptosis (Ikonomidou et al., 2000). This result also was observed in the fetal brain after alcohol ingestion by the pregnant dam. Results correlated positively with the rate at which ethanol was delivered and how long the blood level was maintained at or above toxic threshold. Vulnerability to ethanolinduced apoptosis is associated with the developmental stage of synaptogenesis (Ikonomidou et al., 2000). In humans, synaptogenesis extends from the third trimester of pregnancy to several years thereafter (Goodlett and West, 1992), including adolescence (Barone et al., 2000; Chugani, 1998), which suggests that the developing human brain may be similarly vulnerable. Data reported here are not sufficient to determine whether the APC deficits found in the FAexp group can be attributed to the pervasive brain dysfunctions found in those diagnosed with FAS or FAE. Individuals in the FAexp group did not appear to have the physical characteristics of FAS, and there was no mention of this diagnosis in any medical record, although these subjects were born before FAS was a recognized medical syndrome. Also, all three groups performed as low average intelligence as estimated by the SILS, whereas FAS adults usually perform at a lower level (Streissguth et al., 1991). In addition, the two FAexp subjects without histories of alcohol abuse or treatment were as impaired as the nine FAexp subjects with alcohol abuse diagnoses (t test ⫽ ⫺1.23, p ⬎ 0.25). Therefore, fetal alcohol exposure is a suggested explanation for the very severe deficits in APC found in these 11 subjects. Although this group was defined only by subjects’ reports of their mothers’ drinking behavior without corroborative evidence, the strength of the statistical association suggests that further work should be done with persons of known FAS/FAE histories. Although the results were statistically robust, this was a preliminary examination of APC in primarily male alcoholexposed subjects. These results should be viewed with caution. Because this study used only Veterans Affairs Medical Center patients and employees, a sample that is more representative of the general public, especially one that

368

MONNOT ET AL.

included more women, would allow these results to be generalized. Also, further research is needed to justify the use of the putative category of FAexp. In addition, the use of more comprehensive instruments to measure cognitive functioning could answer the question raised here about whether alcohol exposure is responsible for a specific brain dysfunction: poor ability to comprehend emotion in the voice. Further research should include collateral contacts to confirm family reports of drinking behaviors, more detailed neuropsychological testing to determine extent of deficits attributable to alcohol abuse, physiological assessments of brain activities during perception of affective prosody, and a more representative sample of subjects. Individuals with poor ability to accurately detect emotion in the voice of a communication partner will find it difficult to manage social demands. These include personal relationships and work-related activities such as job interviews, telephone communication, and supervision. Those who can only detect a person’s affect accurately 50% to 70% of the time will make many mistakes during social exchanges. Individuals who acquire such a deficit early in life may never develop the requisite communication, social, and occupational skills necessary for success. This study indicates that early exposure to alcohol, especially through the mechanism of prenatal exposure and early-onset alcohol abuse, causes significant deficits in APC that are likely to impair social competency.

ACKNOWLEDGMENTS We thank William Beatty, PhD, for the concept of testing detoxified alcoholics for affective-prosodic comprehension and production. We also thank the subjects who volunteered for this study; they have helped to advance our understanding of the effects of alcohol on the brain.

REFERENCES American Psychiatric Association (1994) Diagnostic and Statistical Manual of Mental Disorders. 4th ed. American Psychiatric Association, Washington, DC. Ackerman B (1983) Form and function in children’s understanding of ironic utterances. J Exp Child Psychol 35:487–508. Barone SJ, Das KP, Lassiter TL, White LD (2000) Vulnerable processes of nervous system development: A review of markers and methods. Neurotoxicology 21:15–36. Beck AT, Steer RA, Ball R, Ranieri WF (1996) Comparison of Beck Depression Inventories–IA and –II in psychiatric outpatients. J Pers Assess 67:588 –597. Bolinger D (1972) Intonation. Penguin, Hardmondsworth, England. Bolinger D (1980) Language: The Loaded Weapon. Longman Group, London. Bowers D, Bauer RM, Heilman KM (1993) The nonverbal affect lexicon: Theoretical perspectives from neuropsychological studies of affect perception. Neuropsychology 7:433– 444. Bowers D, Coslet HB, Bauer RM, Speedie LJ, Heilman KM (1987) Comprehension of emotional prosody following unilateral hemispheric lesions: Processing defect versus distraction defect. Neuropsychologia 25:317–328.

Buchanan TW, Lutz K, Mirzazade S, Specht K, Shah NJ, Zilles K, Jancke L (2000) Recognition of emotional prosody and verbal components of spoken language: An fMRI study. Brain Res Cogn Brain Res 9:227–238. Cahalan U, Cisin I, Crossely HM (1969) American Drinking Practices. Rutgers Center for Alcohol Studies, New Brunswick, NJ. Cala LA, Jones B, Burns P, Davis RE, Stenhouse N, Mastaglia FL (1983) Results of computerized tomography, psychometric testing and dietary studies in social drinkers, with emphasis on reversibility after abstinence. Med J Aust 2:264 –269. Cermak L, Verfaellie M, Letourneau L, Blackford S, Weiss S, Numan B (1989) Verbal and nonverbal right hemisphere processing by chronic alcoholics. Alcohol Clin Exp Res 13:611– 616. Chugani HT (1998) A critical period of brain development: Studies of cerebral glucose utilization with PET. Prev Med 27:184 –188. Crystal D (1975) The English Tone of Voice. St. Martin’s Press, New York. Doherty CP, Fitzsimons M, Asenbauer B, Staunton H (1999) Discrimination of prosody and music by normal children. Eur J Neurol 6:221–226. Goodlett CR, West JR (1992) Maternal substance abuse and the developing nervous system, in Maternal Substance Abuse and the Developing Nervous System (Zagon I, Slotkin T eds), pp 45–75. Academic Press, San Diego, CA. Gorelick PB, Ross ED (1987) The aprosodias: Further functionalanatomic evidence for the organization of affective language in the right hemisphere. J Neurol Neurosurg Psychiatry 50:553–560. Heilman K (1984) Comprehension of affective and nonaffective prosody. Neurology 34:917–921. Heilman KM, Gilmore RL (1998) Cortical influences in emotion. J Clin Neurophysiol 15:409 – 423. Heilman KM, Scholes R, Watson RT (1975) Auditory affective agnosia: Disturbed comprehension of affective speech. J Neurol Neurosurg Psychiatry 38:69 –72. Holloway F (1994) Low-Dose Alcohol Effects on Human Behavior and Performance: A Review of Post-1984 Research. Office of Aviation Medicine, FAA, Oklahoma City, OK. Ikonomidou C, Bittigau P, Ishimaru MJ, Wozniak DF, Koch C, Genz D, Price MT, Stefovska V, Horster F, Tenkova T, Dikranian K, Olney JW (2000) Ethanol-induced apoptotic neurodegeneration and fetal alcohol syndrome. Science (Wash. DC) 287:1056 –1060. Jones KL, Smith DW, Ulleland CN, Streissguth AP (1973) Pattern of malformation in offspring of chronic alcoholic mothers. Lancet 1:1267– 1271. Kent RD, Read C (1992) The Acoustic Analysis of Speech. Singular Publishing Group, San Diego, CA. Mattson SN, Riley EP (1998) A review of the neurobehavioral deficits in children with fetal alcohol syndrome or prenatal exposure to alcohol. Alcohol Clin Exp Res 22:270 –294. Mattson SN, Riley EP (1999) Implicit and explicit memory functioning in children with heavy prenatal alcohol exposure. J Int Neuropsychol Soc 5:462– 471. Mattson SN, Riley EP, Gramling L, Elis DC, Jones KL (1997) Heavy prenatal alcohol exposure with or without physical features of fetal alcohol syndrome leads to IQ deficits. J Pediatr 131:718 –721. Monnot M (1999) Function of infant-directed speech. Human Nature: An Interdisciplinary Biosocial Perspective 10:415– 443. Aldine de Gruyter, Hawthorne, NY. Monrad-Krohn GH (1948) Dysprosody or altered “melody of language.” Brain 70:405– 415. Monrad-Krohn GH (1963) The third element of speech: Prosody and its disorders, in Problems in Dynamic Neurology (Halpern L ed), pp 101– 118. Hebrew University Press, Jerusalem, Israel. Nixon SJ, Phillips JA (1999) Neurocognitive deficits and recovery in chronic alcohol abuse. CNS Spectrums Int J Neuropsych Med 4:95–108. Nixon SJ, Tivis R, Parsons OA (1992) Interpersonal problem-solving in male and female alcoholics. Alcohol Clin Exp Res 16:684 – 687. Oscar-Berman M (1987) Neuropsychological consequences of alcohol abuse: Questions, hypotheses and models, in Neuropsychology of Alco-

AFFECTIVE PROSODY FUNCTIONING IN ALCOHOLICS

holism: Implications for Diagnosis and Treatment (Parsons O, Butters N, Nathan P eds), pp 256 –269. Guilford, New York. Oscar-Berman M, Hancock M, Mildworf B, Hutner N, Weber DA (1990) Emotional perception and memory in alcoholism and aging. Alcohol Clin Exp Res 14:383–393. Parsons OA (1998) Neurocognitive deficits in alcoholics and social drinkers: A continuum? Alcohol Clin Exp Res 22:954 –961. Parsons OA, Nixon SJ (1998) Cognitive functioning in sober social drinkers: A review of the research since 1986. J Stud Alcohol 59:180 –190. Philippot P, Kornreich C, Blairy S, Baert I, Dulk AD, Bon OL, Streel E, Hess U, Pelc I, Verbanck P (1999) Alcoholics’ deficits in the decoding of emotional facial expression. Alcohol Clin Exp Res 23:1031–1038. Roberts V (1996) Prosody impairment and associated affective and behavioral disturbances in Alzheimer’s disease. Neurology 47:1482–1488. Ross ED (1982) The divided self. The Sciences 22:8 –12. Ross ED (2000) Affective prosody and the aprosodias, in Principles of Behavioral and Cognitive Neurology (Mesulam MM ed), pp 316 –331. Oxford University Press, New York. Ross ED, Orbelo D, Cartwright J, Hansel S, Burgard M, Testa J, Buck R (2001) Affective prosodic deficits in schizophrenia. J Neurol Neurosurg Psychiatry, in press.

369

Ross ED, Mesulam MM (1979) Dominant language functions of the right hemisphere? Prosody and emotional gesturing. Arch Neurol 36:144 –148. Ross ED, Thompson RD, Yenkosky J (1997) Lateralization of affective prosody in brain and the callosal integration of hemispheric language functions. Brain Lang 56:27–54. Sall J, Lehman A (1996) JMP Start Statistics (SAS Institute ed). Duxbury Press, Belmont, CA. Scherer KR, Oshinsky JS (1977) Cue utilization in emotion attributes from auditory stimuli. Motivation and Emotion 1:331–346. Streissguth A (1997) Fetal Alcohol Syndrome. Paul H. Brookes, Baltimore, MD. Streissguth AP, Aase JM, Clarren SK, Randels SP, LaDue RA, Smith DF (1991) Fetal alcohol syndrome in adolescents and adults. JAMA 265: 1961–1967. Van Der Meulen S, Janssen P, Den Os E (1997) Prosodic abilities in children with specific language impairment. J Commun Disord 30:155– 169. Zachary RA (1986) Shipley Institute of Living Scale, Revised Manual. Western Psychological Services, Los Angeles. Zar JH (1984) Biostatistical Analysis. Prentice-Hall, London.

Lihat lebih banyak...

Comentarios

Copyright © 2017 DATOSPDF Inc.