Literature Review

Journal of Personality Disorders, 6(4), 301-312, 1992

VALIDATION OF PERSONALITY DISORDER

ASSESSMENT BY BIOLOGIC AND

FAMILY STUDIES

Larry J. Siever, MD, Robert L. Trestman, PhD, MD,

Jeremy M. Silverman, PhD

ASSESSMENT ISSUES FOR BIOLOGIC AND FAMILIAL VALIDATORS

OF PERSONALITY

Although personality assessment measures must be reliable, internally con

sistent, and psychometrically sound, and must correspond to meaningful clinical entities, validation of these measures ultimately depends on ex

ternal validators outside the domain of the clinical personality traits them

selves. There are a variety of validators that may be employed for a clinical

diagnostic category in psychiatry; these include family history, laboratory tests, treatment response, and long-term outcome. In this paper, illustra

tions of validating evidence for personality disorder diagnoses and criteria

are presented utilizing both biologic and family study validators.

In order for assessments to be meaningful, the personality measures

themselves must be reliable, as must be the validating measures utilized.

Categorical assessments of personality disorder depend on the criteria util

ized. Thus, external validators may be used to assess the validity of specific criteria sets. Dimensional approaches may also be used in personality assessment and may be more accurate reflections of how personality dis

order traits are distributed in clinical samples. Whereas categorical com

parisons involve group analyses, dimensional assessments may depend on

correlational analyses between the continuously distributed traits and

specific validators. In the study of personality disorder, it has been argued that such a dimensional approach may have greater validity (Widiger et al. ,

1986). Dimensional approaches that have been widely utilized include the

five-factor model (Costa & McCrae, 1990), Eysenck personality dimensions

as derived from the Eysenck Personality Questionnaire (EPQ Eysenck &

Eysenck, 1975), and Cloninger's tridimensional scheme as operationalized

by the Tri-dimensional Personality Questionnaire (TPQ Cloninger, 1987).

From the Department of Psychiatry. Veterans Affairs Medical Center, Bronx. NY (L.J.S.); and

the Department of Psychiatry, Mount Sinai School of Medicine, New York, (R.L.T., J.M.S.).

Address correspondence to Larry J. Siever, MD, Department of Psychiatry (1 16A), Bronx VA

Medical Center, 130 West Kingsbridge Rd., Bronx. NY 10468.

An earlier version of this paper was presented at the NIMH Workshop on Assessment of

Personality Disorders, March 1991. New York.

301

302 SIEVER, TRESTMAN, AND SILVERMAN

Other more specific tests have been utilized to assess dimensions of per

sonality related to psychosis proneness (e.g., the Chapman psychosis-

proneness scales Chapman et al., 1976; Chapman & Chapman, 1987), or

impulsivity/aggressiveness (e.g., the Barratt Impulsiveness Scale BIS, Bar-

ratt, 1959; or Buss Durkee Hostility Inventory BDH, Buss & Durkee,

1957).

The use of such scales will depend also on the characteristics of the

samples measured. Personality traits as assessed in "normal" populations

may have different distributions and construct validity than those assessed

in clinical samples of, for example, personality disorder patients. Other

samples that might be used for such studies include volunteers from specif ic populations (e.g., college students) selected by virtue of psychological

profiles or, in the "biologic high-risk' paradigm (Buchsbaum, Coursey, &

Murphy, 1976), selected by virtue of extremes on specific biologic measures.

Similarly, for studies utilizing familial aggregation as a defining variable,

relatives of patients with specific psychiatric disorders may also provide a

validating sample.

FAMILIAL VALIDATORS OF PERSONALITY DISORDERS

Another useful external validation of a personality disorder entity is family

history. Other methods of external validation discussed here for example,

biologic or treatment response variables generally refer to direct measures

of the patient that potentially anchor the criteria for a given disorder to

other factors specifically related to the patient. In contrast, family history data assessment (in addition to indicating a possible familial/genetic com

ponent for a disorder) can be used to examine the integrity or coherence of

an entity through the investigation of its occurrence in relatives with a

related environmental/genetic background.

Although the family history method lacks the sensitivity of the more

direct family study methodology (Andreasen et al., 1977, 1986), it has been

repeatedly demonstrated that useful diagnostic information about the first-

degree relatives of probands can be obtained from knowledgeable family informants, provided that operationally defined categories are specified and

a structured, or semistructured, interview format is employed to ensure

standardized assessment. Furthermore, the utility of this method extends

beyond the ability to make diagnoses relating to DSM-III or III-R Axis I

categories to include the identification of personality disorder traits in

relatives of probands of interest (Siever, Silverman et al., 1990b).

Of course, not all personality disorders will possess familial factors and

this need not to be a requirement for the personality disorder nosology any more than it is for any other type of disorder. Many personality disorders,

and psychiatric disorders in general, are likely to be coherent entities with

specific identifiable etiologies unrelated to familial factors.

Familial factors, however, have been identified for several personality disorders and their further specification can then aid in the validation of

these categories in a number of different ways. First, the identification of

familial factors may help to show familial/genetic connections between dis

orders, providing evidence for a familial spectrum or a cluster of related

BIOLOGIC AND FAMILY STUDIES IN ASSESSMENT 303

disorders. Second, examination of the relatives of probands can aid in the

identification of independent dimensions of personality functioning that

may be found in specific combinations in different personality disorder

categories. Third, the identification of specific family factors in subgroups ofprobands with a disorder may help to identify more homogeneous groups that in turn may lead to the development of more meaningful diagnostic criteria.

The family history investigations of schizotypal personality disorder fea

tures in relatives of schizophrenic and schizotypal probands have added

to the evidence offered by adoption studies for a familial/genetic relation

between schizophrenia and schizotypal personality disorder. Family his

tory criteria for schizophrenia-related personality disorder traits, derived

from DSM-III criteria for schizotypal personality disorder, have been de

veloped (Kendler et al., 1987), and individuals meeting these criteria were

significantly more prevalent among the first-degree relatives of schizo

phrenic probands compared to those of medical controls. These results

offered further evidence for a familial/genetic spectrum of schizophrenia disorders that included relatives with schizotypal personality disorder

characteristics. More recently, family history studies of schizophrenia- related personality disorder traits have been conducted with schizoty

pal and other personality disorder probands. Similar to the relatives of

schizophrenic probands, the relatives of schizotypal personality disor

der probands showed an increased morbid risk for schizophrenia-related

personality disorder traits compared to controls (Siever, Silverman et

al., 1990b).

The family history method can also be used to investigate whether par ticular personality disorder traits run in families. If so, such traits may, in

turn, reflect an underlying familial transmission of personality disorder

related temperamental factors (Siever, Klar, & Coccaro, 1985; also see Akis-

kal, this issue). For example, previous investigations ofborderline personal

ity disorder have reported increases in borderline-related traits, not more

specifically delineated, among the relatives of borderline probands (Baron et

al., 1985; Links, Steiner, & Huxley, 1988; Loranger, Oldham, &Tulis, 1982;

Pope et al., 1983; Zanarini et al., 1988). To investigate whether the di

mensional components that together make up the constellation of border

line traits may be independent familial features, a recent family history

investigation of borderline personality disorder developed independent

family history diagnostic criteria for two hallmark borderline personality disorder traits, affective instability and impulsivity. The increased morbid

risk for both of these categories in the first-degree relatives of borderline

personality disorders compared to the relatives of other clearly non-

borderline personality disorder probands provided evidence that these

traits were indeed familial (Silverman, Pinkham et al., in press). Further

more, these traits were found to be independent of one another in the

relatives of probands; no more relatives than would be expected by chance

were found to have both affective and impulsive personality disorder traits

even though both of these categories were more common in the relatives of

borderline probands compared to controls. Thus, although the criteria for

borderline personality disorder virtually by definition will tend to identify individuals who present with both affective and impulsive personality dis-

304 SIEVER, TRESTMAN, AND SILVERMAN

order symptoms, the symptoms themselves may represent two relatively

distinct dimensions with independent lines of familial transmission.

Improved characterization of the specific familial component (e.g.,

elucidating precisely what personality disorder traits run in families and in

which group of probands) may also help to tease apart previously un

recognized heterogeneous disorders. Recent work with schizotypal and bor

derline personality disorders provides one example of the way family history assessment and analysis can be used to improve the characterization of

valid personality disorder categories. The current criteria for schizotypal

personality disorder were developed to identify individuals with both psy chotic-like and social deficit features, and were not designed to specifically

identify individuals who may share with schizophrenic patients the same

underlying genetic trait. Since schizotypal characteristics ( 1 ) are commonly found in relatives with borderline personality disorder (Zanarini et al.,

1990), a category with no familial relation to schizophrenia, and (2) have

been found in excess among the biological relatives of major affective dis

order probands and schizophrenic probands (Ingraham, 1989; Squires- Wheeler et al., 1988), current criteria for schizotypal personality may well

lack specificity for a genetically based schizophrenia spectrum. Considera

tion of the role of affective instability in independently leading to psychotic- like symptoms has led to the hypothesis that noneffective schizotypal

personality disorder characteristics may better identify a disorder with a

more specific genetic relation to schizophrenia (Siever, Bernstein, & Silver

man, 1991a).

This hypothesis has recently been examined using family history assess

ments of randomly ascertained schizophrenic and personality disorder

probands in which both schizophrenia-related and affective personality disorder traits were assessed (Silverman, Siever et al. 1991). The comorbid

ity of these two groups of personality disorder traits has been assessed in

relatives of different proband groups. A larger proportion of at-risk relatives

of borderline probands, including those probands who also met criteria for

schizotypal personality disorder as well as those who did not, had comorbid

schizophrenia-related and affective personality disorder traits. Alternative

ly, a larger proportion of at-risk relatives of schizophrenic and non-

borderline schizotypal personality disorder probands were found to have

schizophrenia-related personality disorder traits without comorbid affective

personality disorder traits. Data such as these can valuably contribute to

the development of a more valid diagnostic entity. For example, the data

support the possibility that a more restrictive definition of schizotypal

personality disorder might identify a more homogeneous, schizophrenia- related personality disorder. Toward this goal, new criteria for schizotypal

personality disorder have been proposed for DSM-IV (Siever, Bernstein, &

Silverman, 1991a), in which features specifically related to affective symp toms do not contribute to the diagnosis. Such criteria may help reduce the

substantial overlap currently present with borderline personality disorder

and may more specifically identify a disorder with a familial/genetic relation

to schizophrenia. In using biologic variables as external validators, interactions between

different biologic systems may provide the basis for different personality tendencies; this possibility must be considered in the development of realis-

BIOLOGIC AND FAMILY STUDIES IN ASSESSMENT 305

tic models of biologic substrates of these disorders. Thus, validators from

one biologic domain, such as a specific neurotransmitter system, may be

more informative if the role of other relevant biologic systems is considered

as weD. These considerations are illustrated in the following examples.

BIOLOGIC VALIDATORS OF IMPULSIVE/AGGRESSIVE TRAITS

Impulsive/aggressive traits are characteristic of a variety of personality dis

orders, all of which are in the "dramatic cluster of DSM-III-R personality disorders. The self-destructive actions, impulsive behavior (e.g., reckless

driving, shoplifting, binge eating), and the inappropriate intense anger

(e.g., recurrent physical fights or displays of temper) of the borderline

patient all may be manifestations of an underlying tendency toward im

pulsive aggression. However, irritability and aggressiveness are also criteria

for antisocial personality disorder, as is impulsivity (e.g., traveling from

place to place or recklessness involving one's personal safety). Even the

readiness to react to criticism with feelings of rage may be indicative of an

underlying impulsivity in some patients with narcissistic personality dis

order.

A variety of self-assessment questionnaires also reflect traits of impulsive

aggression including, for example, the BIS (Barratt, 1959), BDHI (Buss &

Durkee 1957), Zuckerman Sensation Seeking Scale (Zuckerman, 1979),

and the TPQ (Cloninger, 1987). (See Table 1.) As may be noted in Table 1,

some of the subscales for each of these inventories focus on behavior

reflecting motor disinhibition/aggression whereas others relate more closely to stimulus-seeking/environmental reactivity. Still other characteristics

such as resentment or suspiciousness in the BDHI may not be centrally related to the construct of impulsive aggression. These distinctions may become important in utilizing such scales in

relation to biologic validators, as the biologic substrates of these different

aspects of impulsive aggression may be different, although interactive. For

example, the serotonergic system has been particularly implicated in the

suppression of aggressive or punished behaviors (Olivier et al. , 1987; Sou-

brie, 1989) such that, for example, lesions ofthe serotonergic system in rats

lead to unrestrained killing of mice in their vicinity (Valzelli, 1981). The

noradrenergic system, in contrast, seems to mediate the organism's re

sponse to its environment through arousal following exposure to novel or

threatening stimuli (Aston-Jones & Bloom 1981; Devauges & Sara 1990;

Foote, Bloom, & Aston-Jones, 1983). Thus, these systems are logical choices to examine for potential biological components of impulsive/

aggressive behavior. A combination of heightened reactivity to the envi

ronment leading to excessive irritability and overreactivity, coupled with

disinhibition of aggressive responses to potential threats, may act syn-

ergistically to enhance the likelihood of aggressive behaviors. Indeed, in

animal studies, it has been shown that an intact noradrenergic system is

required for the expression of aggressive behaviors when the serotonergic

system is lesioned (Hodge & Butcher, 1975). These systems, then, are likely candidates for evaluation in the clinical setting in relation to impulsive

aggression.

306 SIEVER, TRESTMAN, AND SILVERMAN

Table 1. Some Self-Report Scales for the Assessment of Impulsivity/Aggression

Barratt Impulsiveness Scale0

Interpersonal Behavior

Motor Impulsivity Risk Taking Self-Assessment

Sensory Stimulation

BussDurkee Hostility Inventory"

Assault

Indirect Hostility

Irritability

Negativism Resentment

Suspicion Verbal Hostility

Guilt

Tri-dimensional Personality Questionnaire'

Harm Avoidance

Novelty Seeking Reward Dependence

Zuckerman Sensation Seeking Scaled

Thrill and Adventure Seeking

Experience Seeking Disinhibition

Boredom Susceptibility

"Barratt, 1959.

bBuss & Durkee, 1957.

cCIoninger, 1987.

dZuckerman, 1979.

With regard to serotonergic function, the prolactin response to the

serotonergic reuptake blocker fenfluramine has been shown to depend on

activity of serotonergic neurons in the hypothalamopituitary axis (Coccaro,

Siever, et al., 1989). Blunted prolactin responses to fenfluramine have been

found in both depressed patients and patients with personality disorder

(Coccaro, Siever et al., 1989; Lopez-Ibor, Lana, 6k Saiz-ruiz, 1990; Siever,

Murphy, et al., 1984). Another serotonergic measure, cerebrospinal fluid

(CSF) concentrations of the serotonin metabolite 5-hydroxyindoleacetic acid (5-HIAA), has been found to be reduced in depressed patients compared to normal controls; the relationship is particularly strong in association

with suicide attempts, often believed to be the outcome of an impulsive action (Asberg et al., 1987; Stanley & Stanley, 1990). Reductions in CSF

5-HIAA have also been associated with impulsive/aggressive behavior (e.g.,

unprovoked or easily provoked verbal and physical assaults) in criminal

offenders and personality disorder patients (Brown, Ebert et al., 1982;

Brown & Linnoila, 1990). These findings raise the question as to whether

reductions in serotonergic functions are specifically related to depression,

suicide attempts, or impulsivity. If related to impulsivity, serotonergic in

dices might therefore constitute an external validator for this trait in some

personality-disordered patients and provide a tool for developing a more

valid nosologic system to diagnose personality disorders.

A study in our laboratory utilized the prolactin response to fenfluramine

BIOLOGIC AND FAMILY STUDIES IN ASSESSMENT 307

as an index of serotonergic activity in both affective and personality dis

order patients in order to evaluate the degree of association between

serotonergic dysfunction and depression, impulsivity, and/or suicide at

tempts (Coccaro, Siever et al. , 1989). Blunted prolactin responses to fenflur

amine were found in both the affective and personality disorder patients. In

the depressed patients, prolactin responses to fenfluramine were blunted in

both acute and remitted depressed patients, suggesting that the

serotonergic dysfunction was not just a concomitant of the state of depres sion but may represent a trait or vulnerability marker in those depressed

patients with blunted responses. However, there was no association with

severity of depression in acutely depressed patients. Furthermore, blunted

prolactin responses to fenfluramine were observed in the cohort ofpersonal

ity disorder patients, who primarily met DSM-III Axis II diagnoses of schizo

typal, paranoid, borderline, and/or histrionic. Whether the patient had a

current or past episode of major depressive disorder had no effect on this

relationship. Thus, the serotonergic dysfunction, although present in de

pressed patients, did not seem to be closely associated with depression per se. Furthermore, blunted prolactin responses to fenfluramine were more

prevalent in suicide attempters in both the depressed and personality cohorts than in non-suicide attempters, suggesting a closer association

with suicide attempt, an act often considered a self-directed form of im

pulsive aggression. In the personality disorder patients, blunted prolactin responses to fen

fluramine were found more frequently in borderline personality disorder

patients compared to patients with other non-borderline personality disor

ders. The blunted prolactin responses to fenfluramine were particularly associated with the diagnostic criteria of angry outbursts, impulsiveness, and self-destructive behavior, but not with affective instability, identity diffusion, or any of the remaining criteria for borderline personality dis

order.

These results suggest that the categorical difference between borderline

and other personality disorder patients was a function of the increased

impulsivity characteristic of these patients. Indeed, significant correlations

were found between inventories of aggression including the interview-based

BrownGoodwin Assessment for History of Lifetime Aggression (Brown,

Goodwin et al., 1979), the Assault and Irritability subscales of the BDHI,

and the Motor Impulsivity subscale of the BIS. In a larger series of patients

extending the initial study, similar correlations were observed between

blunted prolactin responses to fenfluramine and measures of impulsivity/

aggression (Trestman, Lawrence et al., 1991).

These studies suggest that the reduced serotonergic function may be

associated with a disinhibition of aggressive behavior and they are among the first studies demonstrating an association between a personality dis

order trait and a biologic abnormality. These results support the validity of a

dimensional approach, as correlations with impulsivity were observed

across personality disorder patients (in a cohort whose diagnoses were

primarily schizotypal, paranoid, borderline, and/or histrionic). These re

sults also indicate that borderline personality disorder in particular may be a disorder potentially characterized by diminished serotonergic func

tion, although the sample size of antisocial personality disorder patients

308 SIEVER, TRESTMAN, AND SILVERMAN

was too small to adequately test the specificity of this association. It may

be that specificity in the categorization of personality disorders derives

from the interaction of more than one biologic predisposition with en

vironmental factors that influence the expression of these predisposi tions.

Interestingly, there were no significant correlations between impulsivity and prolactin responses to fenfluramine in remitted depressed patients

(Coccaro, Siever et al., 1989). Although indices suggesting reduced

serotonergic function have been associated with suicide attempts, which

might be considered self-directed aggression, in both personality disorder

and depressed patients, the association with externally oriented aggression in serotonergic function is significant only for the personality disorder

patients. Thus, factors other than serotonergic function itself must mediate

the external expression of aggression. As the noradrenergic system is in

volved in arousal and engagement with the environment, it was of interest

to evaluate the activity of this system in depressed and personality disorder

patients. One method to measure the reactivity of this system is to assess

the responsiveness of hypothalamic noradrenergic receptors. The growth hormone (GH) response to clonidine is mediated by postsynaptic

hypothalamic a2 noradrenergic receptors and has frequently been used to

gauge the responsiveness of the noradrenergic system (e.g., Abelson et al.,

1991; Trestman 6k Siever, 1991). GH responses to clonidine that are in

creased in comparison to normal controls may be hypothesized to be related

to increased reactivity or engagement with the environment, whereas de

creased responses might reflect a disengagement from the environment.

Indeed, blunted growth hormone responses to clonidine have been demon

strated in not only acutely depressed patients but also depressed patients who are in remission (Siever, Trestman, et al., in press-b). It is possible that

noradrenergic dysfunction is related to the vegetative disturbances of this

disorder, that is, disturbances of sleep/wakefulness, goal-directed behavior,

engagement with environment (Siever, 1987). In contrast, personality dis

order patients demonstrated normal to increased growth hormone re

sponses to clonidine (Coccaro, Lawrence et al., In press; Lawrence et al., in

press). The magnitude of the growth hormone response to clonidine in

personality disorder patients was correlated with indices of impulsive/

aggression, particularly with the Risk Taking subscale of the BIS and the

Irritability subscale of the BDHI (p<,05). Correlations were also observed

with the Zuckerman Sensation Seeking Scale and the Harm Avoidance scale

(negative correlation, p<,05) and Reward Dependence scale (p<,05) of the

TPQ. These results raise the possibility that increased receptor-mediated

responses to the adrenergic agonist clonidine are associated with increased

responsiveness and engagement with environment. As studies have shown

that an intact noradrenergic system is required for aggression to occur

following serotonergic lesions (Hodge & Butcher, 1975), these clinical re

sults also suggest that where noradrenergic receptor sensitivity is reduced, as in depression, externally oriented aggression is less likely to occur.

Conversely, when adrenergic receptor function is increased, the likelihood

of increased irritability may contribute to the impulsive aggression observed. Thus, these biologic validators may identify two aspects of the

tendency toward impulsive aggression that may jointly contribute to ex-

BIOLOGIC AND FAMILY STUDIES IN ASSESSMENT 309

ternally directed aggression. These results may have important implications for how we conceptualize impulsive behavior and its biologic substrates in

personality disorder patients.

THE DOPAMINERGIC SYSTEM IN SCHIZOTYPAL PERSONALITY

DISORDER

The dopamine system has been implicated in the etiology of the psychotic

symptoms of schizophrenia since the discovery that neuroleptic treatments

reduce psychosis in these patients. Plasma concentrations of homovanillic

acid (HVA), the major metabolite of dopamine, have been used to assess

dopaminergic activity in humans. While plasma HVA may derive largely from peripheral sources, relationships with clinical symptomatology and

central nervous system variables raise the possibility that variance in plas ma HVA may in part reflect variance in brain dopamine. In any case, plasma HVA concentrations correlate with degree of psychotic-like symptomatology in schizophrenic patients, and baseline elevations of plasma HVA predict a

better response to neuroleptic treatments (Davis et al., 1985; Davidson 6k

Davis, 1988; Pickar et al., 1986).

Since schizotypal personality disorder is phenomenologically, genetically, and biologically related to chronic schizophrenia, it was of interest to study

plasma HVA concentrations in schizotypal patients compared to other per

sonality disorder patients and normal controls, and to determine whether

psychotic-like criteria would be particularly associated with elevations in

plasma HVA. Plasma HVA concentrations were determined at 10 a.m. in

unmedicated personality disorder patients ( 1 1 schizotypal patients and 7

patients with personality disorders other than schizotypal, schizoid, and

paranoid) following three days of a low monoamine diet and after an over

night fast. Concentrations of plasma HVA were increased in the schizotypal

personality disorder compared to the normal controls (p<.05), whereas

patients with other non-schizophrenia-related personality disorders were

not significantly different from the normal controls (p<.05). In the per

sonality disorder sample, psychotic-like symptoms correlated significantly with plasma HVA concentration, whereas other non-psychotic-like criteria

for schizotypal personality disorder did not (Siever, Amin et al., 1991b).

A similar pattern of results was observed utilizing cerebrospinal fluid

(CSF) determinations of HVA. These schizotypal patients had higher CSF

HVA concentrations than other personality disorders and the concentra

tions of CSF HVA correlated significantly with the psychotic-like criteria

of the disorder, but not with other criteria unrelated to psychotic-like

symptomatology (Siever, Amin et al., in press-a). In contrast, deficit symp toms such as social isolation were associated with impaired performance of attentional/information processing tasks (Keefe et al., 1989; Siever,

1991; Siever, Keefe, et al., 1990a). Thus, these biologic measures sug

gest the possibility of disentangling two dimensions of schizophrenia- related personality disorder traits and converge with results of genetic studies that independently suggest heritable dimensions of "positive' psy chotic-like symptoms and "negative" or deficit-like symptoms (Kendler,

Ochs et al., 1991).

310 SIEVER, TRESTMAN, AND SILVERMAN

CONCLUSIONS

These studies illustrate the potential utility of employing biologic and family

study validators to further our understanding of personality disorders and

may allow for more therapeutically meaningful nosologic distinctions and

dimensions to be identified. Further, the domains of validation for per

sonality assessment must be chosen with regard to the ultimate application of the diagnostic categories. Thus, for example, (1) personality assessment

might be targeted at identifying populations responsive to a particular

pharmacologic agent, or (2) genetic counseling might be validated by biolog ic and familial techniques. However, it is now clear that personality assess

ment must move beyond simple considerations of reliability or statistical

(factor) analyses; external validation promises to provide a tool with which

to enhance the therapeutic utility of personality disorder diagnoses.

REFERENCES

Abelson, J. L.. Glitz, D., Cameron, O. G..

Lee, M. A., Bronzo, M., & Curtis, G. C.

(1991) Blunted growth hormone re

sponse to clonidine in patients with

generalized anxiety disorder. Archives of General Psychiatry. 48. 157-162.

Andreasen, N. C. Endicott, J., Spitzer. R.

L., et al. (1977). The family history method using diagnostic criteria:

Reliability and validity. Archives of General Psychiatry. 34. 1220-1235.

Andreasen. N., Rice. J., Endicott, J., et al.

(1986). The family history approach to di

agnosis: How useful is it? Archives of General Psychiatry. 43. 421-426.

Asberg. M.. Schalling, D., Traksman-Bendz,

L., et al. (1987). Psychobiology of suicide,

impulsivity, and related phenomena. In

H. Y. Meltzer (Ed.), Psychopharmacolo-

gy. the Third Generation ofProgress (pp.

655-688). New York: Raven Press.

Aston-Jones, G.. & Bloom, R. E. (1981).

Norepinephrine-containing locus coeru-

leus neurons in behaving rats exhibit

pronounced responses to non-noxious

environmental stimuli. Journal of

Neuroscience. 1, 887-900.

Baron, M., Gruen, R., Asnis, L., et al.

(1985). Familial transmission of schizo

typal and borderline personality disor

ders. American Journal of Psychiatry. 142. 927-934.

Barratt, E. S. (1959). Anxiety and im

pulsiveness related to psychomotor effi

ciency. Perceptual and Motor Skills. 9,

191-198.

Brown, G. L., Ebert, M. H., Goyer. P. F., et

al. (1982). Aggression, suicide, and sero

tonin relationships to CSF amine

metabolites American Journal ofPsychi

atry, 139, 741-745.

Brown. G. L., Goodwin, F. K., Ballenger. J.

C, Goyer, P. F., & Major. L. F. (1979).

Aggression in humans correlates with

cerebrospinal fluid amine metabolites.

Psychiatry Research, 1. 131-139.

Brown, G. L.. & Linnoila. M. I. (1990). CSF

serotonin metabolite (5-HIAA) studies in

depression, impulsivity, and violence.

Journal of Clinical Psychiatry. 51, 31-

41.

Buchsbaum, M. S., Coursey, R. D., & Mur

phy, D. L. (1976). The biochemical and

high-risk paradigm: Behavioral and

familial correlates of low platelet mono

amine oxidase activity. Science, 193,

339-341.

Buss. A. H.. & Durkee, A. (1957). An in

ventory for assessing different kinds of

hostility. Journal of Consulting Psychol

ogy, 21, 343-349.

Coccaro, E. F., Lawrence, T., Trestman, R.

L., Gabriel, S., Klar, H. M., & Siever, L. J.

(in press). Growth hormone responses to

intravenous clonidine challenge correlate

with behavioral irritability in psychiatric

patients and in healthy volunteers. Psy

chiatry Research.

Coccaro, E. F., Siever, L. J., Klar, H.. Maur-

er. G., Cochrane, K., Cooper. T. B., Mohs.

R. C, & Davis, K. L. (1989). Serotonergic studies in patients with affective and per

sonality disorders correlates with suicidal

and impulsive aggressive behavior. Ar

chives of General Psychiatry, 46, 587- 599.

Chapman, L. J., & Chapman, J. P. (1987). The search for symptoms predictive of

BIOLOGIC AND FAMILY STUDIES IN ASSESSMENT 311

schizophrenia. Schizophrenia Bulletin,

13, 497-503.

Chapman, L. J., & Chapman, J. P.. &

Raulin, M. L. (1976). Scales for physical and social anhedonia. Journal of Abnor

mal Psychology, 85, 374-382.

Cloninger, C. R. (1987). A systematic method for clinical description and

classification of personality variants: A

proposal. Archives of General Psychi

atry. 44. 573-588.

Costa, P. T., & McRae, R. R. (1990). Per

sonality disorders and the five factor

model of personality. Journal of Per

sonality Disorders. 4. 362-371.

Davidson, M., & Davis. K. L. (1988). A com

parison of plasma homovanillic acid con

centrations in schizophrenic patients and normal controls. Archives ofGeneral

Psychiatry. 45. 561-563.

Davis, K. L., Davidson, M., Mohs, R. C, et

al. (1985). Plasma homovanillic acid con

centrations and the severity of schizo

phrenic illness. Science 227. 1601-1602.

DeVauges, V., & Sara, S. J. (1990). Activa

tion of the noradrenergic system facili

tates an attentional shift in the rat. Be

havior-Brain-Research. 39. 19-28.

Eysenck. H. J.. & Eysenck, S. B. (1975).

Manual of the Eysenck Personality

Questionnaire. London: Hodder and

Stoughton. Foote, S. L., Bloom, F. E., &Aston-Jones, G.

(1983). Nucleus locus coeruleus: New evi

dence of an anatomical and physiological

specificity. Physiological Review, 63,

844.

Hodge, G. K., & Butcher, L. L. (1975).

Catecholamine correlates of isolation-

induced aggression in mice. European

Journal of Pharmacology, 31, 8193.

Ingraham, L. J. (1989). Geneticfactors in

schizophrenia and schizophrenia-like illness from a national sample of

adopted individuals and theirfamilies.

Abstract presented at the 28th annual

meeting of the American College of Neuro-

Psychopharmacology. Keefe, R., Siever, L. J., Mohs, R. C, Peter

son. A., Mahon. T., Bergman, L., & Davis.

K. L. (1989). Eye-tracking, schizophrenic

symptoms, and schizotypal personality

disorder. European Archives of Psychi

atry & Neurological Science. 239. 39--42.

Kendler, K. S., Masterson. C. C, Ungaro, R.,

et al. (1987). A family history of schi

zophrenia-related personality disorders.

American Journal of Psychiatry, 141,

424-A21.

Kendler, K. S., Ochs, A. L., Gorman. A. M..

et al. (1991). The structure of schizotypy: A pilot multitrait twin study. Psychiatry Research 36. 19-36.

Lawrence, T. L., Coccaro. E. F., Bernstein,

D. P., Lawrence, E. K., Trestman, R. L., &

Siever, L. J. (in press). Borderline per

sonality disorder as an impulsive per

sonality disorder: A psychobiologic

perspective. Journal of Personality Dis

orders.

Links, P. S., Steiner, M., & Huxley. G.

(1988). The occurrence of borderline per

sonality disorder in the families ofborder

line patients. Journal of Personality Dis

orders. 2. 14-20.

Lopez-fbor. J. J. Jr.. Lana. F. , & Saiz Ruiz.

J. (1990). Impulsive suicidal behavior

and serotonin. Actas Luso Espandes de

Neurologia Psiquiatria y Ciencias

Afines. 18. 316-325.

Loranger, A. W., Oldham, J. M. & Tulis, E.

H. (1982). Familial transmission of DSM-

III borderline personality disorder. Ar

chives of General Psychiatry. 39. 795

799.

Olivier. B., Mos, J., van der Heyden, J.,

Schipper, J., Tulp, M., Berkelmans, B., &

Bevan. P. (1987). Serotonergic mod

ulation of agonistic behavior. In B.

Olivier. J. Mos, & P. F. Brain (Eds.), Etho-

pharmacoiogy of Agonistic Behavior in

Animals and Humans, (pp. 162186).

Dordrecht: Martinus Nijhoff. Pickar, D.. Labarca, R., Doran, A. R., et al.

(1986). Longitudinal measurement of

plasma homovanillic acid levels in schizo

phrenic patients correlations with psy chosis and response with neuroleptics. Archives ofGeneral Psychiatry. 43. 669

676.

Pope, H. G., Jonas, J. M.. Hudson. J. I., et

al. (1983). The validity of DSM III border

line personality disorder. Archives of General Psychiatry. 40. 23-30.

Siever, L. J. (1987). The role of noradren

ergic mechanisms in the etiology of the

affective disorders. In H. Y. Meltzer (Ed.),

Psychopharmacology : The Third

Generation of Progress (pp. 493504).

New York: Raven Press.

Siever, L. J. (1991). The biology of the

boundaries of schizophrenia. In C. A.

Tamminga & S. C. Schulz (Eds.), Ad

vances in Neuropsychiatry and Psy

chopharmacology. Vol. 1: Schizophrenia Research (pp. 181-191). New York:

Raven Press.

Siever, L. J. (in press). The schizophrenic

312 SIEVER, TRESTMAN, AND SILVERMAN

spectrum personality disorders. In A.

Tasman (Ed.). Annual Review of Psychi

atry Press.

Siever, L. J., Amin, F., Coccaro, E. F., Bern

stein, D., Kavoussi. R. J., Kalus, O., Hor-

vath, T. B., Warne, P., Davidson, M.,

Davis, K. (1991b). Plasma homovanillic

acid in schizotypal personality disorder

patients and controls. American Journal

of Psychiatry, 148. 1246-1248.

Siever, L. J., Amin. F. , Coccaro. E. F., Trest

man, R. L., Silverman, J., Horvath, T. B.,

Mahon, T. R., Knott, P., Davison. M.. &

Davis, K. L. (in press-a). Cerebrospinal

fluid homovanillic acid in schizotypal per

sonality disorder. American Journal of

Psychiatry.

Siever, L. J., Bernstein, D. P., & Silverman,

J. M. (1991a). Schizotypal personality disorder: A review of its current status.

Journal ofPersonality Disorders. 5. 178

193.

Siever. L. J., Keefe, R., Bernstein, D. P..

Coccaro, E. F., Klar. H. M., Zemishlany,

Z., Peterson, A., Davidson, M., Mahon,

T.. Horvath, T., & Mohs. R. (1990a). Eye

tracking impairment in clinically identi

fied schizotypal personality disorder

patients. American Journal of Psychi

atry. 147. 740-745.

Siever, L. J.. Klar, H.. & Coccaro, E. F.

(1985). Psychobiologic substrates of per

sonality. In L. J. Siever & H. Klar (Eds.),

Biologic Response Styles: Clinical Im

plications (pp. 38-66). Washington, DC:

American Psychiatric Press.

Siever, L. J., Murphy, D. L., Slater, S., de la

Vega, E., & Lipper, S. (1984). Plasma pro lactin changes following fenfluramine in

depressed patients compared to controls:

An evaluation of central serotonergic

responsivity in depression. Life Science,

34. 1029-1039.

Siever, L. J.. Silverman, J. M., Horvath, T.

B., et al. (1990b). Increased morbid risk

for schizophrenia-related disorders in

relatives of schizotypal personality dis

ordered patients. Archives of General

Psychiatry. 47. 634-640.

Siever, L. J., Trestman, R. L., Coccaro, E.

F., Bernstein, D. P., Gabriel, S. M.. Owen,

K., Moran. M., Lawrence, T., Rosenthal,

J., & Horvath, T. B. (in press-b). The

growth hormone response to clonidine in

acute and remitted depressed male

patients. Neuropsychopharmacology. Silverman. J. M., Pinkham, I., Horvath, T.

B., Coccaro, E. F., Klar, H., Schear, S.,

Apter, S., Davidson, M.. Mohs. R. C

Siever, L. J. (in press). .Affective and im

pulsive personality disorder traits in the

relatives of borderline personality dis

order patients. American Journal of Psy

chiatry.

Silverman. J. M., Siever, L. S., Pinkham, L.,

Mohs, R. C, & Davis, K. L. (1991).

Identifying the schizophrenia related

phenotype. Abstract presented at the

144th annual meeting of the American

Psychiatric Association.

Soubrie, P. (1989). Reconciling the role of

central serotonin neurons in human and

animal behavior. Behavioral and Brain

Sciences. 9. 319-364.

Squires-Wheeler, E.. Skodol, A. E., Fried

man, D.. Erlenmeyer-Kimling, L. (1988).

The specificity of DSM-III schizotypal per

sonality traits. Psychological Medicine.

18. 757-765.

Stanley. M., & Stanley, B. (1990). Postmor

tem evidence for serotonin's role in

suicide. Journal of Clinical Psychiatry.

51S. 22-28.

Trestman, R. L., Lawrence, T. L., Coccaro,

E. F., Mltropoulou, V., Gabriel, S., Wes

ton, S., & Siever, L. J. (1991). Impulsiv

ity, aggression and prolactin response to

fenfluramine in personality disorders.

Neuroendocrinology Letters, 13, 192.

Trestman, R. L., & Siever, L. J. (1991). The

growth hormone response to clonidine in

depression. In M. Ansseaw, R. von

Frenckell, & G. Franck (Eds.), Biological Markers of Depression: State of the Art

(pp. 4353). New York: Elsevier Science

Publishers.

Valzelli, L. (1981). Psychobiology ofAggres sion and Violence. New York: Raven

Press.

Widiger, T. A., Frances, A., Warner, L., et al.

(1986). Diagnostic criteria for borderline

and schizotypal personality disorders.

Journal of Abnormal Psychology, 95,

43-51.

Zanarini, M. C, Gunderson, J. G., Franken-

burg, F. R. (1990). Cognitive features of

borderline personality disorder. Amer

ican Journal of Psychiatry, 147, 5763.

Zanarini, M. C. Gunderson, J. G., Marino,

M. F.. et al. (1988). DSM-III disorders in

the families of borderline outpatients. Journal ofPersonality Disorders, 2:292- 302.

Zuckerman, M. (1979). Sensation Seeking: Beyond the Optimal Level of Arousal.

Hillsdale, NJ: Lawrence Erlbaum.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.