Elsevier

Biological Psychiatry

Volume 51, Issue 8, 15 April 2002, Pages 621-631
Biological Psychiatry

Original article
Neurodevelopmental patterns of visual P3b in association with familial risk for alcohol dependence and childhood diagnosis

https://doi.org/10.1016/S0006-3223(01)01301-4Get rights and content

Abstract

Background: The P3b component of the event-related potential (ERP) has frequently been reported to be reduced in children and adolescents at high risk for developing alcoholism relative to control children and adolescents without familial loading for alcohol dependence. P300 amplitude changes during development for all children. Previously it has been shown that high-risk offspring display a pattern in which the amplitude is lower at age 8 with a smaller rate of change during adolescence.

Methods: Admixture analysis was applied to data obtained for those children and adolescents having five or more annual assessments of ERPs to determine if multiple P3b growth patterns exist. The P3b amplitude patterns obtained were related to risk status, concurrent presence of childhood psychopathology (internalizing or externalizing), and age of onset to develop a diagnosis.

Results: A pattern characterized by lower P3b amplitude at study entry and a slower rate of change during child and adolescent development (pattern 3) was most often associated with high-risk status in boys and high-risk status in combination with the presence of a childhood diagnosis in girls. Pattern 3 was significantly related to the overall presence of childhood psychopathology (internalizing or externalizing) and to the presence of an Axis I diagnosis at young adult follow-up.

Conclusions: The developmental pattern previously described for offspring at high risk for developing alcoholism because of their familial/genetic background was confirmed. Admixture analysis has refined this observation and suggests that among all children and adolescents tested, three developmental patterns can be identified, one of which is most often seen in association with male high-risk children and adolescents.

Introduction

There has been a long-standing interest in finding trait markers for psychiatric disorders (Zubin and Steinhauer 1981). Several promising neurobiological markers have been investigated including electroencephalographic and event-related potential characteristics, electrodermal response, eye tracking, heart rate, and pupillography (Venables 1991). The event-related potential (ERP) and the P300 component in particular, has been studied as a potential marker for development of psychiatric problems, including such diverse disorders as schizophrenia Pfefferbaum et al 1989, Steinhauer et al 1991 and alcoholism Pfefferbaum et al 1991, Porjesz et al 1987a, Porjesz et al 1987b, Steinhauer et al 1987. The P300 is a scalp positive wave that occurs approximately 300 msec after an informative stimulus occurs (Sutton et al 1965) and is an index of an individual’s capacity to process stimulus information. Two variants of P300 have been described, P3a and P3b. The traditional P3b (Sutton et al 1965) that has been studied most extensively has a parietocentral distribution in comparison with the more anterior scalp distribution of the P3a Courchesne et al 1975, Squires et al 1975. The two types of P300 also differ in the aspects of cognitive processing that they reflect. In contrast with the loading on novelty, or “surprise” in the words of Verleger et al (1994), associated with the more anterior P3a, P3b is associated with more of what Verleger et al (1994) termed “suspense” because it reflects the process whereby the participant awaits the next relevant stimulus. The P3a is elicited in response to deviant stimuli (rare target or rare nontarget). P3b is elicited usually by relevant target stimuli that are typically less frequent than the nontarget condition. A P3 in response to novelty has also been described (Comerchero and Polich 2000). A novelty P3 is elicited by unpredictable unique stimuli that have not been presented previously; however, recent evidence suggests that the P3a and the Novelty P3 are indistinguishable (Simons et al 2001). It is important to distinguish the varying types of P300 because the search for trait markers of psychiatric risk has oftenused the two variants interchangeably, leading to some confusion in the literature. Variations in trait characteristics among various psychiatric disorders can be expected between those in which reduced involuntary responding to novelty is the core trait and those for which the capacity to use attentional resources is the central feature.

In contrast to the increased latency seen in association with environmental factors (e.g., solvent exposure, closed head injury) or organic disease states (e.g., Alzheimer’s disease; Morrow et al 1992, Polich 1991), decrements in P3b amplitude are more often associated with the presence of psychiatric disorders. Reduced amplitude has been reported for schizophrenic patients Pfefferbaum et al 1989, Steinhauer and Zubin 1982, Steinhauer et al 1991 and for patients diagnosed with depression Bruder et al 1995, Yanai et al 1997. Additionally, individuals without psychiatric illness but who carry an especially high loading for a particular psychiatric disorder, especially alcoholism, have been reported to show reduced amplitude when compared with control subjects Begleiter et al 1984, Friedman et al 1995, Hill et al 1995a, Hill and Steinhauer 1993, Steinhauer and Hill 1993. Thus, it is possible that the amplitude of the P3b component may be an inherent characteristic of individuals before they develop psychiatric states that in turn may be related to their vulnerability for incurring these disorders Hill 1994, Hill et al 1987. P3b amplitude appears to be superior to P3 latency as an endophenotypic marker for psychiatric risk because P3b amplitude shows greater heritability than P3b latency Aston and Hill 1990, van Beijsterveldt 1996. Whether reduction in P3b amplitude shows specificity for particular disorders or is a marker of generalized susceptibility to psychiatric disorders is unknown.

Some investigators have argued that for P300 amplitude to be a valid biological marker of susceptibility to a psychiatric disorder it should be present in both child and adult high-risk relatives of psychiatrically diagnosed probands (Porjesz et al 1998); however, alcoholism is one condition in which study of adults may present a problem for understanding whether the P300 component is a trait marker. P3b reductions have been demonstrated to be larger in alcoholics with cortical atrophy than in alcoholics without neuropathological changes (Begleiter et al 1980). Although rigorous investigations have not found significant associations between P3b amplitude in adult alcoholics and lifetime alcohol consumption, conclusions must be tempered by the 20- to 30-year retrospective recall required for such investigations (Pfefferbaum et al 1991). Similarly, the presence of other psychopathology, which is common among alcoholics, may be responsible for the reduction in amplitude of P3b (Hill et al 1999b) and P3a (Costa et al 2000) in adult alcoholics. The presence of these contaminating factors may explain the variable results seen in adult alcoholics, which include reduction of P3b in adult male alcoholics by some Pfefferbaum et al 1991, Porjesz et al 1987a, Porjesz et al 1987b, but not all investigations Hermanutz et al 1981, Hill et al 1999b, Hill et al 1995b, Lille et al 1987, Pfefferbaum et al 1979.

Coupled with the fact that alcoholics tested in ERP paradigms usually have long drinking histories (usually more than 12 years) is the fact that they often are tested after very short periods of abstinence (usually only 2 weeks; Porjesz et al 1987a). Importantly, adult alcoholics tested after a period of 6 months abstinence do not differ in visual or auditory P3b amplitude from controls (Biggins et al 1995) though the alcoholics display longer latency characteristic of exposure to neurotoxic substances (Biggins et al 1995). The critical nosological question is whether P3b reduction, if seen, was present before the alcoholic individual began to drink. Current evidence based on children and adolescents suggests that this is the case.

Developmental studies of normal children are infrequent. Fewer than 200 children and adolescents have been assessed across available studies using either visual or auditory ERPs Courchesne 1977, Kurtzberg et al 1984, Polich et al 1990; however, these cross-sectional studies have demonstrated substantial differences between the ERP characteristics of adults and children Kurtzberg et al 1984, Polich et al 1990. Using samples of minor children, differences in P3b amplitude are seen between children at high- and low-risk for developing alcohol dependence Begleiter et al 1984, Berman et al 1993, Hill et al 1995a, Hill and Steinhauer 1993, Hill et al 1990, Steinhauer and Hill 1993, Whipple et al 1988, although no differences are found between high and low-risk children evaluated with a P3a paradigm (Holguin et al 1998). A meta-analysis performed by Polich et al (1994) suggests that the effect of age is critical in uncovering P3b amplitude differences by alcoholism risk status. The largest effect sizes observed were between high- and low-risk younger children, especially with more difficult ERP paradigms, with few differences being seen in young adulthood. Children at high risk for developing other psychiatric disorders have infrequently been studied; however, one study followed children at high risk for developing schizophrenia, finding lower P3b in childhood was associated with overall higher global impairment at age 18 (Friedman et al 1995).

The presentation modality has been shown to affect the P3b amplitude trajectories of childhood and adolescence (Hill et al 1999c). Before that study, longitudinal data were not available to determine if the same children followed over time would have developmental trajectories that varied by modality (visual or auditory) or by risk group status (high or low risk for alcohol dependence). Growth curve modeling of 635 separate auditory and visual P3b assessments of children and adolescents between the ages of 8 and 18 revealed that auditory P3b showed an increase in amplitude with age followed by a leveling off by late adolescence. In contrast, visual P3b amplitude appears to be greatest at younger ages, declining until about young adulthood. Importantly, the developmental trajectories of P300 amplitude in high-risk offspring from high density for alcoholism families differed from that seen in low-risk offspring (Hill et al 1999c). Male high-risk offspring show a slower rate of change in P3b amplitude over the period of 8 to 18 years than do control children (Hill et al 1999c). Although similar trajectories were seen in high-risk girls, a comorbid childhood diagnosis was required for this pattern to be manifest. The high-risk group, as a whole, entered the study with lower visual P3b amplitudes than control subjects and showed a slower rate of change. Based on data acquired through late adolescence, the theoretical point of convergence when high- and low-risk groups would be indistinguishable was calculated to be at age 18 for the auditory condition and 22 for the visual (Hill et al 1999c). Empirical data (Hill et al, unpublished manuscript) confirm that, as with other developmental milestones that can be delayed during childhood (e.g., onset to walk or talk) but are clearly present by adulthood, the average P3b amplitude for the high- and low-risk groups are virtually identical by young adulthood as was predicted theoretically on the basis of longitudinal growth analyses projected to young adulthood (Hill et al 1999c).

Although the previous report had identified overall differences in the rates of change between high- and low-risk children and adolescents, visual inspection suggested the presence of multiple trajectory patterns. This report is based on an admixture analysis of subject data from the growth curve data set but restricted to participants for whom at least five repeated annual assessment waves of data were available. The goal of this analysis was to determine if risk status confers a different developmental pattern across childhood and adolescence. A secondary goal was to determine if pattern type would be related to the individual diagnoses of the offspring studied. To establish that P3b is a trait marker for psychiatric illness, it is important to rule out contributions from state variables such as concurrent psychiatric illness in the participants studied. It was expected that the results of these analyses might point the way to underlying neurobiological subtypes associated with risk for development of psychiatric illness that might be recognizable in childhood. Identification of developmental variants might potentially provide a means for identifying children at an especially high risk for developing adult psychopathology for targeted interventions.

Section snippets

Subjects

We followed annually 126 children between the ages of 8 and 18 who were offspring of parents enrolled in a large family study. The children were either at high or low risk for developing alcohol dependence based on differing familial loading for alcohol dependence. Some of these children and adolescents have matured into a second initiative designed to address prevalence of psychiatric disorders in young adults (ages 19–30) who vary by family history of alcohol dependence. Data for a subset of

Results

To determine the best-fit model, BIC values, classification quality, and utility of the latent class were examined. The BIC statistics were 3184.64 for the one-class model, 3097.95 for the two-class, 3109.25 for the three-class, 3104.50 for the four-class, 3118.45 for the five-class, and 3130.55 for the six-class solutions. BIC values for a single class or more than four were quite large and not further considered. The average estimated posterior probabilities, conditional on the class

Discussion

Our study demonstrated that varying patterns of developmental change in P3b amplitude occur among children. Using admixture latent growth variable modeling, three distinct patterns could be identified: high intercept P3b with a flat trajectory (Class 1); high intercept P3b with a downward trajectory (Class 2); and low intercept P3b with a flat trajectory (Class 3). Class 1 children displayed amplitudes that were higher than the mean for the entire group (over 35 μV) and remained at this level

Acknowledgements

Supported by the National Institute on Alcohol Abuse and Alcoholism Grant Nos. AA05909, AA08082, and AA11304. We wish to thank all of the individuals involved in acquiring the ERP data, especially Jeannette Locke-Wellman, and Dr. Stuart Steinhauer for his consultation. We also thank the clinicians, especially Lisa Lowers, MA, for maintaining excellent retention in the follow-up. Also, we are grateful to the families who have participated multiple times.

References (69)

  • S.Y Hill et al.

    Developmental delay in P300 production in children at high risk for developing alcohol-related disorders

    Biol Psychiatry

    (1999)
  • S.Y Hill et al.

    Factors predicting the onset of adolescent drinking in families at high risk for developing alcoholism

    Biol Psychiatry

    (2000)
  • S.Y Hill et al.

    Eight year longitudinal follow-up of P300 and clinical outcome in children from high-risk for alcoholism families

    Biol Psychiatry

    (1995)
  • A Pfefferbaum et al.

    Event-related potential changes in chronic alcoholics

    Electroencephalogr Clin Neurophysiol

    (1979)
  • J Polich et al.

    Normal variation of P300 in childrenAge, memory span, and head size

    Int J Psychophysiol

    (1990)
  • B Porjesz et al.

    Event-related brain potentials to high incentive stimuli in abstinent alcoholics

    Alcohol

    (1987)
  • B Porjesz et al.

    The N2 component of the event-related brain potential in abstinent alcoholics

    Electroencephalogr Clin Neurophysiol

    (1987)
  • R.F Simons et al.

    On the relationship of P3a and the Novelty-P3

    Biol Psychiatry

    (2001)
  • N.K Squires et al.

    Two varieties of long-latency positive waves evoked by unpredictable auditory stimuli in man

    Electroencephalogr Clin Neurophysiol

    (1975)
  • S.R Steinhauer et al.

    Event-related potentials in alcoholics and their first-degree relatives

    Alcohol

    (1987)
  • S.R Steinhauer et al.

    Vulnerability to schizophreniaInformation processing in the pupil and event-related potential

  • I Yanai et al.

    Changes in auditory P300 in patients with major depression and silent cerebral infarction

    J Affect Disord

    (1997)
  • C.E Aston et al.

    A segregation analysis of the P300 component of the event-related potential

    Am J Hum Genet

    (1990)
  • R.M Baron et al.

    The moderator-mediator variable distinction in social psychological researchConceptual, strategic, and statistical considerations

    J Pers Soc Psychol

    (1986)
  • L.O Bauer et al.

    Frontal P300 decrements in antisocial personality disorder

    Alcohol Clin Exp Res

    (1994)
  • H Begleiter et al.

    Event-related brain potentials in boys at risk for alcoholism

    Science

    (1984)
  • H Begleiter et al.

    Neuroradiological and neurophysiological evidences of brain deficits in chronic alcoholics

    Acta Psychiatr Scand Suppl

    (1980)
  • C.A Biggins et al.

    Delayed P3A in abstinent elderly male chronic alcoholics

    Alcohol Clin Exp Res

    (1995)
  • G.E Bruder et al.

    Brain event-related potentials to complex tones in depressed patientsRelations to perceptual asymmetry and clinical features

    Psychophysiology

    (1995)
  • S.R Carlson et al.

    Substance dependence and externalizing psychopathology in adolescent boys with small, average, or large P300 event-related potential amplitude

    Psychophysiology

    (1999)
  • W.J Chambers et al.

    The assessment of affective disorders in children and adolescents by semistructured interview. Test-retest reliability of the schedule for affective disorders and schizophrenia for school-age children, present episode version

    Arch Gen Psychiatry

    (1985)
  • M Comerchero et al.

    P3a, perceptual distinctiveness, and stimulus modality

    Cogn Brain Res

    (2000)
  • E Courchesne

    Event-related brain potentialsComparison between children and adults

    Science

    (1977)
  • J.P Feighner et al.

    Diagnostic criteria for use in psychiatry research

    Arch Gen Psychiatry

    (1972)
  • Cited by (57)

    • Genetic psychophysiology: Advances, problems, and future directions

      2014, International Journal of Psychophysiology
    • Cholinergic receptor gene (CHRM2) variation and familial loading for alcohol dependence predict childhood developmental trajectories of P300

      2013, Psychiatry Research
      Citation Excerpt :

      Longitudinal follow up of offspring at ultra high risk for alcohol dependence through their membership in multiplex families has demonstrated an increase in both externalizing and internalizing disorders in comparison to low risk controls (Hill et al., 2008). These studies also found that childhood psychopathology whether externalizing or internalizing is associated with lower P300 amplitude (Hill et al., 1999) with any childhood psychopathology associated with greater likelihood of being in the low amplitude trajectory class (Hill and Shen, 2002). Interestingly, children with internalizing disorder had lower P300 amplitude, and those with both internalizing and externalizing disorders had the lowest P300 amplitude (Hill et al., 1999).

    View all citing articles on Scopus
    View full text