Font and font size: Times New Roman, 12 point.
The Relationship Between Intentional Self-Injurious Behavior and the Loudness Dependence of Auditory Evoked Potential in Research Volunteers
Angelika Marsic,1 Mitchell E. Berman,2 Tammy D. Barry,1 and Michael S. McCloskey3
1The University of Southern Mississippi 2Mississippi State University 3Temple University
Objective: Serotonergic (5-HT) functioning has been shown to be inversely associated with inten- tional self-injurious behaviors. The purpose of this study was to examine the association between three related self-report measures of intentional self-injurious behaviors (suicidal thoughts/behavior, history of nonsuicidal self-injury, history of severe self-harm when angry) and a putative electrophysiological index of 5-HT activity, the loudness dependence of auditory evoked potential (LDAEP). Method: Auditory evoked potentials were recorded from 41 men (mean age = 20.69, standard deviation [SD] = 2.98) during the administration of various tone loudness stimuli, followed by completion of the self-report measures. Results: The component slope was associated with all measures of self-injurious behavior in the expected direction. Conclusion: The LDAEP has the potential to be used as a noninvasive index of intentional self-harm disposition. Additional studies are needed using other popu- lations, including women and treatment-seeking individuals, to determine if the LDAEP more broadly discriminates risk of self-injuring. C© 2014 Wiley Periodicals, Inc. J. Clin. Psychol. 71:250–257, 2015.
Keywords: Self-injury; 5-HT functioning; LDAEP; N1/P2; EEG
The prevalence rate for intentional self-injurious behavior (SIB) ranges from 1% to 4% in adults (Briere & Gil, 1998; Klonsky, Oltmanns, & Turkheimer, 2003; Prinstein, 2008) and from 17% to 38% in college students, with lifetime prevalence estimates of 35% (e.g., Gratz, 2001; Whitlock, Eckenrode, & Silverman, 2006). Not unexpectedly, the rates of SIB are even higher in clinical populations (21%–61% in adolescents and young adults and 21% in adults; e.g., Briere & Gil, 1998; Darche, 1990; DiClemente, Ponton, & Hartley, 1991; Prinstein, 2008). The most severe form of SIB—suicide—is the second leading cause of death among 25- to 34-year-olds, the third leading cause of death among 15- to 24-year-olds, and the 11th leading cause of death overall in the United States (Centers for Disease Control and Prevention [CDC], 2007). There are approximately 100–200 attempts for every completed suicide among young adults aged 15–24 years (Goldsmith, Pellmar, Kleinman, & Bunney, 2002).
One of the most extensively studied biological correlates of SIB is serotonergic (5-HT) neu- rotransmitter activity. Specifically, attenuated 5-HT functioning has been associated with SIBs across the spectrum of lethality (e.g., Arango et al., 1990; Audenaert et al., 2001; Kamali, Oquendo, & Mann, 2001; Malone, Corbitt, Li, & Mann, 1996; McCloskey, Ben-Zeev, Lee, Berman, & Coccaro, 2009). For example, a meta-analysis by Lester (1995) reviewed 27 neuro- chemical studies of the association between 5-HT and SIB involving 1202 psychiatric patients and controls. The results provided strong evidence for the role of serotonin in suicidal be- havior. Individuals who had attempted suicide had lower levels of cerebral spinal fluid (CSF) 5-hydroxyindoleacetic acid (5-HIAA; a 5-HT metabolite) compared to psychiatric controls. As- berg (1997) reviewed 33 studies and found that low levels of CSF 5-HIAA were associated with
Please address correspondence to: Angelika Marsic, 3264 Willowbrook Avenue, Palmdale, CA 93551. E-mail: Angelika.Marsic@gmail.com
JOURNAL OF CLINICAL PSYCHOLOGY, Vol. 71(3), 250–257 (2015) C© 2014 Wiley Periodicals, Inc. Published online in Wiley Online Library (wileyonlinelibrary.com/journal/jclp). DOI: 10.1002/jclp.22136
Self-Injurious Behavior and the LDAEP 251
suicidality in unipolar depression and personality disorders. Diminished levels of CSF 5-HIAA have also been found in depressed patients with a high-lethality suicide attempt compared to depressed individuals with a low-lethality suicide attempt (Mann & Malone, 1997). Further- more, lower CSF 5-HIAA levels have been found in individuals engaging in nonlethal SIBs (López-Ibor, Saiz-Ruiz, & Pérez de los Cobos, 1985).
Commonly used biological indexes of central 5-HT functioning can be costly and invasive (e.g., lumbar puncture; pharmacochallenge). However, a neurophysiological approach that takes advantage of electrical brain wave activity measured at the scalp (the loudness dependence of the auditory evoked potential: LDAEP; Hegerl & Juckel, 1993) may provide a noninvasive means to assess behaviorally relevant central 5-HT functioning. Although auditory evoked potentials are generated by a complex interrelationship of different neurotransmitters, there is mounting evidence that the LDAEP is most likely modulated by serotonergic system activities (Hegerl & Juckel, 1993). The LDAEP is a measure of auditory cortex activity as represented by the auditory evoked potential slopes (Hegerl, Gallinat, & Juckel, 2001). The intensity dependence of the auditory evoked N1/P2 component (i.e., dB level of the tone dependence) has been proposed to be inversely related to central serotonergic activity. That is, low serotonergic innervation of the auditory cortex ostensibly produces a more pronounced LDAEP N1/P2 component (i.e., increased N1/P2 amplitude to increasing intensity tones) and vice versa (Hegerl & Juckel, 1993).
In humans, the N1/P2 component comprises two overlapping subcomponents generated by the superior temporal plane (mainly primary auditory cortex) and the lateral temporal gyri (secondary auditory cortex; Hegerl & Juckel, 1993). The N1/P2 component, occurring about 70–200 ms poststimulus, is used as a combined ratio parameter because it has higher loudness dependence reliability than when loudness dependence is measured separately for N1 and P2.
In addition, the relationship with clinical features and personality factors is stronger with the loudness dependence of the combined parameter than with individual amplitudes (Hegerl, Gallinat, & Mrowinski, 1994). The N1/P2 component also exhibits prominent and stable in- terindividual differences. For example, Hegerl, Prochno, Ulrich, and Muller-Oerlinghausen (1988) found test-retest reliability of .77 for the Cz site (i.e., midline position of the central lobe) and .74 for the amplitude/stimulus intensity function (ASF) slope among healthy partic- ipants. ASF reflects the N1/P2 amplitude changes as the tone intensity increase. Hegerl and Juckel (1993) reported a test-retest correlation of .90 for the intraindividual stability of the in- tensity dependence N1/P2 component, mainly generated by the activity of the primary auditory cortex.
The experimental evidence for a relationship between the LDAEP and 5-HT was first observed in animals (Hegerl et al., 1993; Juckel, Molnar, Hegerl, Csepe, & Karmos, 1997). However, experimental studies in humans (i.e., using pharmacological agents to augment serotonin levels in the brain) of the LDAEP as an index of acute 5-HT changes have yielded mixed results. For example, 5-HT augmentation in a double-blind, placebo-controlled study was shown to produce a significant decrease in N1/P2 slope with increasing tone loudness, lending support for the validity of the LDAEP as a 5-HT index (Nathan, Segrave, Phan, O’Neill, & Croft, 2006). Follow-up studies with healthy participants have failed to replicate these findings (Guille et al., 2008; Uhl et al., 2006), indicating that, at least in healthy subjects, the LDAEP may not be a good indicator of acute changes in central 5-HT activity.
The lack of evidence for acute changes in 5-HT activity as a function of pharmacochallenge with 5-HT agents does not preclude the use of the LDAEP as a valid biological indicator in vulnerable individuals. Several clinical studies have found a strong LDAEP in individuals characterized by psychiatric disorders ostensibly marked by 5-HT dysfunction. For example, Gallinat, Bottlender, and Juckel (2000) found that a significantly higher number of depressive patients fell into a strong LDAEP group (seemingly reflecting attenuated 5-HT activity), and that those same individuals exhibited a significant decrease in depressive symptoms after a selective serotonin reuptake inhibitor (SSRI) treatment compared to depressive patients with a less prominent LDAEP.
Hegerl and colleagues (1998) found that patients with high levels of serotonin syndrome (i.e., enhanced central 5-HT activity) exhibited a weaker LDAEP than those with low serotonin syndrome. Chen and colleagues (2005) found a sharper LDAEP slope in a depression–suicide
252 Journal of Clinical Psychology, March 2015
group as opposed to a depression–nonsuicidal group, demonstrating a potential utility of the LDAEP in discriminating suicidality among depressed individuals (Chen et al., 2005). Based on these findings, O’Neill, Croft, and Nathan (2008) concluded that although evidence for the LDAEP as an indicator of acute serotonergic changes among humans is conflicting in nature, evidence for the LDAEP as a useful biological index of 5-HT functioning in vulnerable individuals is more compelling.
The purpose of this study was to examine the relationship between intentional SIBs and the LDAEP in a sample of young adults. If the LDAEP reflects relatively stable central 5-HT activity, an association between the LDAEP slope and measures of SIBs should emerge. To date no study has examined the relationship between the LDAEP N1/P2 slope and intentional SIBs in a nonclinical population. Given that previous studies have found that an increase in the auditory evoked N1/P2 component slope with increasing tone loudness (i.e., strong LDAEP) is inversely related to indexes of central serotonergic activity, it was expected that strong LDAEP would be positively related to various measures of SIB.
Method
Participants
Forty-one men recruited from undergraduate classes took part in the study. The majority of the participants self-identified as Caucasian (56.1%), followed by African American (39%) and “other” (4.9%) race or ethnicity. Participants ranged in age from 18 to 31 years (mean [M] = 20.69, SD = 2.98). A history of schizophrenia or mood, anxiety, or substance dependence disorder was exclusionary. In addition, any hearing impairment, a history of seizures, and a history of traumatic brain injury were exclusionary. Potential participants were excluded if they were currently under medical treatment. Participants were asked to not consume alcohol or caffeinated beverages in the 24 hours before the study day. The study was reviewed and approved by the Institutional Review Board for the Protection of Human Subjects.
Measures
Health Screening Questionnaire. A brief health screening questionnaire was created for the current study, including items on the participant’s age, gender, and race. Along with this demographic information, items addressing the health and psychiatric exclusion criteria listed above were included.
Suicidal Behaviors Questionnaire (SBQ; Cole, 1988). The SBQ is a four-item self- report measure that assesses suicidal thoughts, plans, and behavior. The SBQ questions are as follows: “Have you ever thought about or attempted to kill yourself?”; “How often have you thought about killing yourself in the past year?”; “Have you ever told someone that you were going to commit suicide, or that you might do it?”; “How likely is it that you will attempt suicide one day?” Items are rated on a Likert-format scale, with values ranging from 0–6, 0–4, 0–2, and 0–4, respectively. Scores range from 0 to 16 (with higher scores implying greater suicidal disposition). The SBQ has adequate internal consistency (α = .80) for a nonclinical sample and good test-retest stability over time (r = .95; Cotton, Peters, & Range, 1995).
Furthermore, the SBQ has good construct validity, as shown by a significant positive corre- lation (r = .69) with the Scale for Suicidal Ideation in a nonclinical sample (Cotton et al., 1995) and with laboratory measures of self-aggression (Berman & Walley, 2003). Internal consistency for the current sample was adequate (α = .71).
Deliberate Self-Harm Inventory (DSHI; Gratz, 2001). The DSHI is a 17-question, self-report scale of nonsuicidal SIBs. The DSHI comprises NSSI behaviors that do not have the goal of ending one’s life (e.g., self-cutting, burning, scratching, biting, and punching); items include, for example, “Have you ever intentionally (i.e., on purpose) carved words into your skin?” and “Have you ever intentionally (i.e., on purpose) used bleach, comet, or oven cleaner to
Self-Injurious Behavior and the LDAEP 253
scrub your skin?” Individuals endorse Yes or No for each item. A DSHI total score is obtained by summing the number of endorsed self-harm behaviors. The DSHI has shown adequate internal consistency (α = .82) and test-retest stability (r = .92; Gratz, 2001). Adequate correlations with related self-report measures of self-harm behaviors have been found (e.g., DSHI and the self-harm items on the Mental Health History Form, r = .49; Gratz, 2001). DSHI internal consistency for the current sample was also adequate (α = .84).
The Life History of Aggression Scale-Self-Aggression subscale (LHA-SA; Coc- caro, Berman, & Kavoussi, 1997). The LHA is an 11-item measure of past aggressive, self-aggressive, and antisocial behaviors. The LHA assesses the frequency and intensity of these behaviors, rather than aggressive traits or ideation, and it provides information about these from age 13 on. For the current study, we used the self-report two-item Self-Aggression subscale of the LHA (e.g., “Deliberately tried to physically hurt yourself in anger or desperation” and “Deliberately tried to end your life or kill yourself in anger or desperation”), which is rated on a 6-point scale, ranging from 0 (no occurrences) to 5 (more events than can be counted), reflecting the total number of occurrences. The LHA-SA was found in previous studies to have somewhat low internal consistency (α = .45) due to gender differences (females, α = .71; males, α = .18) but had adequate inter-rater agreement (r = .84) and test-retest reliability (r = .97; Coccaro et al., 1997). LHA-SA internal consistency for the men in this study, however, was adequate (α = .72).
LDAEP. LDAEP stimulus presentation, data acquisition, and analyses were accomplished using equipment and software obtained from the James Long Company, a 16-channel custom optically isolated bioamplifier. LDAEPs were recorded with 15 electrodes arranged according to the 10–20 electroencephalogram (EEG) electrode system, using M1 as a reference and AFz as ground. Impedances were kept below 5 kΩ throughout the testing. Pure sinus tones (1000 Hz, some with 100 ms duration with 10 ms rise and 10 ms fall time, and some with some with 50 ms duration with 10 ms rise and 10 ms fall time, inter stimulus interval (ISI) randomized between 1800 and 2200 ms) of five intensities (60, 70, 80, 90, 100 dB) were presented biaurally in a pseudorandomized form by headphones.
Data were collected with a sampling rate of 500 Hz and an analogous bandpass filter (0.16– 50 Hz). Seventy sweeps of each stimulus intensity and time duration were presented (700 sweeps in all, with 350 sweeps of 50 ms tone duration, and 350 sweeps of 100 ms tone duration). Poststimulus peak latencies were determined between 80–120 ms for N1 and 150–230 ms for P2 components.
Procedure
Upon arrival, participants completed the informed consent process, after which a brief screening interview was administered. If the participant did not meet any exclusionary criteria, then he was instructed to complete the demographic questionnaire, which was computer administered. Next, the participant was prepared for the EEG recording. An appropriately sized electrocap compris- ing 15 electrodes, following the International 10–20 system, was fitted on the participant’s head. The scalp was prepared by application of a mildly abrasive gel (OmniPrep). Electrooculography (EOG) electrodes were placed on the outer canthi of each eye and on the supraorbital and infraorbital ridge of the left eye, to allow for detection and removal of ocular artifacts.
According to lab standards, each electrode site displayed impedance of less than 5 kΩ, while the impedance on the EOG sites were kept at less than 10 kΩ. The left mastoid electrode site was used as a reference site during the collection phase. However, during the analysis, the right mastoid was averaged with the left mastoid to serve as the final reference to avoid the left or right hemisphere bias that is often found when using just one reference site (Luck, 2005).
The participant was instructed to refrain from moving his eyes during testing to ensure mini- mal contamination of the data. Specifically, a fixation point was displayed on the screen for the duration of the EEG experiment, and the participant was asked to softly focus on that point and refrain from any eye movement other than regular blinking. In addition, the participant was asked to refrain from making any body movements. After the EEG data collection, the
254 Journal of Clinical Psychology, March 2015
participant completed the self-report measures of SIB (SBQ, SHI, LHA-SA). Finally, the par- ticipant was debriefed and psychology course research credit was applied.
Results
