Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Neurotoxicol Teratol. Author manuscript; available in PMC 2010 November 1.
Published in final edited form as:
PMCID: PMC2765041

Children With and Without Gestational Cocaine Exposure: A Neurocognitive Systems Analysis


An extensive preclinical literature documents the deleterious effect of cocaine on prenatal brain development and subsequent function in animals. One mechanism by which cocaine affects brain development is by a reduction in uterine blood flow and subsequent fetal hypoxia [65,87]. Cocaine also disrupts the neural ontogeny of the monaminergic neurotransmitters, which include dopamine, norepinephrine, and serotonin [45,62]. This disruption may affect brain development globally as well as impact the structure and function of specific systems. In this regard, anatomical, physiological, and behavioral abnormalities have been found in animals with gestational cocaine exposure (GCE) [26,3840,79]. The systems most reliably affected by GCE in animals include the executive/attentional systems of prefrontal cortex and memory [3840].

In contrast, the empirical literature on humans is inconsistent, with some studies finding cognitive impairments and others finding no effect of GCE [8,11,35,49,50,56,57,66,73,77,83]. One complicating factor is that children with GCE are more likely to have a number of other developmental risk factors, both medical and psychosocial, than children without such exposure. For example, cocaine using mothers are more likely to use other drugs, to have poorer nutrition, and to have less stable home lives [55,81]. Failure to control for these potential confounds would inflate estimates of the impact of GCE.

An earlier review of the effects of GCE on human brain and behavioral development in young children was published by Frank, et al [34]. After excluding studies that lacked control groups, failed to blind examiners to group membership, did not prospectively enroll participants, or suffered from other serious methodological limitations, the authors were left with 36 studies. On the basis of these studies, all of which were conducted with children aged six years or younger, the authors concluded that: “there is no convincing evidence that prenatal cocaine exposure is associated with developmental toxic effects that are different in severity, scope or kind from the sequelae of multiple other risk factors.” This review demonstrated that many findings once thought to be specific effects of gestational cocaine exposure are correlated with other factors, including prenatal exposure to tobacco, marijuana and/or alcohol and the quality of the child’s environment [25,34,80].

More recently published studies, controlling for a number of medical and social risk factors, report finding differences between exposed and non-exposed children on measures of language [5,57,67], attention [1,68,72,75] and perceptual reasoning [77]. For many of these studies, effect sizes are small [5,57,67], with findings often limited to single rather than multiple components of language and attention [68,75]. However, these effects may prove meaningful because underlying abnormalities in cognitive processing, such as language processing, attention, and perceptual reasoning, are associated with learning disorders and associated problems in school [2].

Some studies examining the relation between cocaine exposure status and school performance have found a higher incidence of learning disability in exposed subjects [56,66]. Morrow et al [66] report that cocaine exposed children were 2.8 times more likely to develop a learning disability by age 7. On the other hand, in the same cohort as for this report, Hurt et al [49] examined the grade retention patterns and reported no effect of cocaine exposure on grade retention patterns through grade 4 after controlling for important covariates. In this study, while GCE was associated with a higher rate of commission errors on a distractibility task, and higher distractibility was associated with higher rates of grade retention, no reliable relation between cocaine exposure and grade retention was established. Similarly, Singer et al, [77] in a recent, well controlled evaluation examining both IQ and school achievement, found cocaine effects on only the Perceptual Reasoning composite score of the Wechsler Intelligence Scale for Children – Fourth Edition and reported no effects on domains of academic achievement. Given these varied results, continued investigation of effects of gestational cocaine exposure on specific cognitive processes through school age are necessary for improved understanding of the long term outcomes of these at risk children.

Across various studies, cocaine effects have been identified in the cognitive domains of language [5,57,67], attention [1,68,72,75] and perceptual reasoning. The present study was designed to assess effects of GCE on these cognitive skills specifically, as well as on other important cognitive domains, while controlling for important covariates that are likely to impact development. Each domain that is assessed in this report is either prefrontally mediated and responsible for the control of attention, working memory, and decision making, or non-prefrontally mediated and responsible for processing language, declarative memory, spatial cognition, and visual cognition. The study also was motivated by the need to address GCE effects at older ages, given the possibility of so called “sleeper” or latent effects [36,63,78]. The subjects for the investigation reported here were middle school-aged children with and without GCE. They were assessed with tasks adapted from the cognitive neuroscience literature to tap seven specific neurocognitive (NC) systems: Cognitive Control, Working Memory, Reward Processing, Language, Memory, Spatial Cognition, and Visual Cognition. Multivariate analysis of these NC systems allowed for examination of general multivariate effects on cognitive function as well as specific between-subject effects on individual NC systems.


2.1. Subjects

Our subjects were 120, primarily African American, middle school-aged children, approximately half with gestational cocaine exposure and half without. All were recruited at birth from a single inner-city hospital, were born at or near term (≥ 34 weeks), and had 5-minute Apgar scores of at least 5. The children were of low socioeconomic status (SES) as defined by maternal receipt of public medical assistance. Birth characteristics for GCE and Control subjects included in this report are shown in Table 1. None of the children had Fetal Alcohol Syndrome or any syndrome known to be associated with developmental delay. As previously reported [47,48], mothers were native English speakers and had no past or present indication of major psychiatric illness as determined by medical chart review at time of enrollment. Maternal use of cocaine, as well as amphetamines, opiates, barbiturates, and benzodiazepines, was ascertained by interview, medical record review at time of birth, and by maternal and infant urine specimens. Mothers who reported use of drugs other than tobacco, alcohol, marijuana, and cocaine were excluded from enrollment in the study [47,48]. Mothers’ estimated days of use of cocaine use was also ascertained through maternal interview at time of enrollment in the study. Only mothers who reported cocaine use in at least two trimesters of pregnancy were included. While use of cigarettes, alcohol, and marijuana during pregnancy was also documented by interview and medical record review, specific information on frequency and amount of use of these substances was not.

Table 1
Subject Characteristics by GCE Status

Since enrollment, the children have been evaluated semi-annually for measurements of growth, development, language, cognitive, and social-emotional outcomes [47,50,51]. Two hundred twenty-four (105 GCE and 119 Controls) subjects were enrolled at birth. During the ensuing years, five subjects died (1 GCE and 4 Controls), with an additional early attrition of approximately 40% of Controls and 42% of GCE. Since this early attrition (largely related to inadequate number of study personnel secondary to fiscal constraints), cohort number has been stable for the past 8 years or more, with 120 participants comprising the sample for the current report of NC function. GCE children lost to follow-up did not differ on any birth or maternal characteristics from those GCE children retained, including such variables as level of prenatal care, maternal drug screen positive for cocaine, birth weight, head circumference, exposure to alcohol, cigarettes or marijuana, admission to NICU, and discharge to biological mother (p> 0.20). Further, those lost to follow-up did not differ from those retained in number of days of maternal cocaine use during pregnancy (p=0.61). The only statistically significant difference found between those lost to follow-up and those retained was in the Control group, with more males (60%) lost to follow up than females (40%) (p= 0.042).

Assent was obtained from all participating subjects and informed consent was obtained from their caregivers. The project was approved by the Institutional Review Boards of the University of Pennsylvania and the Children’s Hospital of Philadelphia.

2.2. NC Assessment

NC functioning was evaluated at subject mean age of 12.1 years, using a battery of ‘pencil and paper’ and computerized tasks designed to tax three frontal and four non-frontally mediated NC systems defined by anatomical and functional criteria. While we understand that the whole brain is working during performance of each task, our strategy was to select tasks that disproportionately tax particular systems as evidenced in the cognitive neuroscience literature cited below. We selected two representative tasks per NC system, with each task chosen to be as different as possible from the other in terms of stimulus and response types. Brief descriptions of NC systems assessed and tasks are listed below, with details available in other published reports [31,32].

Executive Systems (Frontal)

The Cognitive Control system, closely linked to the anterior cingulate cortex in imaging and lesions studies, plays a crucial role in monitoring for conflict between the individual’s automatic responses and the correct response. In addition, this system has been linked to the ability to summon additional attention needed to regulate responses. Tasks: Go-NoGo; Number Stroop [17,21].

The Working Memory system, closely linked to the dorsolateral prefrontal cortex in imaging and lesions studies, requires the ability to hold the present context, rules, or goals in mind as the individual performs a complex task. Tasks: Spatial Working Memory; Letter Two-Back [18,19].

The Reward Processing system, associated with the ventromedial prefrontal cortex in imaging and lesions studies, underlies an individual’s ability to resist the pull of reward stimuli. Tasks: Delay task from the Gordon Diagnostic System; Intra/Extra Dimensional (ID/ED) Shift task [18,43].

Non-executive Systems (Non-frontal)

The Language system, shown in imaging and lesion studies to be linked to the left perisylvian cortical area, underlies two main and distinct components of language: lexical semantics and syntax. Tasks: Peabody Picture Vocabulary Test (PPVT); Test for the Reception of Grammar (TROG) [13,27].

The Memory system, shown in imaging research to map onto the medial temporal cortical area, is required for incidental or one trial learning. Tasks: Incidental Word Learning; Incidental Face Learning [60].

The Spatial Cognition system, shown in imaging and lesion studies to activate the parietal cortical area [14,70], subserves the representation and manipulation of spatial information. Tasks: Line Orientation; Mental Rotation [28].

The Visual Cognition system, linked in imaging and lesion studies to the occipito-temporal brain region, subserves the segmentation and recognition of shapes. Tasks: Shape Detection from Warrington and James’ Visual Object & Space Perception (VOSP) Battery; Face Perception from Mooney’s Test of Visual Closure [64,84].

2.3. Cocaine Exposure

Two measures of cocaine exposure were examined. In our main analyses we included a dichotomous, “yes” or “no” variable for cocaine exposure status based on maternal interviews and urine drug screens at birth. To examine the possibility of a dose effect, we computed an estimation of days of gestational cocaine exposure. During the enrollment interview, at the time of the child’s birth, mothers were asked to indicate frequency of cocaine use by trimester on a 10 item scale that ranged from: “no use” to “2–3 times per week” to “three or more times per day”. Days of use were estimated from these responses.

2.4. Other Measures

Other Exposures

At time of enrollment, mothers were asked about use of alcohol, cigarettes, and marijuana during pregnancy and use of these drugs during pregnancy was documented in yes / no format for each participant. These dichotomous indicators of other prenatal exposures were used in analysis.

Measure of Childhood Intelligence

Children’s Full Scale Intelligence Quotient (FSIQ) was assessed at age 6 years using the Wechsler Preschool and Primary Scale of Intelligence – Revised (WPPSI-R) [85].

Measure of Childhood Experience

Children’s home environments were evaluated at age 4 years (4.1 ± 0.2) and 8 years (8.4 ± 0.5) using the Home Observation for Measurement of the Environment (HOME) Inventory [15]. The HOME is a 1-hour structured, in-home parental interview and observational checklist with 8 subscales measuring specific aspects of the child’s home life. In-home assessments were completed by a trained member of the research team, masked to child exposure status. Studies in cognitive neuroscience have shown that environmental characteristics of parental nurturance and environmental are related to neurocognitive outcomes [33]. We have found similar relationships in our cohort, with HOME composites related to both general and specific aspects of cognitive function [31]. We included these important variables in the current analysis. The Environmental Stimulation composite incorporated HOME subscales most analogous to experiential factors that vary with environmental enrichment in animal studies. The Parental Nurturance composite incorporated HOME subscales that measured the warmth and availability of parental care. Composite scores were calculated by averaging the Z scores of relevant subscales for each age. Further details are described in Farah et al [31]. The means of the four and eight year Environmental Stimulation composites and the four and eight year Parental Nurturance composites were used in this report.

Foster Care Placement

At each study visit, between birth and time of NC testing, the child’s primary caregiver was recorded. We found 4 types of primary caregiver in our cohort: biological mother, biological father, kinship foster care and non-kinship foster care. These were then collapsed into two categories: ever in foster care and never in foster care. These categories were used to compute the dichotomous yes/no variable used in analysis.

Measure of Maternal Depression

Children’s primary caregivers were administered the Beck Depression Inventory-Second Edition [7], (BDI-II) during the visit one year prior to the child’s NC evaluation. The BDI-II is an instrument designed to assess depressive symptoms in adults.

2.5. Testing Procedure

The NC task battery, as well as all evaluations included in this report, was administered individually by a licensed psychologist unaware of child exposure status. The neurocognitive evaluation was completed in one study visit lasting an average of 2.5 hours, with 1 or 2 short break periods scheduled within the task administration. Also, at each study visit, demographic information was updated to include information about the current primary caregiver and household environment.

2.6. Statistical Analyses

To reduce the effect of outliers, data from each NC task were winsorized, that is, the two most extreme values at each end of the distribution of all children’s scores were replaced with the third most extreme value at each end.

In order to express performance on a common scale for purposes of analysis, the data for each task were transformed to Z scores defined relative to the distribution of all 120 children. Composite scores for each NC system were then created by averaging the Z scores for the two tasks used to assess each system.

Effects of GCE on NC function were tested using multivariate analysis of covariance (MANCOVA). The seven NC system composite scores were entered as the multivariate dependent variable. Using Multiple Analysis of Variance (MANOVA), we examined the relation between the seven NC system composite scores and individual birth and school-age characteristics shown in Table 1 and Table 2. Table 3 shows the results of univariate testing of the relations between NC outcome and important covariates. Those covariates that were related to outcome at p< 0.20 were subsequently examined using MANCOVA. Of variables measured at birth, this included only prenatal exposure to alcohol. Of those measured during school age years, it included age at testing, foster care placement, caregiver current cocaine use, and the Parental Nurturance HOME composite.

Table 2
Neurocognitive Test Scores by GCE Status*
Table 3
Relationships between Multivariate Neurocognitive Variable and Individual Covariates by MANOVA.

We tested the relation of GCE to NC outcome using a series of MANCOVA models designed to maximize our degrees of freedom for analysis by eliminating nonsignificant covariates (p>.10) after each model analysis (Table 4 Models 1 to 5). Because age is robustly associated with cognitive development, this variable was included in all analyses. GCE status, our main independent variable of interest was also included in all models. While MANOVA showed no statistically significant effects of gender on outcome (see Table 3), we included this variable in model 2 to determine if there were differential effects of GCE on outcomes by gender [8,11,39], a finding previously reported by others. Also, because others have reported an association between cognitive outcomes and prenatal exposure to cigarettes and marijuana [22,37,46,57], and environmental stimulation in the home [31,44,53], we included these variables in our analysis.

Table 4
MANCOVA Models with NC System Composite Scores as Dependent Variable – Multivariate p-values

We first examined the effects of GCE alone, including no covariates in the model (Model 1), followed by a MANCOVA that included age at testing, gender, prenatal exposure to cigarettes, alcohol or marijuana, (Model 2). Next, we added the two composites of childhood experience from the HOME: Environmental Stimulation and Parental Nurturance composites (Model 3), followed by caregiver related variables: foster care placement and caregiver current cocaine use (Model 4). Finally, a dose response effect was examined by analyzing GCE as a continuous rather than categorical variable in the GCE group (Model 5).

2.7. Power Analysis

Assuming a correlation coefficient of 0.4 among the seven NC outcomes and a sample size of 120, there was 80% power (type I error = 0.05) to detect a significant difference between the GCE and Control if there was at least one NC outcome with an effect size of 0.5 or greater. With a sample size of 120, inclusion of up to 6 additional variables is acceptable for MANCOVA analysis of the effects of GCE.


First, we present descriptive statistics for individual NC tasks. The means, standard deviations and ranges of scores for each group in each task are presented in Table 2 for the purpose of indicating that performance differences would not be obscured by ceiling or floor effects. By independent samples t-test there were no differences between groups (data not shown). As shown in Table 2, standard scores for the language measures were in the low average range while working memory and reward processing task scores were in the average range, with all scores falling below the mean for the standardization sample. This pattern of performance is consistent with previous reports on our cohort as well as other groups of disadvantaged children showing poorest performance in verbal tasks.

Subjects were close to the ceiling on one task of visual cognition, the VOSP, as indicated by mean scores of 19.1 out of 20. As this could impair our ability to discern the effects of GCE on performance, data from this task were excluded from the current analysis. Also, due to its late addition to the battery, the Gordon Delay task was available for only 80 subjects. Thus, data from this task were also excluded from the current analysis. For the Visual Cognition and Reward Processing systems, after exclusion of the VOSP and Gordon Delay tasks respectively, the Z score for the single remaining task was used in subsequent data analyses

3.1. Effects of GCE on NC function, with and without covariates

A MANOVA, simply testing the effects of GCE on the seven NC system composite scores, showed no overall effect of GCE on NC function, F(7,108)=1.09, p=0.37 (Table 4, Model 1). After adjustment for multiple comparisons, there were no significant univariate effects of GCE on individual NC systems (Data not shown).

When age at testing, gender and the exposures to cigarettes, alcohol and marijuana were added to the analysis (Table 4, Model 2), there was still no overall effect of GCE, F(7,100)= 0.96, p= 0.47. Age showed a strong effect, F(7,100)= 3.23, p= 0.004, and gender, F(7,100)= 0.88, p= 0.53, showed no effect on NC outcome. There was no interaction effect between GCE and gender, F(7,100)= 0.37, p= 0.92. Exposure to alcohol, F(7,100)= 2.16, p= 0.045, was significantly related to the multivariate outcome; however, prenatal exposure to cigarettes, F(7,100)= 2.027, p= 0.059, showed only a borderline association with the outcomes and prenatal exposure to marijuana, F(7,100)=0.48, p= 0.85, showed no effect on outcome.

Next, we examined environmental covariates using HOME composites. One hundred and sixteen of the 120 subjects who completed the NC evaluation also had HOME scores available for these analyses. In the MANCOVA retaining GCE status, age at testing, prenatal cigarette and alcohol exposure (with gender and the exposure to marijuana removed), we added HOME Parental Nurturance and Environmental Stimulation composites (Table 4, Model 3). No effects of GCE were seen, F(7,97)= .968, p= 0.46. The Environmental Stimulation composite, F(7,97)= 2.51, p= 0.020, as well as age at testing, F(7,97)= 4.36, p< 0.001, was associated with overall NC function. The Parental Nurturance composite was not related to overall NC function, F(7,97)= 0.95, p=0.47, and prenatal alcohol exposure was no longer associated with outcome, F(7,97)=1.56, p= 0.16, in this model.

Finally, we evaluated whether current caregiver variables, foster care placement and caregiver current cocaine use, would impact our outcome measure. Of the 120 children, 34 had a history of foster care placement (See Table 1), of these only 9 (6 GCE and 3 Controls) were ever in non-kinship care with 2 subjects (1 GCE and 1 Control) in non-kinship foster care for the majority of their lives. Because of low prevalence non-kinship care, we collapsed our foster care placement data to reflect two groups: ever in foster care and never in foster care. To investigate these relationships, we performed a MANCOVA retaining GCE status, age at testing, prenatal cigarette exposure, and the HOME Environmental Stimulation composite, excluding the Parental Nurturance composite, gender and prenatal alcohol exposure, and adding foster care placement and caregiver current cocaine use variables to the model (Table 4, Model 4). No overall effect of GCE, F(7,99)= 1.34, p= 0.24, foster care placement, F(7, 99)= 1.51, p= 0.17, or caregiver current cocaine use, F(7, 99)= 1.74, p= 0.11, was seen. Age at testing, F(7,99)= 3.66, p= 0.001, and the HOME Environmental Stimulation composite, F(7, 99)= 2.58, p= 0.017, continued to show a strong effect on outcome. Prenatal cigarette exposure effects were nonsignificant, F(7, 99)= 1.92, p= 0.074.

3.3. Further Analyses

To examine the possibility of a dose effect of cocaine exposure on the multivariate NC outcome, the number of days of GCE (median days of GCE = 99), and age at testing and HOME Environmental Stimulation composite were modeled using MANCOVA. Only GCE subjects were included in this analysis. The cigarette exposure variable, significant at p<.10 in each model analysis, was not included in this model because only 2 of the 55 GCE children had no prenatal cigarette exposure. MANCOVA, with days of cocaine exposure, showed no overall effect of days of GCE on NC function, F(7,43) =0.98, p= 0.46, and in this group, age at testing, F(7,43) =1.55, p= 0.18, and HOME Environmental Stimulation composite, F(7,43) =0.80, p= 0.59, were not related to the outcome (Table 4, Model 5).

Because we did not find effects of GCE on NC outcome, we performed an analysis to determine if this NC battery, designed to tax specific frontal and non-frontal NC systems as shown by neuroimaging and lesion studies, served as a valid index of cognitive ability. We used two indicators of validity, the increase in NC test scores with age and the relation of NC scores to scores on other tests of cognitive functioning [3]. In a MANCOVA including age at testing and Wechsler Full Scale IQ score as covariates, the effects of both age at testing, F(7,96)= 6.01, p< 0.001, and IQ, F(7,96)= 19.23, p<0.001, were seen. Six of the seven NC composites showed a relation to age (all p<0.05, except for the visual cognition composite, p>0.3) and all seven NC composite scores showed a relation to IQ (all p<0.02). These results add to the validation of our NC battery as we have shown a developmentally appropriate increase in cognitive functioning with age and a relation to other tests of cognitive ability [3].


For this cohort of middle school-aged children, we found no evidence of impaired NC function caused by gestational cocaine exposure, despite the fact that our sample size was adequate to detect a statistically and clinically significant difference (effect size of 0.5) and we used a NC battery shown to be sensitive to age and IQ. This is an unexpected result in view of the well-established preclinical effects of cocaine on uterine blood flow and fetal oxygenation [65,87] as well as its effects on developing monaminergic neural systems [26,45,62]. While previous studies in humans have shown specific effects of GCE on language [5,57,67], attention [1,68,72,75] and perceptual reasoning [77], we found no difference between groups even with isolation of specific cognitive systems for evaluation of cocaine effects. Controlling for multiple confounding child and environmental variables also did not reveal any effects of GCE. Since most of the GCE children in our cohort were heavily exposed [82,88] (median days of maternal cocaine use during pregnancy was 99), the lack of GCE effects in our sample is even more striking. Nonetheless, after addressing factors important for isolation of GCE effects on children, we found no difference between children with GCE and Controls.

While we reported early differences between GCE and Controls in birth weight, head circumference, gestational age, as well as maternal age, and later history of foster care placement, we still found no effects of GCE on NC outcome in this sample of low SES children. Our results may reflect, in part, the limited variation in our cohort of gestational age and foster care, two variables known to impact development. These special characteristics of our cohort limited exploration of certain questions. First, the children in this cohort were limited to those born at term or near term so that the serious effects associated with prematurity could be minimized. These children were not at increased risk of developmental problems secondary to prematurity as they were enrolled only if they were greater than or equal to 34 weeks gestational age at birth. Thus, we could not examine the compound risk of GCE and prematurity. We did, however, evaluate effects of gestational age as shown in Table 3 and found no effects in our cohort of children born at term or near term. Second, between birth and age at testing for this report, all but two of the children (1 GCE and 1 Control) who were placed in foster care were adopted by relatives or placed with familiar adults (grandparents, aunts, uncles).

While children placed in non-kinship foster or adoptive care have been shown to have better outcomes than GCE children in kinship placements [77], we were unable to examine such effects in our cohort because of small cell size for the non-kinship foster care category (n=2). Perhaps with a larger sample size we would have had enough variance in caregiver categories to examine such caregiver effects. Future studies with a larger sample of GCE children in non-kinship foster care will allow for much needed investigation of kinship versus non-kinship caregiver effects.

The difficulty of interpreting our null results is somewhat eased in the present case because of our ability to detect effects of age, child IQ, and aspects of earlier childhood experience (HOME composites) on NC function with the same cohort and task battery. According to Anastasi, the relation found here between the NC outcome and age at testing and IQ would be expected for any valid measure of cognitive function [3]. Further, using the NC task battery described here in an investigation of NC correlates of SES, we previously reported an overall SES effect as well as significant group differences (low vs. middle SES) in four of the NC systems [32].

Other investigations of the effect of GCE on cognitive abilities have found differences between cocaine exposed and non-exposed children after controlling for environmental factors known to be associated with cognitive outcomes [6,8,12,24,34,35,57,76,77]. While we were able to consider some of the variables commonly associated with poorer outcomes such as foster care placement, caregiver current cocaine use, and characteristics of the home environment, we did not have data on lead exposure or iron deficiency anemia, which have been shown in other studies to be associated with cognitive outcomes [77]. We also had limited data regarding amount and frequency of use of cigarettes, alcohol and marijuana during pregnancy. In some studies, environmental variables such as these have been reported to have a more powerful effect on a wider range of outcomes than gestational cocaine exposure [9,10,23,71,74]. Our findings notwithstanding, the effects of gestational cocaine exposure that have been identified by other researchers are not to be ignored as the clinical significance of those findings has not yet been determined.

Our results show a strong relation between early home environment and later NC outcome. This finding is consistent with other investigations of longitudinal effects of the home environment on child cognitive and academic outcomes [11,16,29,31,44,52,53,77]. As stated earlier, the HOME evaluations were completed twice, once at age 4 years and again at age 8 years. Our composite was computed using age 4 and 8 year home subscales, thus providing an average measure of the child’s experience. NC testing was completed at age 12 years and it is likely that for some participants characteristics of the home environment may have changed both between and after these assessments. Bradley et al (2001) found that relations between characteristics of the home and cognitive and behavioral outcomes are stronger at younger ages, due to the increasing influence of other environments, such as school. We are limited in our ability to examine concurrent relationships between these types of variables at age 12 because we do not have measures of the home environment or measures of the quality of other environments, such as school at this time. Future studies are needed to examine the relative impact of current versus early home environment on outcomes. While it is true that prenatal exposures increase the risk for poorer outcomes, other variables such as characteristics of the home have significant short term and long term effects on cognitive outcome.

Statistically significant effects of prenatal exposure to cigarettes, alcohol and marijuana on cognitive outcome have reported by other researchers [22,37,46,57]. We show marginally significant effects of prenatal exposure to cigarettes and alcohol on NC outcome in Model 2 and 3 of our analyses. These effects did not remain when important covariates were added to the models due to small sample size and multicollinearity. In our cohort only two of the CGE children were not exposed to cigarettes during pregnancy and only two of the Control children were exposed to marijuana as shown in Table 1. Other studies of cocaine exposed children have shown this pattern of polysubstance exposure [5,30,77]. Interaction effects can not be reliably evaluated when cell sizes for these various subsets of the GCE and Control groups are so small. With a larger group of participants it is more likely that we would have been able to evaluate these interactions.

Our results may have been affected by the early attrition in our cohort. While analysis of differences in birth characteristics between subjects retained and those lost to follow up did not reveal any differences in the measured variables, it may be that the children lost to follow-up are different in characteristics not measured, such as foster care placement or caregiver current cocaine use. It also is possible that with a higher retention we would have been able to examine non-kinship foster care effects in our cohort. As noted earlier, within the group of 120 included in this report only two were in non-kinship care, leaving too few subjects to evaluate the interesting effects reported by others [77].

The present results leave many questions unanswered, among them the reason for the discrepancy between the effects of GCE in humans and preclinical models. Furthermore, to the extent that our findings represent good news by suggesting that children with GCE may not experience NC impairments as a result of their exposure, the good news must be regarded as very tentative. We are tentative because some prenatal insults do not manifest themselves until maturity or later [36,63,78] as brain development in humans does not cease until at least the early twenties [20,41,42,59,69]. In fact, recent imaging analyses on this cohort show decreased volume in the caudate, a dopaminergically mediated brain area [4]. The possibility remains that the GCE and Control children of the present cohort may yet diverge in their NC development. Finally, the effects of GCE may be less apparent in simple protocols employing the assessment of cognitive function, such as the one used by us, than in more complex and stressful situations requiring the regulation of arousal and emotion along with cognitive performance [61]. For example, it may be that pairing a stress inducing protocol, such as the Trier Social Stress test [54] with a task of working memory [58,86], may reveal such an effect. Nevertheless, our results are informative in that they address NC functioning at one of the oldest ages so far reported in the GCE literature, and in that NC functioning was assessed by tasks that have been shown in other studies to engage specific NC systems. Replication of this type of investigation using a larger cohort of participants is needed to confirm our results. At present, results reported here both add support to the finding that GCE is not as potent a risk factor for child NC development as anticipated, as well as add to the literature identifying postnatal environmental factors as consistently potent influences on cognitive development.


The project described was supported by NIDA RO1DA18913 and Grant Number UL1RR024134 from the National Center For Research Resources. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Center For Research Resources or the National Institutes of Health.


Beck Depression Inventory - II
Cambridge Neuropsychological Test Automated Battery
Full Scale Intelligence Quotient
Gestational Cocaine Exposure
Home Observation for Measurement of the Environment
Intra Dimensional / Extra Dimensional
Multivariate Analysis of Covariance
Multivariate Analysis of Variance
Peabody Picture Vocabulary Test
Socioeconomic Status
Test for the Reception of Grammar
Visual Object & Space Perception
Wechsler Preschool and Primary Scale of Intelligence – Revised Edition


Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


1. Accornero VH, Amado AJ, Morrow CE, Xue L, Anthony JC, Bandstra ES. Impact of prenatal cocaine exposure on attention and response inhibition as assessed by continuous performance tests. J Dev Behav Pediatr. 2007;28:195–205. [PMC free article] [PubMed]
2. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Text Revision. Fourth Edition. Washington, DC: American Psychiatric Association; 2000.
3. Anastasi A. Psychological Testing. Macmillan Publishing Company; 1988.
4. Avants BB, Hurt H, Giannetta JM, Epstein CL, Shera DM, Rao H, Wang J, Gee JC. Effects of heavy in utero cocaine exposure on adolescent caudate morphology. Pediatr Neurol. 2007;37:275–279. [PubMed]
5. Bandstra ES, Morrow CE, Vogel AL, Fifer RC, Ofir AY, Dausa AT, Xue L, Anthony JC. Longitudinal influence of prenatal cocaine exposure on child language functioning. Neurotoxicol Teratol. 2002;24:297–308. [PubMed]
6. Bandstra ES, Vogel AL, Morrow CE, Xue L, Anthony JC. Severity of prenatal cocaine exposure and child language functioning through age seven years: a longitudinal latent growth curve analysis. Subst Use Misuse. 2004;39:25–59. [PMC free article] [PubMed]
7. Beck AT, Steer RA, Brown GK. Beck Depression Inventory II, Manual. San Antonio, TX: The Psychological Corporation; 1996.
8. Beeghly M, Martin B, Rose-Jacobs R, Cabral H, Heeren T, Augustyn M, Bellinger D, Frank DA. Prenatal cocaine exposure and children's language functioning at 6 and 9.5 years: moderating effects of child age, birthweight, and gender. J Pediatr Psychol. 2006;31:98–115. [PMC free article] [PubMed]
9. Bellinger D, Leviton A, Waternaux C, Needleman H, Rabinowitz M. Longitudinal analyses of prenatal and postnatal lead exposure and early cognitive development. N Engl J Med. 1987;316:1037–1043. [PubMed]
10. Bellinger D, Sloman J, Leviton A, Rabinowitz M, Needleman HL, Waternaux C. Low-level lead exposure and children's cognitive function in the preschool years. Pediatr. 1991;87:219–227. [PubMed]
11. Bennett DS, Bendersky M, Lewis M. Children's cognitive ability from 4 to 9 years old as a function of prenatal cocaine exposure, environmental risk, and maternal verbal intelligence. Dev Psychol. 2008;44:919–928. [PMC free article] [PubMed]
12. Bennett DS, Bendersky M, Lewis M. Children's intellectual and emotional-behavioral adjustment at 4 years as a function of cocaine exposure, maternal characteristics, and environmental risk. Dev Psychol. 2002;38:648–658. [PMC free article] [PubMed]
13. Bishop DVM. The Test for Reception of Grammar. University of Manchester, Age and Cognitive Performance Research Centre; 1983.
14. Booth J, MacWhinney B, et al. Functional organization of activation patterns in children: fMRI during 3 different cognitive tasks. Prog Neuropsychopharmacol Biol Psychiatry. 1999;23:271–294. [PubMed]
15. Bradley RH. The HOME Inventory: Review and reflections. In: Reese HW, editor. Advances in Child Development and Behavior. San Diego: Academic Press, Inc; 1994. pp. 241–288. [PubMed]
16. Bradley RH, Convyn RF, Burchinal M, McAdoo HP, Coll CG. The home environments of children in the United States part II: relations with behavioral development through age thirteen. Child Dev. 2001;72:1868–1886. [PubMed]
17. Bush G, Whalen PJ, Rosen BR, Jenike MA, McInerney SC, Rauch SL. The Counting Stroop: an interference task specialized for functional neuroimaging-validation study with functional MRI. Hum Brain Map. 1998;6:270–282. [PubMed]
18. Cambridge Cognition L. Cambridge Neuropsychological Test Automated Battery (CANTAB) Cambridge, UK: 1999.
19. Casey BJ, Cohen JD, Jezzard P, Turner R, Noll DC, Trainor RJ, Giedd J, Kaysen D, Hertz-Pannier L, Rapoport JL. Activation of prefrontal cortex in children during a nonspatial working memory task with functional MRI. NeuroImage. 1995;2:221–229. [PubMed]
20. Casey BJ, Getz S, Galvan A. The adolescent brain. Dev Review. 2008;28:62–77. [PMC free article] [PubMed]
21. Casey BJ, Trainor JL, Schubert AB, Nystrom LE, Giedd JN, Castellanos FX, Haxby JV, Noll DC, Cohen JD, Forman SD, et al. A developmental functional MRI study of prefrontal activation during performance of a go-no-go task. J Cog Neurosci. 1997;9:835–847. [PubMed]
22. Cornelius MD, Ryan CM, Day NL, Goldschmidt L, Willford JA. Prenatal tobacco effects on neuropsychological outcomes among preadolescents. J Dev Behav Pediatr. 2001;22:217–225. [PubMed]
23. Coscia JM, Christensen BK, Henry RR, Wallston K, Radcliffe J, Rutstein R. Effects of home environment, socioeconomic status, and health status on cognitive functioning in children with HIV-1 infection. J Pediatr Psychol. 2001;26:321–329. [PubMed]
24. Delaney-Black V, Covington C, Templin T, Kershaw T, Nordstrom-Klee B, Ager J, Clark N, Surendran A, Martier S, Sokol RJ. Expressive language development of children exposed to cocaine prenatally: literature review and report of a prospective cohort study. J Commun Disord. 2000;33:463–480. quiz 480-1. [PubMed]
25. Delaney-Black V, Covington CY, Nordstrom-Klee B, Sokol RJ. Prenatal cocaine exposure as a risk factor for later developmental outcomes. JAMA. 2001;286:46–47. [PubMed]
26. Dow-Edwards D, Mayes L, Spear L, Hurd Y. Cocaine and development: clinical, behavioral, and neurobiological perspectives - A symposium report. Neurotoxicol Teratol. 1999;21:481–490. [PubMed]
27. Dunn LM, Dunn LM. Examiner's Manuel for the Peabody Picture Vocabulary Test-III. Circle Pines, MN: American Guidance Service, Inc.; 1997.
28. Ekstrom R, French JW, Harman HH. Manual for Kit of Factor-Referenced Cognitive Tests. Princeton, NJ: Educational Testing Service; 1976.
29. Evans GW, Schamberg MA. Childhood poverty, chronic stress, and adult working memory. Proc Natl Acad Sci U S A. 2009;106:6545–6549. [PubMed]
30. Eyler FD, Behnke M, Conlon M, Woods NS, Wobie K. Birth outcome from a prospective, matched study of prenatal crack/cocaine use: I. Interactive and dose effects on health and growth. Pediatr. 1998;101:229–237. [PubMed]
31. Farah M, Betancourt L, Shera D, Savage J, Giannetta J, Brodsky N, Malmud EK, Hurt H. Environmental Stimulation, Parental Nurturance and Cognitive Development in Humans. Dev Sci. 2008;11:793–801. [PubMed]
32. Farah MJ, Shera DM, Savage JH, Betancourt L, Giannetta JM, Brodsky NL, Malmud EK, Hurt H. Childhood poverty: Specific associations with neurocognitive development. Brain Res. 2006;1110:166–174. [PubMed]
33. Francis D, Diorio J, Liu D, Meaney MJ. Nongenomic transmission across generations of maternal behavior and stress responses in the rat. Science. 1999;286:1155–1158. [PubMed]
34. Frank DA, Augustyn M, Knight WG, Pell T, Zuckerman B. Growth, development, and behavior in early childhood following prenatal cocaine exposure: a systematic review. JAMA. 2001;285:1613–1625. [PMC free article] [PubMed]
35. Frank DA, Rose-Jacobs R, Beeghly M, Wilbur M, Bellinger D, Cabral H. Level of prenatal cocaine exposure and 48-month IQ: importance of preschool enrichment. Neurotoxicol Teratol. 2005;27:15–28. [PubMed]
36. Fried PA. Conceptual issues in behavioral teratology and their application in determining long-term sequelae of prenatal marihuana exposure. J Child Psychol Psychiatry. 2002;43:81–102. [PubMed]
37. Fried PA, Smith AM. A literature review of the consequences of prenatal marihuana exposure. An emerging theme of a deficiency in aspects of executive function. Neurotoxicol Teratol. 2001;23:1–11. [PubMed]
38. Gabriel M, Taylor C. Prenatal exposure to cocaine impairs neuronal coding of attention and discriminative learning. Ann N Y Acad Sci. 1998;846:194–212. [PubMed]
39. Gendle MH, Strawderman MS, Mactutus CF, Booze RM, Levitsky DA, Strupp BJ. Impaired sustained attention and altered reactivity to errors in an animal model of prenatal cocaine exposure. Brain Res Dev Brain Res. 2003;147:85–96. [PubMed]
40. Gendle MH, White TL, Strawderman M, Mactutus CF, Booze RM, Levitsky DA, Strupp BJ. Enduring effects of prenatal cocaine exposure on selective attention and reactivity to errors: evidence from an animal model. Behav Neurosci. 2004;118:290–297. [PubMed]
41. Giedd JN, Bloomenthal J. Brain development during childhood and adolescence: A longitudinal MRI study. Nat Neurosci. 1999;2:861–863. [PubMed]
42. Giedd JN, Snell JW, Lange N, Rajapakse JC, Casey BJ, Kozuch PL, Vaituzis AC, Vauss YC, Hamburger SD, Kaysen D, et al. Quantitative magnetic resonance imaging of human brain development: ages 4–18. Cereb Cortex. 1996;6:551–560. [PubMed]
43. Gordon M, McClure FD, Aylward GP. The Gordon Diagnostic System: Instruction Manual and Interpretive Guide. DeWitt, NY: Gordon System, Inc.; 1996.
44. Gottfried AE, Fleming JS, Gottfried AW. Role of cognitively stimulating home environment in children's academic intrinsic motivation: a longitudinal study. Child Dev. 1998;69:1448–1460. [PubMed]
45. Harvey JA. Cocaine effects on the developing brain: current status. Neurosci Biobehav Rev. 2004;27:751–764. [PubMed]
46. Huizink AC, Mulder EJ. Maternal smoking, drinking or cannabis use during pregnancy and neurobehavioral and cognitive functioning in human offspring. Neurosci Biobehav Rev. 2006;30:24–41. [PubMed]
47. Hurt H, Brodsky NL, Betancourt L, Braitman LE, Malmud E, Giannetta J. Cocaine-exposed children: follow-up through 30 months. J Dev Behav Pediatr. 1995;16:29–35. [PubMed]
48. Hurt H, Brodsky NL, Braitman LE, Giannetta J. Natal status of infants of cocaine users and control subjects: a prospective comparison. J Perinatol. 1995;15:297–304. [PubMed]
49. Hurt H, Brodsky NL, Roth H, Malmud E, Giannetta JM. School performance of children with gestational cocaine exposure. Neurotoxicol Teratol. 2005;27:203–211. [PubMed]
50. Hurt H, Malmud E, Betancourt L, Braitman LE, Brodsky NL, Giannetta J. Children with in utero cocaine exposure do not differ from control subjects on intelligence testing. Arch Pediatr Adolesc Med. 1997;151:1237–1241. [PubMed]
51. Hurt H, Malmud E, Betancourt L, Brodsky NL, Giannetta J. A prospective evaluation of early language development in children with in utero cocaine exposure and in control subjects. J Pediatr. 1997;130:310–312. [PubMed]
52. Hurt H, Malmud E, Braitman LE, Betancourt LM, Brodsky NL, Giannetta JM. Inner-city achievers: who are they? Arch Pediatr Adolesc Med. 1998;152:993–997. [PubMed]
53. Jimerson S, Egeland B, Teo A. A longitudinal study of Achievement trajectories: Factors associated with change. J Ed Psychol. 1999;91:116–126.
54. Kirschbaum C, Pirke KM, Hellhammer DH. The 'Trier Social Stress Test'--a tool for investigating psychobiological stress responses in a laboratory setting. Neuropsychobio. 1993;28:76–81. [PubMed]
55. LaGasse LL, Seifer R, Lester BM. Interpreting research on prenatal substance exposure in the context of multiple confounding factors. Clin Perinatol. 1999;26:39–54. vi. [PubMed]
56. Levine TP, Liu J, Das A, Lester B, Lagasse L, Shankaran S, Bada HS, Bauer CR, Higgins R. Effects of prenatal cocaine exposure on special education in school-aged children. Pediatr. 2008;122:e83–e91. [PMC free article] [PubMed]
57. Lewis BA, Kirchner HL, Short EJ, Minnes S, Weishampel P, Satayathum S, Singer LT. Prenatal cocaine and tobacco effects on children's language trajectories. Pediatr. 2007;120:e78–e85. [PubMed]
58. Luethi M, Meier B, Sandi C. Stress effects on working memory, explicit memory, and implicit memory for neutral and emotional stimuli in healthy men. Front Behav Neurosci. 2008;2:5. [PMC free article] [PubMed]
59. Luna B, Thulborn KR, Munoz DP, Merriam EP, Garver KE, Minshew NJ, Keshavan MS, Genovese CR, Eddy WF, Sweeney JA. Maturation of widely distributed brain function subserves cognitive development. NeuroImage. 2001;13:786–793. [PubMed]
60. Mayes AR, Meudell PR, Neary D. Do amnesiacs adopt inefficient encoding strategies with faces and random shapes? Neuropsychologia. 1980;18:527–541. [PubMed]
61. Mayes LC. A behavioral teratogenic model of the impact of prenatal cocaine exposure on arousal regulatory systems. Neurotoxicol Teratol. 2002;24:385–395. [PubMed]
62. Mayes LC. Developing brain and in utero cocaine exposure: effects on neural ontogeny. Dev Psychopathol. 1999;11:685–714. [PubMed]
63. Mayes LC, Fahy T. Prenatal Drug Exposure and Cognitive Development. In: Sternberg RJ, Grigorenko EL, editors. Environmental Effects on Cognitive Abilities. Mahwah, New Jersey: Lawrence Erlbaum Associates; 2001. pp. 189–219.
64. Mooney CM. Age in the development of closure ability in children. Can J Psychol. 1957;11:219–226. [PubMed]
65. Moore TR, Sorg J, Miller L, Key TC, Resnik R. Hemodynamic effects of intravenous cocaine on the pregnant ewe and fetus. Am J Obstet Gynecol. 1986;155:883–888. [PubMed]
66. Morrow CE, Culbertson JL, Accornero VH, Xue L, Anthony JC, Bandstra ES. Learning disabilities and intellectual functioning in school-aged children with prenatal cocaine exposure. Dev Neuropsychol. 2006;30:905–931. [PMC free article] [PubMed]
67. Morrow CE, Vogel AL, Anthony JC, Ofir AY, Dausa AT, Bandstra ES. Expressive and receptive language functioning in preschool children with prenatal cocaine exposure. J Pediatr Psychol. 2004;29:543–554. [PMC free article] [PubMed]
68. Noland JS, Singer LT, Short EJ, Minnes S, Arendt RE, Kirchner HL, Bearer C. Prenatal drug exposure and selective attention in preschoolers. Neurotoxicol Teratol. 2005;27:429–438. [PubMed]
69. Paus T. Mapping brain maturation and cognitive development during adolescence. Trends Cogn Sci. 2005;9:60–68. [PubMed]
70. Ratcliff G. Spatial thought, mental rotation and the right cerebral hemisphere. Neuropsychologia. 1979;17:49–54. [PubMed]
71. Rauh VA, Whyatt RM, Garfinkel R, Andrews H, Hoepner L, Reyes A, Diaz D, Camann D, Perera FP. Developmental effects of exposure to environmental tobacco smoke and material hardship among inner-city children. Neurotoxicol Teratol. 2004;26:373–385. [PMC free article] [PubMed]
72. Richardson GA, Conroy ML, Day NL. Prenatal cocaine exposure: effects on the development of school-age children. Neurotoxicol Teratol. 1996;18:627–634. [PubMed]
73. Rivkin MJ, Davis PE, Lemaster JL, Cabral HJ, Warfield SK, Mulkern RV, Robson CD, Rose-Jacobs R, Frank DA. Volumetric MRI study of brain in children with intrauterine exposure to cocaine, alcohol, tobacco, and marijuana. Pediatrics. 2008;121:741–750. [PMC free article] [PubMed]
74. Sameroff AJ, Seifer R, Baldwin A, Baldwin C. Stability of intelligence from preschool to adolescence: the influence of social and family risk factors. Child Dev. 1993;64:80–97. [PubMed]
75. Savage J, Brodsky NL, Malmud E, Giannetta JM, Hurt H. Attentional functioning and impulse control in cocaine-exposed and control children at age ten years. J Dev Behav Pediatr. 2005;26:42–47. [PubMed]
76. Singer LT, Minnes S, Short E, Arendt R, Farkas K, Lewis B, Klein N, Russ S, Min MO, Kirchner HL. Cognitive outcomes of preschool children with prenatal cocaine exposure. JAMA. 2004;291:2448–2456. [PubMed]
77. Singer LT, Nelson S, Short E, Min MO, Lewis B, Russ S, Minnes S. Prenatal cocaine exposure: drug and environmental effects at 9 years. J Pediatr. 2008;153:105–111. [PMC free article] [PubMed]
78. Sobrian SK, Marr L, Ressman K. Prenatal cocaine and/or nicotine exposure produces depression and anxiety in aging rats. Prog Neuropsychopharmacol Biol Psychiatry. 2003;27:501–518. [PubMed]
79. Spear LP, Kirstein CL, Frambes NA. Cocaine effects on the developing central nervous system: behavioral, psychopharmacological, and neurochemical studies. Ann NY Acad Sci. 1989;562:290–307. [PubMed]
80. Stanwood GD, Levitt P. Prenatal cocaine exposure as a risk factor for later developmental outcomes. JAMA. 2001;286(45):46–47. author reply. [PubMed]
81. Tronick EZ, Beeghly M. Prenatal cocaine exposure, child development, and the compromising effects of cumulative risk. Clin Perinatol. 1999;26:151–171. [PubMed]
82. Tronick EZ, Frank DA, Cabral H, Zuckerman BS. A dose response effect to in utero cocaine exposure on infant neurobehavioral functioning. Pediatr Res. 1994;35:28A.
83. Warner TD, Behnke M, Eyler FD, Padgett K, Leonard CM, Hou W, Garavan CW, Schmalfuss I, Blackband S. Diffusion Tensor Imaging of Frontal White Matter and Executive Functioning in Cocaine-Exposed Children. Pediatrics. 2006;118:2014–2024. [PMC free article] [PubMed]
84. Warrington EK, James M. Visual Object and Space Perception Battery. Suffolk, England: Thames Valley Test Co.; 1991.
85. Wechsler D. The Wechsler Preschool and Primary Scale of Intelligence- Revised. San Antonio, TX: The Psychological Corp; 1989.
86. Wolf OT, Schommer NC, Hellhammer DH, McEwen BS, Kirschbaum C. The relationship between stress induced cortisol levels and memory differs between men and women. Psychoneuroendocrino. 2001;26:711–720. [PubMed]
87. Woods JR, Jr, Plessinger MA, Clark KE. Effect of cocaine on uterine blood flow and fetal oxygenation. JAMA. 1987;257:957–961. [PubMed]
88. Zuckerman B, Frank DA, Hingson R, Amaro H, Levenson SM, Kayne H, Parker S, Vinci R, Aboagye K, Fried LE, et al. Effects of maternal marijuana and cocaine use on fetal growth. N Engl J Med. 1989;320:762–768. [PubMed]