i P300 Amplitude Decrements in Children from Families of Alcoholic Female Probands Shirley Y. Hill, Diane Muka, Stuart Steinhauer, and Jeannette Locke A total o[ 70 age- and gender-matched children and their relatives who werefromJamilies that were either at high or low risk for developing ah'oholism were studied and the children evaluated using event-related potential paradigms from both the auditory and visual modali- ties. The high-risk children were ascertained through .~amilies at exceptionally high risk for ]bmale alcoholism (haft o/all female first- and second-degree relatives were alcoholic), while low-risk control/iunilies had been selectedfiJr absence o[ alcoholism in first- and second-de- gree relatives. Results indicate that reduced amplitude q{ P300 and greater negativity of N250 characteri:e childret~ fl'om high-risk families when evaluated with an auditory task. The visual task discriminated high- and low-risk groups /or male children only, consistent with earlier findings for high-ri.sk families ascertained through a male alcoholic. Further, daughters of alcoholic mothers (biological .[ather nonaleoholic J display significantly lower P300 than matched controls, indicating that transmi.ssion ~{alcoholism risk may be possible from mother to daughter without the necessity qfpatermd alcoholism. Key Words: P300. N250, ERP, women, alcoholism, high-risk children BIOL PSYCHIATRY 1995;38:622-632 Introduction There is increasing recognition of the fact that there may be two types of alcoholism in women. One type appears to have an early onset, is characterized by a higher density of relatives who are alcoholic, and ma~. have a poorer progno- sis (Hill and Smith 1991: Lex et al 1991: Glenn and Nixon 1991 ). If this form is genetically mediated, one might expect biological variation among these women and possibly their offspring. Previously, differences in the P300 component of the event-related potential (ERP) between alcoholic women from high-density families and controls have been reported (Hill and Steinhauer 1993a I. If the biological marker ( P300 decrement) is predictive of later development of alcohol- ism, studies of children who have not yet become dependent FI'tlII1 the Dt2p~lYll/ICrll i/1 I:','~Chhklt]~ I Ill'~kl,]lN ~1 Ir;'ltl~htlt<_th ~]edlk I] (~CTIILq PiU',bU rgh. PA. Address reprint requests, to [)l Shirhc> 5 tltlL Dcpaltmcni ut P%'chiatr). WPIC. I nl~ el ~,11~ ol Pittsburgh Sch~×l/ul Medicine ~sll ()Haia Stlcel. PittnbknIgh. P~, 1~,~1 Recci\ed Augu-q 22. 1994: ic\i,cd No~ clnl~¢l 14. 1'4t)4 would be most informative. However, minor children who are relatives of female alcoholics have not previously been characterized using event-related potentials. The P300 component of the event-related potential has received considerable attention as a possible neurophysio- logical risk marker for the development of alcoholism. P300 is a scalp-positive wave occurring approximately 300 msec after an informative event. The event-related potential is of interest for two reasons. First, long-latency components of ERPs. including P300, are associated with particular sen- sory and cognitive aspects of information processing (Sub ton et al 1965: Donchin 1979). Second, the ERP wavefonn appears to be under genetic control (Aston and Hill 1990; Polich and Burns 1987; Rogers and Deary 1991; Surwillo 1980). The promise of utilizing P300 as a marker for alco- holism risk has been clouded by controversy surrounding the fact that some research groups have presented strong evidence for P300 being a risk marker (Begleiter et al 1984; Hill and Steinhauer 1993b; Steinhauer and Hill 1993; @ ] 9t}5 Socicl} ~i Biol(~gi~M I,'%~htah? 0006-32231951509.50 SSDI 0006-3223(94)00384-FP300 Amplitude Decrements in Children BIOL PSYCHIATRY 623 1995;38:622-632 Whipple et al 1988). Other groups have found moderate support (Elmasian et al 1982: O'Connor et al 1987); still others have found no difference between high- and low-risk groups (Baribeau et al 1987; Neville and Schmidt 1985; Polich and Bloom 1987: 1988: Polich et a11988a; 1988b). To date, six studies have employed prepubescent boys (Begleiter et al 1984; Begleiter et al 1987; Hill et al 1990; Hill and Steinhauer 1993b: Whipple et al 1988; Steinhauer and Hill 1993), while the majority of other studies have utilized young adult college students. Only two studies have contrasted both pre- and postpubescent boys and girls (Hill and Steinhauer 1993b; Steinhauer and Hill 1993). In these studies, P300 decrements were found using both auditory and visual modalities among high-risk as compared to low- risk controls. Previous findings from this laboratory (Hill et al 1990: Steinhauer and Hill 1993) have also included the observation that there was significantly increased late nega- tivity in the auditory ERP, measured by the peak of the N250 component, over anterior regions (Fz and Cz) in the high- risk children as compared to children of low-risk families. The late anterior negativity typically decreases with devel- opment (Friedman et al 1984). These findings are of consid- erable interest, as they suggest that children ascertained through male alcoholic proband families might be charac- terized by a delay in maturational processes. Understanding the critical differences in methodologies leading to these divergent results is of considerable interest. The age of the high-risk subjects employed (prepubescent. postpubescent minors, and young adults), as well as the proportion of males and females utilized, has varied across studies. Differences in task difficulty and the modality of stimulus presentation are additional factors that have not been systematically varied using the same subject pool. Finally, criteria for "high-risk" status has varied consi- derably across studies. We have reserved the use of the term "high-risk" to denote cases where the familial constellation is of sufficient density to produce lifetime risk of 50c/~ or greater for young males. For an individual to have a pre- dicted recurrence risk of this magnitude, multiple relatives must be alcoholic (Aston and Hill 1990). Most studies using the Family History Positive (FHP) and Family History Neg- ative (FHN) dichotomy to determine presence or absence of