- Email: [email protected]
Motor response selection in children with fetal alcohol spectrum disorders
Neurotoxicology and Teratology 28 (2006) 278 – 285 www.elsevier.com/locate/neutera
Motor response selection in children with fetal alcohol spectrum disorders Roger W. Simmons a,⁎, Jennifer D. Thomas b , Susan S. Levy a , Edward P. Riley b a
Motor Control Laboratory, Department of Exercise and Nutritional Sciences, San Diego State University, San Diego, CA 92182, USA b Center for Behavioral Teratology, Department of Psychology, San Diego State University, San Diego, CA 92120, USA Received 26 July 2005; received in revised form 6 December 2005; accepted 10 January 2006 Available online 9 March 2006
Abstract Previous work has reported timing delays in motor response selection in children with prenatal exposure to alcohol when the information load involved responding to two stimulus choices. The present study examined whether the delay in response selection extends to conditions in which the information load is increased to four and eight stimulus choices. Twenty children aged between 12 and 17 years with fetal alcohol spectrum disorders (FASD) were compared to 17 non-alcohol-exposed controls (NC) on a reaction time (RT) task involving 1, 2, 4 or 8 visual stimulus choices. The task demands required the participant to release a response key as fast as possible when the stimulus light electronically paired with the response key was activated. With the number of stimulus choices expressed on a logarithmic scale, there was a significant and linear increase in RT for the FASD children as predicted by information processing theory. Additionally, the increase in RT for the FASD group was comparable to that observed for the NC children at each level of stimulus choice examined. It was concluded that FASD adolescents require additional time to process increasing amount of information, but that the time required for motor response selection is not delayed relative to control group performance. © 2006 Elsevier Inc. All rights reserved. Keywords: Response selection; Fetal alcohol spectrum disorders; Reaction time
1. Introduction Fetal alcohol syndrome (FAS) is defined by a triad of characteristics of pre- and postnatal growth retardation, central nervous system (CNS) dysfunction and a specific pattern of facial features [26]. FAS children experience a range of cognitive and behavioral dysfunctions, as do children with histories of prenatal exposure to alcohol (PEA) but who do not have the physical anomalies associated with FAS [34,36]. In addition to cognitive and intellectual problems [50], both FAS and PEA children prenatally exposed to alcohol (PEA) also manifest poor motor skills, including atypical motor timing [49,58]. According to the constructs of information processing theory, one determinant of motor timing is the time required to make a decision to respond to an external stimulus, which is known as motor response selection [47].
⁎ Corresponding author. Tel.: +1 619 594 5543; fax: +1 619 594 6553. E-mail address: [email protected] (R.W. Simmons). 0892-0362/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.ntt.2006.01.008
The history of investigating response selection processing can be traced back to 1885 when Merkel [37] found a negatively accelerated function between reaction time (RT) and the number of stimulus choices. Many years later, it was reported that the RT function is linear when the number of stimulus choices is expressed on a log2 scale (in units of bits) [20], with the slope of the function reflecting delays in the speed of CNS processing as information load increases. The predicted relationship between RT and the number of stimulus choices is known as Hick's law [20] and, although some exceptions have been reported [31,61], the law has proven to be robust and generalizes to a variety of conditions and tasks using non-clinical human participants [7,14,23,25,51,57] and certain animal species [58]. Hick's law has also been tested using clinical groups including Alzheimer and Parkinson's patients [16,21,32], melancholic depressives [44] and non-melancholic depressives [4], schizophrenics [29], mentally retarded children [59] and adults [6], individuals with multiple sclerosis [27], and children prenatally exposed to alcohol [49]. In the context of prenatal alcohol exposure, anomalies in the cerebrum [46], cerebellum [3,35,45,46], corpus callosum [26,42,62] and basal ganglia [13,34,42,62] resulting from the
R.W. Simmons et al. / Neurotoxicology and Teratology 28 (2006) 278–285
teratogenic effects of alcohol are thought to slow decision making speed, as estimated by RT. For example, one group of RT studies required alcohol-exposed children to view continuous visual displays and respond as fast as possible when a target stimulus randomly appeared within the context of the display. Results of this work have provided limited support for the existence of RT delays in alcohol-exposed children [12,53–56]. Continuous display studies, however, involve the child attending to a visual stimulus for an extended period of time (e.g. 12–15 min), which is conceptually different from traditional measures of simple reaction time (SRT), involving one stimulus-response pairing, and choice reaction time (CRT), involving multiple stimulus-response pairings, in which the stimulus and response occur in a relatively brief time period (e.g. 1–4 s). To date, only one study has used this type of RT paradigm to investigate information processing time in children with prenatal alcohol exposure. Results revealed comparable SRT values for the alcohol-exposed children and controls, but the alcohol-exposed group was significantly delayed during a two-stimulus CRT task [49]. Additionally, the investigators fractionated SRT and CRT into premotor RT and motor RT. Premotor RT defines the time between the onset of the stimulus and the onset of neuromuscular activity and provides a measure of the time taken to complete CNS processing [41]. In contrast, motor RT, defined as the time between the onset of neuromuscular activity as indexed by electromyographic (EMG) activity and the first observable movement (the release of the RT key), indexes electromechanical delays inherent to the peripheral neuromuscular system. The advantage of fractionating SRT and CRT lies in generating an estimate of central processing time minus delays associated with the initiation of neuromuscular activity. Results revealed comparable premotor SRT for both groups, but the alcohol-exposed group produced a significantly longer premotor CRT. This latter finding confirmed the existence of significant CNS processing deficits in the alcohol-exposed child when information load (i.e. the number of stimulus choices) was doubled. A similar delay in reaction time has been reported for alcohol-exposed children when responding to higher cognitive loads of a Sternberg task [9]. The alcohol-exposed group was also significantly delayed in motor RT for both the SRT and CRT conditions, thereby indicating additional RT delays at the peripheral neuromuscular level. What remains unknown is whether the response selection delay reported for the alcohol-exposed children during the two CRT task extends to conditions with additional levels of information processing, and whether the resulting RT function is linear as proposed by motor response selection theory. Based on our previous work with two stimulus choices, and the expectation that additional information processing will also involve alcohol impacted CNS structures, it is predicted that adolescent children prenatally exposed to alcohol will manifest a linear increase in RT as the number of stimuli double (i.e. increases of 1-bit of information), and these children will be significantly slower in responding during multiple stimulus conditions than non-alcohol-exposed controls.
279
2. Methods 2.1. Participants Two groups of children between the ages of 12 and 17 years, with full-scale IQs (FSIQ) above 70 served as participants in the study. This age group was chosen for their ability to sustain attention to the multiple repetitions of the stimulus conditions and because they possess levels of fine motor control sufficient to independently move the four fingers of each hand during multiple stimulus choice conditions. One group of children was recruited from a pool of participants registered at the Center of Behavioral Teratology (CBT) at San Diego State University. These children had previously been evaluated with a comprehensive neuropsychological battery (including intelligence quotient testing—IQ) and by a pediatric dysmorphologist (Kenneth Lyons Jones, MD). The average time between IQ assessment and participation in the experiment was 2.2 years (S.D. ± 2.0). Using established criteria of specific craniofacial anomalies (e.g. short palpebral fissures, long smooth philtrum, broad nasal bridge and epicanthic folds), pre- and/or postnatal growth deficiency and CNS dysfunction [25], six children received a diagnosis of fetal alcohol syndrome (FAS) by the dysmorphologist. An additional 14 children who did not present any of the triad of characteristics associated with FAS but who had experienced in utero exposure to alcohol [36] were classified by the dysmorphologist as having PEA. Nine of these children presented evidence of having one of the facial characteristics associated with FAS, while another child experienced growth deficiency. Although specific details regarding the degree and frequency of prenatal alcohol exposure were not always available, abusive use of alcohol by each child's mother was confirmed through maternal self-reports and access to medical or social service records. Statistical analyses revealed the FAS and PEA children to have comparable RTs for all levels of information load. This result is consistent with previous work reporting similarity of motor performance [43] for the two groups. Consequently, the FAS and PEA children were combined into a single alcohol-exposed group, which falls under the umbrella term of fetal alcohol spectrum disorders (FASD) [40,52]. This combined alcohol-exposed group will be indicated in the present study by the abbreviation FASD. A second group of 17 children aged 12 and 17 years inclusively, comprised the non-alcohol-exposed control group (NC) and were also recruited from a pool of participants registered at the CBT. All children in the NC group had previously been tested using the same evaluation scales used for the alcohol-exposed children. Evaluation involved a comprehensive neuropsychological battery and screening for prenatal alcohol and other teratogenic agents through a telephone interview and completion of a parent questionnaire that included information concerning alcohol consumption during pregnancy. An average of 2.7 years (S.D. ± 2.3) occurred between IQ assessment and completion of the task for the NC group. Potential control participants were excluded for the following reasons: prenatal exposure to alcohol or other known teratogens, closed head injury with significant loss of memory, a
280
R.W. Simmons et al. / Neurotoxicology and Teratology 28 (2006) 278–285
FSIQ score below 70 or a diagnosis of attention deficit hyperactivity disorder (ADHD). ADHD children were excluded from the FASD and NC groups because of the purported link between ADHD and structural damage in the CNS [17]. None of the control participants had any known neurological problems or exposure to teratogens. Therefore, 20 children were included in the FASD group and 17 in the NC group. The NC group was closely matched to the FASD children on age, gender, socioeconomic status (measured using the Hollingshead Four Factor Index, 1975) and ethnicity. Demographic information for the FASD and NC groups is presented in Table 1. 2.2. Apparatus The RT apparatus consisted of two rectangular response modules (21 × 15 × 8cm) placed adjacent to each other on a table in front of the seated participant. Positioned on top of each response module were four square (1.5 × 1.5 cm) response buttons arranged 2.5 cm apart in an ergonomically compatible semi-circular pattern. The top surface of each response module was inclined towards the participant at an angle of 20° to the horizontal plane. With the base of the palm resting on a wrist pad, the participants could comfortably place the four fingers of each hand on the designated response buttons. Each response button was electronically paired with a color-coded circular stimulus light (1.5 cm in diameter) located 3 cm above the response button on the top surface of the response module. There were four stimulus light-response button pairings for each response module resulting in a maximum of eight stimulus-response alternatives. For testing purposes, response buttons were numbered 1 through 8 and corresponded to the eight fingers beginning with the left little finger of the participant's right hand through the little finger of the left hand, respectively. The thumb of each hand was not used for responding. The participant was unaware of this numbering system and the
experimenter used this key designation only to determine the order of stimulus light presentation. The experimenter sat opposite the participant and controlled activation of the stimulus lights from two remote control panels hidden from the participant's view by a wooden screen. 2.3. Procedures After obtaining informed consent from the legal guardian and assent from the child, two tests were completed by the participant, one of which was the response selection test. The second test was conducted as part of an ongoing research project at the CBT and did not interfere with performance of the response selection test. The presentation order of the two tests was systematically counterbalanced across subjects. An interval of approximately 7 min separated the two tests and eliminated any fatigue resulting from completion of the first test. Participants were seated in front of the response modules with the seat height adjusted to a position where the child could comfortably place the finger(s) on the appropriate response key (s). Prior to testing, participants were provided task related instructions and a demonstration, and allowed a maximum of five practice trials for each of the four-stimulus conditions examined. The experimenter provided a verbal ‘ready’ command followed by a randomly determined foreperiod of 1–4 s. The task consisted of the participants holding down 1, 2, 4 or 8 response keys with their fingers and releasing a response key as fast as possible when the stimulus light paired with the response key was activated. Release of the response key (as opposed to depressing the key) controlled for any time delay associated with finger movement. The elapsed time between stimulus light activation and lifting the finger from the response key defined RT and was measured to the nearest millisecond using two electronic timers. The 1, 2, 4 and 8 stimulus light choices represented four-stimulus conditions corresponding to
Table 1 Demographic information by group (mean ± S.D.)
Gender (M, %/F, %) Age, M (S.D.) Range FSIQ a, M (S.D.) Range VIQ a, M (S.D.) Range PIQ a, M (S.D.) Range Hollingshead Score b Hand dominance (L, %/R, %) Ethnicity (n, %) Asian African American Caucasian Native American Hispanic descent
FASD (N = 20)
NC (N = 17)
9 (45.0%):11 (55.0%) 14.2 (1.4) 12.0–17.0 88.7 (10.0) 70–109 87.6 (11.8) 58–101 92.0 (13.3) 69–119 46.9 (7.3) 4 (20%):16 (80%)
6 (35.3%):11 (69.2%) 14.2 (1.7) 12.0–17.75 101.2 (9.9) 84–118 104.6 (9.9) 85–119 97.4 (12.7) 79–121 49.9 (13.0) 0 (0%):17 (100%)
2 (10.0%) 2 (10.0%) 10 (50.0%) 2 (10.0%) 4 (20.0%)
1 (5.9%) 4 (23.5%) 9 (52.9%) 0 (0%) 1 (17.6%)
p N 0.54 p N 0.87 p b 0.002 p b 0.002 p N 0.22 p N 0.27
a Intelligence scores were derived from either the Wechsler Preschool and Primary Scales of Intelligence Revised or the Wechsler Intelligence Scale for Children-III depending on the child's age at the study entry. b Socioeconnomic status was estimated using the Hollingshead Four Factor Index of Social Status (Hollingshead, 1975, unpublished data).
R.W. Simmons et al. / Neurotoxicology and Teratology 28 (2006) 278–285
FSIQ
VIQ
PIQ
FASD group 0 bit 1 bit 2 bits 3 bits
−0.352 −0.328 −0.400 −0.126
− 0.133 − 0.256 − 0.233 0.051
−0.423 −0.283 −0.422 −0.221
NC group 0 bit 1 bit 2 bits 3 bits
−0.345 −0.502 * −0.497 * −0.482 *
− 0.315 − 0.188 − 0.176 0.048
−0.194 −0.505 ⁎ −0.516 ⁎ −0.664 ⁎⁎
⁎ p b 0.05. ⁎⁎ p b 0.01.
information loads of 0, 1, 2 and 3 bits of information, respectively. Participants completed a block of 24 trials for each stimulus condition. In the CRT conditions (i.e. 1, 2 or 3 bits of information), the order of stimulus light activation was randomized within each trial block with each light being presented with equal frequency. The presentation order of the four-stimulus conditions was counterbalanced across participants with a 2-min rest period between trial blocks. When responding during the 0-, 1-, 2- and 3-bit stimulus conditions, the participants used either the right index finger (or left, if the individual was left-handed), the two index fingers, the index and middle fingers of each hand, or all four fingers of each hand, respectively. A mistrial was considered to have occurred if the participant lifted the wrong finger or lifted more than one finger in response to a stimulus light being activated. Mistrials were repeated at the end of the trial block. Mistrials were marked on a data record sheet at the time of testing and analyzed at a later time together with other variables of interest. At the conclusion of testing, each child received a monetary award. 2.4. General analyses Raw scores for each stimulus condition were examined and discarded if the trial score was equal to or greater than 2.5 S.D. from the participant's mean score for the stimulus condition. The average number of trials discarded for the FASD and NC groups was 1.7 and 2.1, respectively. Estimates of internal consistency reliability for trials within each stimulus condition were separately calculated for each group. Frequencies of response errors within each group were calculated and compared using Mann–Whitney and Wilcoxon's nonparametric tests. Reaction time data were analyzed to determine if gender and age should be included in the main analysis. Results indicated no significant main or interaction effects for either variable (p N 0.05) and they were not included as between-subject variables. Analysis of variance (ANOVA) procedures were used to analyze the dependent variable of RT produced by correctly responding to stimulus light activation. Group (FASD versus NC) was the between-subject factor and stimulus condition (0-,
1-, 2- and 3-bit stimulus conditions) was the within-subject factor. Data were analyzed using a 2 (group) × 4 (stimulus condition) mixed design [28] with repeated measures on the last factor. Assumptions of sphericity and homogeneity of variance were examined using Mauchly's and Levene's tests, respectively. Alpha was set to 0.05 for all tests of significance and post hoc comparisons were performed using Bonferroni t-test procedures. 3. Results 3.1. Demographic information Descriptive data for participants are presented in Table 1. Pairwise comparisons of group demographic data revealed no significant differences between the age (p N 0.87) or socioeconomic status of FASD and NC participants (p N 0.27). Analysis of IQ, as measured by the Wechsler Intelligence Scale for ChildrenIII or the Wechsler Preschool and Primary Scales of IntelligenceRevised, indicated significantly lower full-scale IQ (FSIQ) and verbal IQ (VIQ) scores for the FASD group (p b 0.002). In contrast, performance IQ (PIQ) was not statistically different for the two groups (p N 0.05). An examination of correlations among the FSIQ, VIQ and PIQ scores, and the dependent variable of RT produced results that are presented in Table 2. FSIQ, VIQ and PIQ were not significantly correlated with RT at any level of stimulus condition for the FASD group. For the NC group, FSIQ significantly correlated with RT at the 1 (r = −0.50), 2 (r = −0.50) and 3 (r = −0.48) bit stimulus condition levels. A similar relationship between PIQ and RT was found for the NC group at the 1 (r = −0.51), 2 (r = −0.52) and 3 (r = −0.66) bit stimulus condition levels. Thus, for the NC group, higher FSIQ and PIQ was associated with slower RT. 3.2. Error rates Error rates for both groups increased as the number of stimulus alternatives increased. Wilcoxon signed-rank tests indicated that all comparisons of the average number of errors between adjacent stimulus conditions within the FASD and NC groups were significantly different (p b 0.05). The mean number of errors recorded for the 0-, 1-, 2- and 3-bit stimulus conditions for the FASD group were 0.4, 1.0, 2.4 and 4.8 errors, respectively. Comparable data for the NC group were 0.1, 0.8, 2.9 and 650 Mean Reaction Time (ms)
Table 2 Correlations among FSIQ, VIQ, PIQ and RT mean scores
281
600 550 500 FASD 450
NC
400 350 300 0
1
2
3
Stimulus Conditions (bits)
Fig. 1. Mean reaction time (± S.E.M.) for the FASD and NC groups as a function of stimulus conditions.
282
R.W. Simmons et al. / Neurotoxicology and Teratology 28 (2006) 278–285
Mean Reaction Time (ms)
700 650 600
FASD
550 500
NC FASDY
450
FASDO
400
NCY NCO
350 300 0
1
2
3
Stimulus Conditions (bits)
Fig. 2. Mean reaction time data for young (5–7 years) and old (8–10 years) FASD and NC children from Simmons et al. [49] superimposed on data presented in Fig. 1.
3.8 errors. Mann–Whitney tests conducted to examine group differences in response errors revealed no significant differences for any of the stimulus conditions (1-bit, p N 0.55; 2-bit, p N 0.63; 3-bit, p N 0.84). A low frequency of errors for the 0-bit stimulus condition precluded comparison of average error rates at this level. 3.3. Trial reliability Internal consistency reliability for the response selection trials was separately calculated for each stimulus condition for both groups. Intraclass correlation coefficients (ICC) above 0.80 indicated reliable consistency of responses across trials [39]. For the FASD group, ICCs were in the acceptable range in all four-stimulus conditions (range of R = 0.88 to R = 0.97). Similar ICCs were established for the NC group (range of R = 0.90 to R = 0.97). These data indicate participants of both groups maintained consistent levels of RT across trials and stimulus conditions without evidence of learning effects. 3.4. Reaction time Average RT values for the FASD and NC groups are presented in Fig. 1. Mauchly's test indicated the data met the assumption of sphericity (p N 0.05) and Levene's test indicated group variances were equal (p N 0.05) at each of the fourstimulus conditions. The analysis of RT revealed a significant main effect for stimulus condition, F(3,105) = 226.7, p b 0.001, η2 = 0.87, with RT being directly related to increased information load. Bonferroni post hoc t-tests indicated the fastest RT occurred for the 0-bit stimulus condition (mean SRT = 359 ms) with significant (p b 0.05) increases in RT for the 1-, 2- and 3-bit stimulus conditions (mean CRTs = 417, 500 and 583 ms, respectively). Furthermore, with the number of stimulus choices logarithmically scaled, the slopes of the RT functions for the two groups were linear as predicted by Hick's law. The functions for the FASD and NC groups were y = 338.8 + 30.7x and y = 359.4 + 30.9x, respectively. There was no significant effect for group, F(1,35) = 1.45, p N 0.23, or group × stimulus condition interaction, F(3,105) = 0.79, p N 0.49. To verify this latter result, confidence intervals (CI) were constructed and compared for the FASD and NC groups for each stimulus condition [48]. None of the CI comparisons
(CIΔ − 57.46 to 14.63) indicated significant separation between the two groups, thereby confirming the finding that the FASD and NC groups were comparable in motor response selection processing. In an earlier study, Simmons et al. [49] also reported comparable RTs for the two groups but only for the 0-bit stimulus condition. When information load increased to 1-bit, the FASD group responded with significantly slower CRT. This result contrasts with the comparable CRT produced by the two groups for the 1-bit stimulus condition in the present study. Fig. 2 presents the results of the two studies with data superimposed. Data from the first study [49] were organized according to age divisions (5–7 and 8–10 years) originally used by investigators for statistical analyses for the FASD and NC groups. This procedure generated four subgroups: FASDyoung (FASDY), FASD-old (FASDO), NC-young (NCY) and NC-old (NCO). The 12–17-year-old age group used in the current study was also examined according to age subgroupings of 12–14 and 15–17 years, but was considered as a single age group when no significant age effects were indicated. The data in Fig. 2 indicate an age effect combined with a developmental delay for the FASD children. For example, the FASDO and NCO groups of the Simmons et al. [49] study have SRT values similar to those produced by the FASD and NC groups of the present experiment. For the 1-bit stimulus condition, however, the FASDO group experiences a significant (p b 0.009) timing deficit in comparison to the NCO of the previous study and FASD and NC groups of the present study. This CRT deficit is eliminated by the time FASD children reach 12–17 years of age, as would be predicted if alcohol-exposed children were experiencing a delay in development. Similarly, in comparison to the older children of the earlier study, the FASDY and NCY groups were significantly slower at the 0- and 1-bit stimulus conditions (p b 0.001), which is consistent with an age effect. But the FASDY children also show evidence of a developmental delay as indicated by the elevated SRT and CRT values. 4. Discussion Selective structures of the CNS such as the motor cortex [10,63], the rostral cingular motor area [24,25], striatum [25,38], cerebellum [18], globus pallidus [2] and corpus callosum [33] are involved in the generation of RT. Many of these CNS structures are adversely affected by prenatal exposure to alcohol and it is not surprising that deficits in motor timing have been reported for this clinical group [49,60]. What remains unclear is whether the observed temporal delays associated with a task involving an information load of 1-bit extends to higher cognitive loads, as is the case for other groups with CNS dysfunction [30]. The purpose of the present study was to investigate this question using a RT paradigm in which FASD adolescents had to respond to stimuli representing 0-, 1-, 2- and 3-bits of information. Consistent with the predictions of Hick's law and our first prediction, results indicated a significant and linear increase in RT for both the FASD and NC children for each additional bit of
R.W. Simmons et al. / Neurotoxicology and Teratology 28 (2006) 278–285
information to be processed. However, contrary to our second prediction, the time required by the FASD group to complete motor response selection was not statistically different from the processing time of the NC group. Interestingly, the slopes of the RT functions produced by the FASD and NC groups were estimated at 30.7 and 30.0ms/ bit of information, respectively. These values are considerably faster than the generally accepted processing rate of 150 ms/bit associated with Hick's law. Although the reason for this difference is unclear, it has been suggested that extensive practice [8,19,54,59] and high stimulus-response compatibility conditions [14] can significantly decrease the speed of information processing. As indicated by the high consistency reliabilities for all stimulus conditions, learning effects across trials was not a factor for either the FASD or NC groups even though the higher stimulus choice conditions represented a novel response and S-R compatibility was held constant across groups. Fig. 2 presents data from two separate experiments both of which use RT to assess response selection at information loads of 0- and 1-bit of information. As previously noted, FASD children had significantly slower responses at the 1-bit stimulus level in the first study but not the second study. The data illustrate a possible explanation for the difference in results that centers on a developmental delay effect that differentially impacts the two groups. It is well established that RT decreases through childhood, peaks in the second decade of life and progressively increases throughout the remainder of the lifespan [22]. This age related shift is evident in Fig. 2, as is the developmental delay associated with the FASD children. This result has particular relevance to the literature documenting various motor deficits in alcohol-exposed children. Much of this work has studied children less than 10 years of age [1,5] responding to a variety of motor skills [11,15]. Relatively little work has been completed using adolescents or adults with FASD and, while it remains possible that the proposed agedevelopmental delay phenomenon is specific to response selection processing, investigators must be cautious in assuming that motor deficits documented in young children persist into adolescence and adulthood. However, it remains possible the shift in RT with age is not a true developmental delay effect but is due to the CRT conditions not being sufficiently challenging to the older FASD children to produce discernable differences from the NC group performance. The data indicated a significant relation between IQ and RT for the control group but not the ALC group. The reason for this result is not immediately known but does not appear to be the product of a restrictive range in IQ values for either the NC or ALC groups. The result does indicate the RT paradigm used in the study is sensitive to detecting increases in RT for the alcohol-exposed children as a function of number of stimulus choices without the potential confounding effect of IQ. In the present study, the use of a relatively small sample size may have increased the probability of a type 2 error. Based on our previous work in motor timing behavior using FASD children [49,60], a large effect size was expected in the present study. In the absence of this effect size and given the small
283
difference in RT between the two groups at all stimulus conditions, it is unlikely a larger sample size would sufficiently increase statistical power to the extent that significant differences between groups would be established. Furthermore, the finding of null effects through use of parametric procedures was verified by application of confidence interval analysis. Both procedures supported the conclusion of comparable group performance and based on the outcome of these tests it is reasonable to assume the observed similarity of the RT data was a reliable effect. In summary, the results indicate both adolescent FASD and NC children produce a linear increase in RT as the amount of information processing increases, and RT values for each level of information processing examined are statistically comparable for the two groups. When considered with results from an earlier study using a similar RT paradigm, the data indicate a developmental delay phenomenon that adversely affects young children with FASD but which is no longer evident by adolescence. Acknowledgement This manuscript was supported in part by grants AA14017 and AA07456 awarded by the National Institute on Alcohol Abuse and Alcoholism. References [1] C.M. Adnams, P.W. Kodituwakku, A. Hay, C.D. Molteno, D. Viljoen, P.A. May, Patterns of cognitive-motor development in children with fetal alcohol syndrome from a community in South Africa, Alcohol Clin. Exp. Res. 25 (2001) 557–562. [2] M. Alamy, E. Trouche, A. Nieoullon, E. Legallet, Globus pallidus and motor initiation: the bilateral effects of unilateral quisqualic acid-induced lesion on reaction times in monkeys, Exp. Brain Res. 99 (1994) 247–258. [3] S.L. Archibald, C. Fennema-Notestine, A. Gamst, E.P. Riley, S.N. Mattson, T.L. Jernigan, Brain dysmorphology in individuals with severe prenatal alcohol exposure, Dev. Med. Child Neurol. 43 (2001) 148–154. [4] J.M. Azorin, P. Benhaim, T. Hasbroucq, C.A. Possamai, Stimulus preprocessing and response selection in depression: a reaction time study, Acta Psychol. (Amst) 89 (1995) 95–100. [5] H.M. Barr, A.P. Streissguth, B.L. Darby, P.D. Sampson, Prenatal exposure to alcohol, caffeine, tobacco and aspirin: effects on fine and gross motor performance in 4 year old children, Dev. Psychol. 26 (1990) 339–348. [6] N. Brewer, T. Nettelbeck, Speed and accuracy in the choice reaction time of mildly retarded adults, Am. J. Ment. Defic. 84 (1979) 55–61. [7] D.E. Broadbent, M. Gregory, On the interaction of S-R compatibility with other variables affecting reaction time, Br. J. Psychol. 56 (1965) 61–67. [8] D. Bunce, S.W. MacDonald, D.F. Hultsch, Inconsistency in serial choice decision and motor reaction times dissociate in younger and older adults, Brain Cogn. 56 (2004) 320–327. [9] M.J. Burden, S.W. Jacobsen, J.L. Jacobsen, Relation of prenatal alcohol exposure to cognitive processing speed and efficiency in childhood, Alcohol Clin. Exp. Res. 29 (2005) 1473–1483. [10] L. Carbonnell, T. Hasbroucq, J. Grapperon, F. Vidal, Response selection and motor areas: a behavioral and electrophysiological study, Clin. Neurophysiol. 115 (2004) 2164–2174. [11] L.S. Chandler, G.A. Richardson, J.D. Gallagher, N.L. Day, Prenatal exposure to alcohol and marijuana: effects on motor development of preschool children, Alcohol Clin. Exp. Res. 20 (1996) 455–461.
284
R.W. Simmons et al. / Neurotoxicology and Teratology 28 (2006) 278–285
[12] C.D. Coles, K.A. Platzman, C.L. Raskind-Hood, R.T. Brown, A. Falek, I.E. Smith, A comparison of children affected by prenatal alcohol exposure and attention deficit, hyperactivity disorder, Alcohol Clin. Exp. Res. 21 (1997) 150–161. [13] C.L. Coulter, R.W. Leech, G.B. Schaefer, B.W. Scheithauer, R.A. Brumback, Midline cerebral dysgenesis, dysfunction of the hypothalamic–pituitary axis, and fetal alcohol effects, Arch. Neurol. 50 (1993) 771–775. [14] P. Dassonville, S.M. Lewis, H.E. Foster, J. Ashe, Choice and stimulusresponse compatibility affect duration of response selection, Brain Res. Cogn. Brain Res. 7 (1999) 235–240. [15] P.A. Fried, B. Watkinson, 36- and 48-month neurobehavioral follow-up of children prenatally exposed to marijuana, cigarettes, and alcohol, J. Dev. Behav. Pediatr. 11 (1990) 49–58. [16] J. Gauntlett-Gilbert, V.J. Brown, Reaction time deficits and Parkinson's disease, Neurosci. Biobehav. Rev. 22 (1998) 865–881. [17] J.N. Giedd, F.X. Castellanos, B.J. Casey, P. Kozuch, A.C. King, S.D. Hamburger, J.L. Rapoport, Quantitative morphology of the corpus callosum in attention deficit hyperactivity disorder, Am. J. Psychiatry 151 (1994) 665–669. [18] S.E. Grill, M. Hallett, L.M. McShane, Timing of onset of afferent responses and of use of kinesthetic information for control of movement in normal and cerebellar impaired subjects, Exp. Brain Res. 113 (1997) 33–47. [19] M.A. Hart, T.G. Reeve, Influence of practice on response-selection and response-implementation processes involved in the response-interference effect, Acta Psychol. (Amst) 109 (2002) 177–194. [20] W.E. Hick, On the rate of gain of information, Q. J. Exp. Psychol. 4 (1952) 11–26. [21] S. Hocherman, R. Moont, M. Schwartz, Response selection and execution in patients with Parkinson's disease, Brain Res. Cogn. Brain Res. 19 (2004) 40–51. [22] D.F. Hultsch, S.W. MacDonald, R.A. Dixon, Variability in reaction time performance of younger and older adults, J. Geront., B Psychol. Sci. Soc. Sci. 57 (2002) P101–P115. [23] R. Hyman, Stimulus information as a determinant of reaction time, J. Exp. Psychol. 45 (1953) 188–196. [24] Y. Isomura, Y. Ito, T. Akazawa, A. Nambu, M. Takada, Neural coding of “attention for action” and “response selection” in primate anterior cingulate cortex, J. Neurosci. 23 (2003) 8002–8012. [25] M. Jahanshahi, R.G. Brown, C.D. Marsden, Simple and choice reaction time and the use of advance information for motor preparation in Parkinson's disease, Brain 115 (Pt 2) (1992) 539–564. [26] K.L. Jones, D.W. Smith, Recognition of the fetal alcohol syndrome in early infancy, Lancet 2 (1973) 999–1001. [27] R. Kail, Speed of information processing in patients with multiple sclerosis, J. Clin. Exp. Neuropsychol. 20 (1998) 98–106. [28] R. Kirk, Experimental Design: Procedures for the Behavioral Sciences, Brooks/Cole, Belmont, CA, 1995. [29] S. Krieger, S. Lis, B. Gallhofer, Cognitive subprocesses and schizophrenia: A. Reaction-time decomposition, Acta Psychiatr. Scand. Suppl. (2001) 18–27. [30] Y. Kutukcu, W.J. Marks Jr., D.S. Goodin, M.J. Aminoff, Simple and choice reaction time in Parkinson's disease, Brain Res. 815 (1999) 367–372. [31] L.E. Longstreth, N. el-Zahhar, M.B. Alcorn, Exceptions to Hick's law: explorations with a response duration measure, J. Exp. Psychol. Gen. 114 (1985) 417–434. [32] R.K. Mahurin, F.J. Pirozzolo, Application of Hick's law of response speed in Alzheimer and Parkinson diseases, Percept. Mot. Skills 77 (1993) 107–113. [33] J.L. Mathias, E.D. Bigler, N.R. Jones, S.C. Bowden, M. BarrettWoodbridge, G.C. Brown, D.J. Taylor, Neuropsychological and information processing performance and its relationship to white matter changes following moderate and severe traumatic brain injury: a preliminary study, Appl. Neuropsychol. 11 (2004) 134–152. [34] S.N. Mattson, E.P. Riley, Brain anomalies in fetal alcohol syndrome, in: L. E. Abel (Ed.), Fetal Alcohol Syndrome: From Mechanism to Prevention, CRC Press, Boca Raton, FL, 1996, pp. 51–68.
[35] S.N. Mattson, E.P. Riley, E.R. Sowell, T.L. Jernigan, D.F. Sobel, K.L. Jones, A decrease in the size of the basal ganglia in children with fetal alcohol syndrome, Alcohol Clin. Exp. Res. 20 (1996) 1088–1093. [36] S.N. Mattson, E.P. Riley, L. Gramling, D.C. Delis, K.L. Jones, Heavy prenatal alcohol exposure with or without physical features of fetal alcohol syndrome leads to IQ deficits, J. Pediatr. 131 (1997) 718–721. [37] J. Merkel, Die zeitlichen Verhaltnisse det Willensthautigkeit, in: R.S. Woodworth (Ed.), Experimental Psychology, Holt, New York, 1938. [38] T. Muller, E. Eising, W. Kuhn, T. Buttner, H.H. Coenen, H. Przuntek, Delayed motor response correlates with striatal degeneration in Parkinson's disease, Acta Neurol. Scand. 100 (1999) 227–230. [39] J.C. Nunnally, I.H. Bernstein, Psychometric Theory, McGraw-Hill, New York, 1994. [40] K.D. O'Malley, J. Nanson, Clinical implications of a link between fetal alcohol spectrum disorder and attention-deficit hyperactivity disorder, Can. J. Psychiatry 47 (2002) 349–354. [41] A.J. Raynor, Fractioned reflex and reaction time in children with developmental coordination disorder, Motor Control 2 (1998) 114–124. [42] E.P. Riley, S.N. Mattson, E.R. Sowell, T.L. Jernigan, D.F. Sobel, K.L. Jones, Abnormalities of the corpus callosum in children prenatally exposed to alcohol, Alcohol Clin. Exp. Res. 19 (1995) 1198–1202. [43] T.M. Roebuck-Spencer, S.N. Mattson, S.D. Marion, W.S. Brown, E.P. Riley, Bimanual coordination in alcohol-exposed children: role of the corpus callosum, J. Int. Neuropsychol. Soc. 10 (2004) 536–548. [44] M.A. Rogers, M.A. Bellgrove, E. Chiu, C. Mileshkin, J.L. Bradshaw, Response selection deficits in melancholic but not nonmelancholic unipolar major depression, J. Clin. Exp. Neuropsychol. 26 (2004) 169–179. [45] P.D. Sampson, A.P. Streissguth, F.L. Bookstein, R.E. Little, S.K. Clarren, P. Dehaene, J.W. Hanson, J.M. Graham Jr., Incidence of fetal alcohol syndrome and prevalence of alcohol-related neurodevelopmental disorder, Teratology 56 (1997) 317–326. [46] H.H. Samson, Microcephaly and Fetal Alcohol Syndrome: Human and Animal Studies, in: J.R. West (Ed.), Alcohol and Brain Development, Oxford University Press, New York, 1988, pp. 167–183. [47] R.A. Schmidt, T.D. Lee, Motor Control and Learning: A behavioral emphasis, Human Kinetics, Champaign ILL, 1999. [48] M.A. Serlin, R.C. Seaman, Equivalence confidence intervals for twogroup comparisons of means, Psychol. Methods 3 (1996) 403–411. [49] R.W. Simmons, T. Wass, J.D. Thomas, E.P. Riley, Fractionated simple and choice reaction time in children with prenatal exposure to alcohol, Alcohol Clin. Exp. Res. 26 (2002) 1412–1419. [50] I.M. Smith, C.D. Coles, J. Lancaster, P.M. Fernhoff, The effect of volume and duration of prenatal ethanol exposure on neonatal physical and behavioral development, Neurobehav. Toxicol. Teratol. 8 (1986) 375–381. [51] I.M. Spigel, Lift reaction time and topographic compatibility of the S-R field, J. Gen. Psychol. 72 (1965) 165–172. [52] A.P. Streissguth, K. O'Malley, Neuropsychiatric implications and longterm consequences of fetal alcohol spectrum disorders, Semin. Clin. Neuropsychiatry 5 (2000) 177–190. [53] A.P. Streissguth, H.M. Barr, D.C. Martin, C.S. Herman, Effects of maternal alcohol, nicotine, and caffeine use during pregnancy on infant mental and motor development at eight months, Alcohol Clin. Exp. Res. 4 (1980) 152–164. [54] A.P. Streissguth, D.C. Martin, H.M. Barr, B.M. Sandman, G.L. Kirchner, B.L. Darby, Intrauterine alcohol and nicotine exposure: attention and reaction time in 4-year-old children, Dev. Psychol. 20 (1984) 533–541. [55] A.P. Streissguth, H.M. Barr, P.D. Sampson, J.C. Parrish-Johnson, G.L. Kirchner, D.C. Martin, Attention, distraction and reaction time at age 7 years and prenatal alcohol exposure, Neurobehav. Toxicol. Teratol. 8 (1986) 717–725. [56] A.N. Streissguth, F.L. Bookstein, P.D. Sampson, H.M. Barr, Attention: prenatal alcohol and continuities of vigilance and attentional problems from 4 through 14 years, Dev. Psychopathol. 7 (1995) 419–446. [57] W.H. Teichner, M.J. Krebs, Laws of visual choice reaction time, Psychol. Rev. 81 (1974) 75–98. [58] C. Vickrey, A. Neuringer, Pigeon reaction time, Hick's law, and intelligence, Psychon. Bull. Rev. 7 (2000) 284–291.
R.W. Simmons et al. / Neurotoxicology and Teratology 28 (2006) 278–285 [59] M.G. Wade, K.M. Newell, S.A. Wallace, Decision time and movement time as a function of response complexity in retarded persons, Am. J. Ment. Defic. 83 (1978) 135–144. [60] T.S. Wass, R.W. Simmons, J.D. Thomas, E.P. Riley, Timing accuracy and variability in children with prenatal exposure to alcohol, Alcohol Clin. Exp. Res. 26 (2002) 1887–1896. [61] K.F. Widaman, J.S. Carlson, Procedural effects on performance on the Hicks paradigm: bias in reaction time and movement time parameters, Intell 13 (1989) 63–85.
285
[62] K. Wisniewski, M. Dambska, J.H. Sher, Q. Qazi, A clinical neuropathological study of the fetal alcohol syndrome, Neuropediatrics 14 (1983) 197–201. [63] J. Yordanova, V. Kolev, J. Hohnsbein, M. Falkenstein, Sensorimotor slowing with ageing is mediated by a functional dysregulation of motorgeneration processes: evidence from high-resolution event-related potentials, Brain 127 (2004) 351–362.
Motor response selection in children with fetal alcohol spectrum disorders Roger W. Simmons a,⁎, Jennifer D. Thomas b , Susan S. Levy a , Edward P. Riley b a
Motor Control Laboratory, Department of Exercise and Nutritional Sciences, San Diego State University, San Diego, CA 92182, USA b Center for Behavioral Teratology, Department of Psychology, San Diego State University, San Diego, CA 92120, USA Received 26 July 2005; received in revised form 6 December 2005; accepted 10 January 2006 Available online 9 March 2006
Abstract Previous work has reported timing delays in motor response selection in children with prenatal exposure to alcohol when the information load involved responding to two stimulus choices. The present study examined whether the delay in response selection extends to conditions in which the information load is increased to four and eight stimulus choices. Twenty children aged between 12 and 17 years with fetal alcohol spectrum disorders (FASD) were compared to 17 non-alcohol-exposed controls (NC) on a reaction time (RT) task involving 1, 2, 4 or 8 visual stimulus choices. The task demands required the participant to release a response key as fast as possible when the stimulus light electronically paired with the response key was activated. With the number of stimulus choices expressed on a logarithmic scale, there was a significant and linear increase in RT for the FASD children as predicted by information processing theory. Additionally, the increase in RT for the FASD group was comparable to that observed for the NC children at each level of stimulus choice examined. It was concluded that FASD adolescents require additional time to process increasing amount of information, but that the time required for motor response selection is not delayed relative to control group performance. © 2006 Elsevier Inc. All rights reserved. Keywords: Response selection; Fetal alcohol spectrum disorders; Reaction time
1. Introduction Fetal alcohol syndrome (FAS) is defined by a triad of characteristics of pre- and postnatal growth retardation, central nervous system (CNS) dysfunction and a specific pattern of facial features [26]. FAS children experience a range of cognitive and behavioral dysfunctions, as do children with histories of prenatal exposure to alcohol (PEA) but who do not have the physical anomalies associated with FAS [34,36]. In addition to cognitive and intellectual problems [50], both FAS and PEA children prenatally exposed to alcohol (PEA) also manifest poor motor skills, including atypical motor timing [49,58]. According to the constructs of information processing theory, one determinant of motor timing is the time required to make a decision to respond to an external stimulus, which is known as motor response selection [47].
⁎ Corresponding author. Tel.: +1 619 594 5543; fax: +1 619 594 6553. E-mail address: [email protected] (R.W. Simmons). 0892-0362/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.ntt.2006.01.008
The history of investigating response selection processing can be traced back to 1885 when Merkel [37] found a negatively accelerated function between reaction time (RT) and the number of stimulus choices. Many years later, it was reported that the RT function is linear when the number of stimulus choices is expressed on a log2 scale (in units of bits) [20], with the slope of the function reflecting delays in the speed of CNS processing as information load increases. The predicted relationship between RT and the number of stimulus choices is known as Hick's law [20] and, although some exceptions have been reported [31,61], the law has proven to be robust and generalizes to a variety of conditions and tasks using non-clinical human participants [7,14,23,25,51,57] and certain animal species [58]. Hick's law has also been tested using clinical groups including Alzheimer and Parkinson's patients [16,21,32], melancholic depressives [44] and non-melancholic depressives [4], schizophrenics [29], mentally retarded children [59] and adults [6], individuals with multiple sclerosis [27], and children prenatally exposed to alcohol [49]. In the context of prenatal alcohol exposure, anomalies in the cerebrum [46], cerebellum [3,35,45,46], corpus callosum [26,42,62] and basal ganglia [13,34,42,62] resulting from the
R.W. Simmons et al. / Neurotoxicology and Teratology 28 (2006) 278–285
teratogenic effects of alcohol are thought to slow decision making speed, as estimated by RT. For example, one group of RT studies required alcohol-exposed children to view continuous visual displays and respond as fast as possible when a target stimulus randomly appeared within the context of the display. Results of this work have provided limited support for the existence of RT delays in alcohol-exposed children [12,53–56]. Continuous display studies, however, involve the child attending to a visual stimulus for an extended period of time (e.g. 12–15 min), which is conceptually different from traditional measures of simple reaction time (SRT), involving one stimulus-response pairing, and choice reaction time (CRT), involving multiple stimulus-response pairings, in which the stimulus and response occur in a relatively brief time period (e.g. 1–4 s). To date, only one study has used this type of RT paradigm to investigate information processing time in children with prenatal alcohol exposure. Results revealed comparable SRT values for the alcohol-exposed children and controls, but the alcohol-exposed group was significantly delayed during a two-stimulus CRT task [49]. Additionally, the investigators fractionated SRT and CRT into premotor RT and motor RT. Premotor RT defines the time between the onset of the stimulus and the onset of neuromuscular activity and provides a measure of the time taken to complete CNS processing [41]. In contrast, motor RT, defined as the time between the onset of neuromuscular activity as indexed by electromyographic (EMG) activity and the first observable movement (the release of the RT key), indexes electromechanical delays inherent to the peripheral neuromuscular system. The advantage of fractionating SRT and CRT lies in generating an estimate of central processing time minus delays associated with the initiation of neuromuscular activity. Results revealed comparable premotor SRT for both groups, but the alcohol-exposed group produced a significantly longer premotor CRT. This latter finding confirmed the existence of significant CNS processing deficits in the alcohol-exposed child when information load (i.e. the number of stimulus choices) was doubled. A similar delay in reaction time has been reported for alcohol-exposed children when responding to higher cognitive loads of a Sternberg task [9]. The alcohol-exposed group was also significantly delayed in motor RT for both the SRT and CRT conditions, thereby indicating additional RT delays at the peripheral neuromuscular level. What remains unknown is whether the response selection delay reported for the alcohol-exposed children during the two CRT task extends to conditions with additional levels of information processing, and whether the resulting RT function is linear as proposed by motor response selection theory. Based on our previous work with two stimulus choices, and the expectation that additional information processing will also involve alcohol impacted CNS structures, it is predicted that adolescent children prenatally exposed to alcohol will manifest a linear increase in RT as the number of stimuli double (i.e. increases of 1-bit of information), and these children will be significantly slower in responding during multiple stimulus conditions than non-alcohol-exposed controls.
279
2. Methods 2.1. Participants Two groups of children between the ages of 12 and 17 years, with full-scale IQs (FSIQ) above 70 served as participants in the study. This age group was chosen for their ability to sustain attention to the multiple repetitions of the stimulus conditions and because they possess levels of fine motor control sufficient to independently move the four fingers of each hand during multiple stimulus choice conditions. One group of children was recruited from a pool of participants registered at the Center of Behavioral Teratology (CBT) at San Diego State University. These children had previously been evaluated with a comprehensive neuropsychological battery (including intelligence quotient testing—IQ) and by a pediatric dysmorphologist (Kenneth Lyons Jones, MD). The average time between IQ assessment and participation in the experiment was 2.2 years (S.D. ± 2.0). Using established criteria of specific craniofacial anomalies (e.g. short palpebral fissures, long smooth philtrum, broad nasal bridge and epicanthic folds), pre- and/or postnatal growth deficiency and CNS dysfunction [25], six children received a diagnosis of fetal alcohol syndrome (FAS) by the dysmorphologist. An additional 14 children who did not present any of the triad of characteristics associated with FAS but who had experienced in utero exposure to alcohol [36] were classified by the dysmorphologist as having PEA. Nine of these children presented evidence of having one of the facial characteristics associated with FAS, while another child experienced growth deficiency. Although specific details regarding the degree and frequency of prenatal alcohol exposure were not always available, abusive use of alcohol by each child's mother was confirmed through maternal self-reports and access to medical or social service records. Statistical analyses revealed the FAS and PEA children to have comparable RTs for all levels of information load. This result is consistent with previous work reporting similarity of motor performance [43] for the two groups. Consequently, the FAS and PEA children were combined into a single alcohol-exposed group, which falls under the umbrella term of fetal alcohol spectrum disorders (FASD) [40,52]. This combined alcohol-exposed group will be indicated in the present study by the abbreviation FASD. A second group of 17 children aged 12 and 17 years inclusively, comprised the non-alcohol-exposed control group (NC) and were also recruited from a pool of participants registered at the CBT. All children in the NC group had previously been tested using the same evaluation scales used for the alcohol-exposed children. Evaluation involved a comprehensive neuropsychological battery and screening for prenatal alcohol and other teratogenic agents through a telephone interview and completion of a parent questionnaire that included information concerning alcohol consumption during pregnancy. An average of 2.7 years (S.D. ± 2.3) occurred between IQ assessment and completion of the task for the NC group. Potential control participants were excluded for the following reasons: prenatal exposure to alcohol or other known teratogens, closed head injury with significant loss of memory, a
280
R.W. Simmons et al. / Neurotoxicology and Teratology 28 (2006) 278–285
FSIQ score below 70 or a diagnosis of attention deficit hyperactivity disorder (ADHD). ADHD children were excluded from the FASD and NC groups because of the purported link between ADHD and structural damage in the CNS [17]. None of the control participants had any known neurological problems or exposure to teratogens. Therefore, 20 children were included in the FASD group and 17 in the NC group. The NC group was closely matched to the FASD children on age, gender, socioeconomic status (measured using the Hollingshead Four Factor Index, 1975) and ethnicity. Demographic information for the FASD and NC groups is presented in Table 1. 2.2. Apparatus The RT apparatus consisted of two rectangular response modules (21 × 15 × 8cm) placed adjacent to each other on a table in front of the seated participant. Positioned on top of each response module were four square (1.5 × 1.5 cm) response buttons arranged 2.5 cm apart in an ergonomically compatible semi-circular pattern. The top surface of each response module was inclined towards the participant at an angle of 20° to the horizontal plane. With the base of the palm resting on a wrist pad, the participants could comfortably place the four fingers of each hand on the designated response buttons. Each response button was electronically paired with a color-coded circular stimulus light (1.5 cm in diameter) located 3 cm above the response button on the top surface of the response module. There were four stimulus light-response button pairings for each response module resulting in a maximum of eight stimulus-response alternatives. For testing purposes, response buttons were numbered 1 through 8 and corresponded to the eight fingers beginning with the left little finger of the participant's right hand through the little finger of the left hand, respectively. The thumb of each hand was not used for responding. The participant was unaware of this numbering system and the
experimenter used this key designation only to determine the order of stimulus light presentation. The experimenter sat opposite the participant and controlled activation of the stimulus lights from two remote control panels hidden from the participant's view by a wooden screen. 2.3. Procedures After obtaining informed consent from the legal guardian and assent from the child, two tests were completed by the participant, one of which was the response selection test. The second test was conducted as part of an ongoing research project at the CBT and did not interfere with performance of the response selection test. The presentation order of the two tests was systematically counterbalanced across subjects. An interval of approximately 7 min separated the two tests and eliminated any fatigue resulting from completion of the first test. Participants were seated in front of the response modules with the seat height adjusted to a position where the child could comfortably place the finger(s) on the appropriate response key (s). Prior to testing, participants were provided task related instructions and a demonstration, and allowed a maximum of five practice trials for each of the four-stimulus conditions examined. The experimenter provided a verbal ‘ready’ command followed by a randomly determined foreperiod of 1–4 s. The task consisted of the participants holding down 1, 2, 4 or 8 response keys with their fingers and releasing a response key as fast as possible when the stimulus light paired with the response key was activated. Release of the response key (as opposed to depressing the key) controlled for any time delay associated with finger movement. The elapsed time between stimulus light activation and lifting the finger from the response key defined RT and was measured to the nearest millisecond using two electronic timers. The 1, 2, 4 and 8 stimulus light choices represented four-stimulus conditions corresponding to
Table 1 Demographic information by group (mean ± S.D.)
Gender (M, %/F, %) Age, M (S.D.) Range FSIQ a, M (S.D.) Range VIQ a, M (S.D.) Range PIQ a, M (S.D.) Range Hollingshead Score b Hand dominance (L, %/R, %) Ethnicity (n, %) Asian African American Caucasian Native American Hispanic descent
FASD (N = 20)
NC (N = 17)
9 (45.0%):11 (55.0%) 14.2 (1.4) 12.0–17.0 88.7 (10.0) 70–109 87.6 (11.8) 58–101 92.0 (13.3) 69–119 46.9 (7.3) 4 (20%):16 (80%)
6 (35.3%):11 (69.2%) 14.2 (1.7) 12.0–17.75 101.2 (9.9) 84–118 104.6 (9.9) 85–119 97.4 (12.7) 79–121 49.9 (13.0) 0 (0%):17 (100%)
2 (10.0%) 2 (10.0%) 10 (50.0%) 2 (10.0%) 4 (20.0%)
1 (5.9%) 4 (23.5%) 9 (52.9%) 0 (0%) 1 (17.6%)
p N 0.54 p N 0.87 p b 0.002 p b 0.002 p N 0.22 p N 0.27
a Intelligence scores were derived from either the Wechsler Preschool and Primary Scales of Intelligence Revised or the Wechsler Intelligence Scale for Children-III depending on the child's age at the study entry. b Socioeconnomic status was estimated using the Hollingshead Four Factor Index of Social Status (Hollingshead, 1975, unpublished data).
R.W. Simmons et al. / Neurotoxicology and Teratology 28 (2006) 278–285
FSIQ
VIQ
PIQ
FASD group 0 bit 1 bit 2 bits 3 bits
−0.352 −0.328 −0.400 −0.126
− 0.133 − 0.256 − 0.233 0.051
−0.423 −0.283 −0.422 −0.221
NC group 0 bit 1 bit 2 bits 3 bits
−0.345 −0.502 * −0.497 * −0.482 *
− 0.315 − 0.188 − 0.176 0.048
−0.194 −0.505 ⁎ −0.516 ⁎ −0.664 ⁎⁎
⁎ p b 0.05. ⁎⁎ p b 0.01.
information loads of 0, 1, 2 and 3 bits of information, respectively. Participants completed a block of 24 trials for each stimulus condition. In the CRT conditions (i.e. 1, 2 or 3 bits of information), the order of stimulus light activation was randomized within each trial block with each light being presented with equal frequency. The presentation order of the four-stimulus conditions was counterbalanced across participants with a 2-min rest period between trial blocks. When responding during the 0-, 1-, 2- and 3-bit stimulus conditions, the participants used either the right index finger (or left, if the individual was left-handed), the two index fingers, the index and middle fingers of each hand, or all four fingers of each hand, respectively. A mistrial was considered to have occurred if the participant lifted the wrong finger or lifted more than one finger in response to a stimulus light being activated. Mistrials were repeated at the end of the trial block. Mistrials were marked on a data record sheet at the time of testing and analyzed at a later time together with other variables of interest. At the conclusion of testing, each child received a monetary award. 2.4. General analyses Raw scores for each stimulus condition were examined and discarded if the trial score was equal to or greater than 2.5 S.D. from the participant's mean score for the stimulus condition. The average number of trials discarded for the FASD and NC groups was 1.7 and 2.1, respectively. Estimates of internal consistency reliability for trials within each stimulus condition were separately calculated for each group. Frequencies of response errors within each group were calculated and compared using Mann–Whitney and Wilcoxon's nonparametric tests. Reaction time data were analyzed to determine if gender and age should be included in the main analysis. Results indicated no significant main or interaction effects for either variable (p N 0.05) and they were not included as between-subject variables. Analysis of variance (ANOVA) procedures were used to analyze the dependent variable of RT produced by correctly responding to stimulus light activation. Group (FASD versus NC) was the between-subject factor and stimulus condition (0-,
1-, 2- and 3-bit stimulus conditions) was the within-subject factor. Data were analyzed using a 2 (group) × 4 (stimulus condition) mixed design [28] with repeated measures on the last factor. Assumptions of sphericity and homogeneity of variance were examined using Mauchly's and Levene's tests, respectively. Alpha was set to 0.05 for all tests of significance and post hoc comparisons were performed using Bonferroni t-test procedures. 3. Results 3.1. Demographic information Descriptive data for participants are presented in Table 1. Pairwise comparisons of group demographic data revealed no significant differences between the age (p N 0.87) or socioeconomic status of FASD and NC participants (p N 0.27). Analysis of IQ, as measured by the Wechsler Intelligence Scale for ChildrenIII or the Wechsler Preschool and Primary Scales of IntelligenceRevised, indicated significantly lower full-scale IQ (FSIQ) and verbal IQ (VIQ) scores for the FASD group (p b 0.002). In contrast, performance IQ (PIQ) was not statistically different for the two groups (p N 0.05). An examination of correlations among the FSIQ, VIQ and PIQ scores, and the dependent variable of RT produced results that are presented in Table 2. FSIQ, VIQ and PIQ were not significantly correlated with RT at any level of stimulus condition for the FASD group. For the NC group, FSIQ significantly correlated with RT at the 1 (r = −0.50), 2 (r = −0.50) and 3 (r = −0.48) bit stimulus condition levels. A similar relationship between PIQ and RT was found for the NC group at the 1 (r = −0.51), 2 (r = −0.52) and 3 (r = −0.66) bit stimulus condition levels. Thus, for the NC group, higher FSIQ and PIQ was associated with slower RT. 3.2. Error rates Error rates for both groups increased as the number of stimulus alternatives increased. Wilcoxon signed-rank tests indicated that all comparisons of the average number of errors between adjacent stimulus conditions within the FASD and NC groups were significantly different (p b 0.05). The mean number of errors recorded for the 0-, 1-, 2- and 3-bit stimulus conditions for the FASD group were 0.4, 1.0, 2.4 and 4.8 errors, respectively. Comparable data for the NC group were 0.1, 0.8, 2.9 and 650 Mean Reaction Time (ms)
Table 2 Correlations among FSIQ, VIQ, PIQ and RT mean scores
281
600 550 500 FASD 450
NC
400 350 300 0
1
2
3
Stimulus Conditions (bits)
Fig. 1. Mean reaction time (± S.E.M.) for the FASD and NC groups as a function of stimulus conditions.
282
R.W. Simmons et al. / Neurotoxicology and Teratology 28 (2006) 278–285
Mean Reaction Time (ms)
700 650 600
FASD
550 500
NC FASDY
450
FASDO
400
NCY NCO
350 300 0
1
2
3
Stimulus Conditions (bits)
Fig. 2. Mean reaction time data for young (5–7 years) and old (8–10 years) FASD and NC children from Simmons et al. [49] superimposed on data presented in Fig. 1.
3.8 errors. Mann–Whitney tests conducted to examine group differences in response errors revealed no significant differences for any of the stimulus conditions (1-bit, p N 0.55; 2-bit, p N 0.63; 3-bit, p N 0.84). A low frequency of errors for the 0-bit stimulus condition precluded comparison of average error rates at this level. 3.3. Trial reliability Internal consistency reliability for the response selection trials was separately calculated for each stimulus condition for both groups. Intraclass correlation coefficients (ICC) above 0.80 indicated reliable consistency of responses across trials [39]. For the FASD group, ICCs were in the acceptable range in all four-stimulus conditions (range of R = 0.88 to R = 0.97). Similar ICCs were established for the NC group (range of R = 0.90 to R = 0.97). These data indicate participants of both groups maintained consistent levels of RT across trials and stimulus conditions without evidence of learning effects. 3.4. Reaction time Average RT values for the FASD and NC groups are presented in Fig. 1. Mauchly's test indicated the data met the assumption of sphericity (p N 0.05) and Levene's test indicated group variances were equal (p N 0.05) at each of the fourstimulus conditions. The analysis of RT revealed a significant main effect for stimulus condition, F(3,105) = 226.7, p b 0.001, η2 = 0.87, with RT being directly related to increased information load. Bonferroni post hoc t-tests indicated the fastest RT occurred for the 0-bit stimulus condition (mean SRT = 359 ms) with significant (p b 0.05) increases in RT for the 1-, 2- and 3-bit stimulus conditions (mean CRTs = 417, 500 and 583 ms, respectively). Furthermore, with the number of stimulus choices logarithmically scaled, the slopes of the RT functions for the two groups were linear as predicted by Hick's law. The functions for the FASD and NC groups were y = 338.8 + 30.7x and y = 359.4 + 30.9x, respectively. There was no significant effect for group, F(1,35) = 1.45, p N 0.23, or group × stimulus condition interaction, F(3,105) = 0.79, p N 0.49. To verify this latter result, confidence intervals (CI) were constructed and compared for the FASD and NC groups for each stimulus condition [48]. None of the CI comparisons
(CIΔ − 57.46 to 14.63) indicated significant separation between the two groups, thereby confirming the finding that the FASD and NC groups were comparable in motor response selection processing. In an earlier study, Simmons et al. [49] also reported comparable RTs for the two groups but only for the 0-bit stimulus condition. When information load increased to 1-bit, the FASD group responded with significantly slower CRT. This result contrasts with the comparable CRT produced by the two groups for the 1-bit stimulus condition in the present study. Fig. 2 presents the results of the two studies with data superimposed. Data from the first study [49] were organized according to age divisions (5–7 and 8–10 years) originally used by investigators for statistical analyses for the FASD and NC groups. This procedure generated four subgroups: FASDyoung (FASDY), FASD-old (FASDO), NC-young (NCY) and NC-old (NCO). The 12–17-year-old age group used in the current study was also examined according to age subgroupings of 12–14 and 15–17 years, but was considered as a single age group when no significant age effects were indicated. The data in Fig. 2 indicate an age effect combined with a developmental delay for the FASD children. For example, the FASDO and NCO groups of the Simmons et al. [49] study have SRT values similar to those produced by the FASD and NC groups of the present experiment. For the 1-bit stimulus condition, however, the FASDO group experiences a significant (p b 0.009) timing deficit in comparison to the NCO of the previous study and FASD and NC groups of the present study. This CRT deficit is eliminated by the time FASD children reach 12–17 years of age, as would be predicted if alcohol-exposed children were experiencing a delay in development. Similarly, in comparison to the older children of the earlier study, the FASDY and NCY groups were significantly slower at the 0- and 1-bit stimulus conditions (p b 0.001), which is consistent with an age effect. But the FASDY children also show evidence of a developmental delay as indicated by the elevated SRT and CRT values. 4. Discussion Selective structures of the CNS such as the motor cortex [10,63], the rostral cingular motor area [24,25], striatum [25,38], cerebellum [18], globus pallidus [2] and corpus callosum [33] are involved in the generation of RT. Many of these CNS structures are adversely affected by prenatal exposure to alcohol and it is not surprising that deficits in motor timing have been reported for this clinical group [49,60]. What remains unclear is whether the observed temporal delays associated with a task involving an information load of 1-bit extends to higher cognitive loads, as is the case for other groups with CNS dysfunction [30]. The purpose of the present study was to investigate this question using a RT paradigm in which FASD adolescents had to respond to stimuli representing 0-, 1-, 2- and 3-bits of information. Consistent with the predictions of Hick's law and our first prediction, results indicated a significant and linear increase in RT for both the FASD and NC children for each additional bit of
R.W. Simmons et al. / Neurotoxicology and Teratology 28 (2006) 278–285
information to be processed. However, contrary to our second prediction, the time required by the FASD group to complete motor response selection was not statistically different from the processing time of the NC group. Interestingly, the slopes of the RT functions produced by the FASD and NC groups were estimated at 30.7 and 30.0ms/ bit of information, respectively. These values are considerably faster than the generally accepted processing rate of 150 ms/bit associated with Hick's law. Although the reason for this difference is unclear, it has been suggested that extensive practice [8,19,54,59] and high stimulus-response compatibility conditions [14] can significantly decrease the speed of information processing. As indicated by the high consistency reliabilities for all stimulus conditions, learning effects across trials was not a factor for either the FASD or NC groups even though the higher stimulus choice conditions represented a novel response and S-R compatibility was held constant across groups. Fig. 2 presents data from two separate experiments both of which use RT to assess response selection at information loads of 0- and 1-bit of information. As previously noted, FASD children had significantly slower responses at the 1-bit stimulus level in the first study but not the second study. The data illustrate a possible explanation for the difference in results that centers on a developmental delay effect that differentially impacts the two groups. It is well established that RT decreases through childhood, peaks in the second decade of life and progressively increases throughout the remainder of the lifespan [22]. This age related shift is evident in Fig. 2, as is the developmental delay associated with the FASD children. This result has particular relevance to the literature documenting various motor deficits in alcohol-exposed children. Much of this work has studied children less than 10 years of age [1,5] responding to a variety of motor skills [11,15]. Relatively little work has been completed using adolescents or adults with FASD and, while it remains possible that the proposed agedevelopmental delay phenomenon is specific to response selection processing, investigators must be cautious in assuming that motor deficits documented in young children persist into adolescence and adulthood. However, it remains possible the shift in RT with age is not a true developmental delay effect but is due to the CRT conditions not being sufficiently challenging to the older FASD children to produce discernable differences from the NC group performance. The data indicated a significant relation between IQ and RT for the control group but not the ALC group. The reason for this result is not immediately known but does not appear to be the product of a restrictive range in IQ values for either the NC or ALC groups. The result does indicate the RT paradigm used in the study is sensitive to detecting increases in RT for the alcohol-exposed children as a function of number of stimulus choices without the potential confounding effect of IQ. In the present study, the use of a relatively small sample size may have increased the probability of a type 2 error. Based on our previous work in motor timing behavior using FASD children [49,60], a large effect size was expected in the present study. In the absence of this effect size and given the small
283
difference in RT between the two groups at all stimulus conditions, it is unlikely a larger sample size would sufficiently increase statistical power to the extent that significant differences between groups would be established. Furthermore, the finding of null effects through use of parametric procedures was verified by application of confidence interval analysis. Both procedures supported the conclusion of comparable group performance and based on the outcome of these tests it is reasonable to assume the observed similarity of the RT data was a reliable effect. In summary, the results indicate both adolescent FASD and NC children produce a linear increase in RT as the amount of information processing increases, and RT values for each level of information processing examined are statistically comparable for the two groups. When considered with results from an earlier study using a similar RT paradigm, the data indicate a developmental delay phenomenon that adversely affects young children with FASD but which is no longer evident by adolescence. Acknowledgement This manuscript was supported in part by grants AA14017 and AA07456 awarded by the National Institute on Alcohol Abuse and Alcoholism. References [1] C.M. Adnams, P.W. Kodituwakku, A. Hay, C.D. Molteno, D. Viljoen, P.A. May, Patterns of cognitive-motor development in children with fetal alcohol syndrome from a community in South Africa, Alcohol Clin. Exp. Res. 25 (2001) 557–562. [2] M. Alamy, E. Trouche, A. Nieoullon, E. Legallet, Globus pallidus and motor initiation: the bilateral effects of unilateral quisqualic acid-induced lesion on reaction times in monkeys, Exp. Brain Res. 99 (1994) 247–258. [3] S.L. Archibald, C. Fennema-Notestine, A. Gamst, E.P. Riley, S.N. Mattson, T.L. Jernigan, Brain dysmorphology in individuals with severe prenatal alcohol exposure, Dev. Med. Child Neurol. 43 (2001) 148–154. [4] J.M. Azorin, P. Benhaim, T. Hasbroucq, C.A. Possamai, Stimulus preprocessing and response selection in depression: a reaction time study, Acta Psychol. (Amst) 89 (1995) 95–100. [5] H.M. Barr, A.P. Streissguth, B.L. Darby, P.D. Sampson, Prenatal exposure to alcohol, caffeine, tobacco and aspirin: effects on fine and gross motor performance in 4 year old children, Dev. Psychol. 26 (1990) 339–348. [6] N. Brewer, T. Nettelbeck, Speed and accuracy in the choice reaction time of mildly retarded adults, Am. J. Ment. Defic. 84 (1979) 55–61. [7] D.E. Broadbent, M. Gregory, On the interaction of S-R compatibility with other variables affecting reaction time, Br. J. Psychol. 56 (1965) 61–67. [8] D. Bunce, S.W. MacDonald, D.F. Hultsch, Inconsistency in serial choice decision and motor reaction times dissociate in younger and older adults, Brain Cogn. 56 (2004) 320–327. [9] M.J. Burden, S.W. Jacobsen, J.L. Jacobsen, Relation of prenatal alcohol exposure to cognitive processing speed and efficiency in childhood, Alcohol Clin. Exp. Res. 29 (2005) 1473–1483. [10] L. Carbonnell, T. Hasbroucq, J. Grapperon, F. Vidal, Response selection and motor areas: a behavioral and electrophysiological study, Clin. Neurophysiol. 115 (2004) 2164–2174. [11] L.S. Chandler, G.A. Richardson, J.D. Gallagher, N.L. Day, Prenatal exposure to alcohol and marijuana: effects on motor development of preschool children, Alcohol Clin. Exp. Res. 20 (1996) 455–461.
284
R.W. Simmons et al. / Neurotoxicology and Teratology 28 (2006) 278–285
[12] C.D. Coles, K.A. Platzman, C.L. Raskind-Hood, R.T. Brown, A. Falek, I.E. Smith, A comparison of children affected by prenatal alcohol exposure and attention deficit, hyperactivity disorder, Alcohol Clin. Exp. Res. 21 (1997) 150–161. [13] C.L. Coulter, R.W. Leech, G.B. Schaefer, B.W. Scheithauer, R.A. Brumback, Midline cerebral dysgenesis, dysfunction of the hypothalamic–pituitary axis, and fetal alcohol effects, Arch. Neurol. 50 (1993) 771–775. [14] P. Dassonville, S.M. Lewis, H.E. Foster, J. Ashe, Choice and stimulusresponse compatibility affect duration of response selection, Brain Res. Cogn. Brain Res. 7 (1999) 235–240. [15] P.A. Fried, B. Watkinson, 36- and 48-month neurobehavioral follow-up of children prenatally exposed to marijuana, cigarettes, and alcohol, J. Dev. Behav. Pediatr. 11 (1990) 49–58. [16] J. Gauntlett-Gilbert, V.J. Brown, Reaction time deficits and Parkinson's disease, Neurosci. Biobehav. Rev. 22 (1998) 865–881. [17] J.N. Giedd, F.X. Castellanos, B.J. Casey, P. Kozuch, A.C. King, S.D. Hamburger, J.L. Rapoport, Quantitative morphology of the corpus callosum in attention deficit hyperactivity disorder, Am. J. Psychiatry 151 (1994) 665–669. [18] S.E. Grill, M. Hallett, L.M. McShane, Timing of onset of afferent responses and of use of kinesthetic information for control of movement in normal and cerebellar impaired subjects, Exp. Brain Res. 113 (1997) 33–47. [19] M.A. Hart, T.G. Reeve, Influence of practice on response-selection and response-implementation processes involved in the response-interference effect, Acta Psychol. (Amst) 109 (2002) 177–194. [20] W.E. Hick, On the rate of gain of information, Q. J. Exp. Psychol. 4 (1952) 11–26. [21] S. Hocherman, R. Moont, M. Schwartz, Response selection and execution in patients with Parkinson's disease, Brain Res. Cogn. Brain Res. 19 (2004) 40–51. [22] D.F. Hultsch, S.W. MacDonald, R.A. Dixon, Variability in reaction time performance of younger and older adults, J. Geront., B Psychol. Sci. Soc. Sci. 57 (2002) P101–P115. [23] R. Hyman, Stimulus information as a determinant of reaction time, J. Exp. Psychol. 45 (1953) 188–196. [24] Y. Isomura, Y. Ito, T. Akazawa, A. Nambu, M. Takada, Neural coding of “attention for action” and “response selection” in primate anterior cingulate cortex, J. Neurosci. 23 (2003) 8002–8012. [25] M. Jahanshahi, R.G. Brown, C.D. Marsden, Simple and choice reaction time and the use of advance information for motor preparation in Parkinson's disease, Brain 115 (Pt 2) (1992) 539–564. [26] K.L. Jones, D.W. Smith, Recognition of the fetal alcohol syndrome in early infancy, Lancet 2 (1973) 999–1001. [27] R. Kail, Speed of information processing in patients with multiple sclerosis, J. Clin. Exp. Neuropsychol. 20 (1998) 98–106. [28] R. Kirk, Experimental Design: Procedures for the Behavioral Sciences, Brooks/Cole, Belmont, CA, 1995. [29] S. Krieger, S. Lis, B. Gallhofer, Cognitive subprocesses and schizophrenia: A. Reaction-time decomposition, Acta Psychiatr. Scand. Suppl. (2001) 18–27. [30] Y. Kutukcu, W.J. Marks Jr., D.S. Goodin, M.J. Aminoff, Simple and choice reaction time in Parkinson's disease, Brain Res. 815 (1999) 367–372. [31] L.E. Longstreth, N. el-Zahhar, M.B. Alcorn, Exceptions to Hick's law: explorations with a response duration measure, J. Exp. Psychol. Gen. 114 (1985) 417–434. [32] R.K. Mahurin, F.J. Pirozzolo, Application of Hick's law of response speed in Alzheimer and Parkinson diseases, Percept. Mot. Skills 77 (1993) 107–113. [33] J.L. Mathias, E.D. Bigler, N.R. Jones, S.C. Bowden, M. BarrettWoodbridge, G.C. Brown, D.J. Taylor, Neuropsychological and information processing performance and its relationship to white matter changes following moderate and severe traumatic brain injury: a preliminary study, Appl. Neuropsychol. 11 (2004) 134–152. [34] S.N. Mattson, E.P. Riley, Brain anomalies in fetal alcohol syndrome, in: L. E. Abel (Ed.), Fetal Alcohol Syndrome: From Mechanism to Prevention, CRC Press, Boca Raton, FL, 1996, pp. 51–68.
[35] S.N. Mattson, E.P. Riley, E.R. Sowell, T.L. Jernigan, D.F. Sobel, K.L. Jones, A decrease in the size of the basal ganglia in children with fetal alcohol syndrome, Alcohol Clin. Exp. Res. 20 (1996) 1088–1093. [36] S.N. Mattson, E.P. Riley, L. Gramling, D.C. Delis, K.L. Jones, Heavy prenatal alcohol exposure with or without physical features of fetal alcohol syndrome leads to IQ deficits, J. Pediatr. 131 (1997) 718–721. [37] J. Merkel, Die zeitlichen Verhaltnisse det Willensthautigkeit, in: R.S. Woodworth (Ed.), Experimental Psychology, Holt, New York, 1938. [38] T. Muller, E. Eising, W. Kuhn, T. Buttner, H.H. Coenen, H. Przuntek, Delayed motor response correlates with striatal degeneration in Parkinson's disease, Acta Neurol. Scand. 100 (1999) 227–230. [39] J.C. Nunnally, I.H. Bernstein, Psychometric Theory, McGraw-Hill, New York, 1994. [40] K.D. O'Malley, J. Nanson, Clinical implications of a link between fetal alcohol spectrum disorder and attention-deficit hyperactivity disorder, Can. J. Psychiatry 47 (2002) 349–354. [41] A.J. Raynor, Fractioned reflex and reaction time in children with developmental coordination disorder, Motor Control 2 (1998) 114–124. [42] E.P. Riley, S.N. Mattson, E.R. Sowell, T.L. Jernigan, D.F. Sobel, K.L. Jones, Abnormalities of the corpus callosum in children prenatally exposed to alcohol, Alcohol Clin. Exp. Res. 19 (1995) 1198–1202. [43] T.M. Roebuck-Spencer, S.N. Mattson, S.D. Marion, W.S. Brown, E.P. Riley, Bimanual coordination in alcohol-exposed children: role of the corpus callosum, J. Int. Neuropsychol. Soc. 10 (2004) 536–548. [44] M.A. Rogers, M.A. Bellgrove, E. Chiu, C. Mileshkin, J.L. Bradshaw, Response selection deficits in melancholic but not nonmelancholic unipolar major depression, J. Clin. Exp. Neuropsychol. 26 (2004) 169–179. [45] P.D. Sampson, A.P. Streissguth, F.L. Bookstein, R.E. Little, S.K. Clarren, P. Dehaene, J.W. Hanson, J.M. Graham Jr., Incidence of fetal alcohol syndrome and prevalence of alcohol-related neurodevelopmental disorder, Teratology 56 (1997) 317–326. [46] H.H. Samson, Microcephaly and Fetal Alcohol Syndrome: Human and Animal Studies, in: J.R. West (Ed.), Alcohol and Brain Development, Oxford University Press, New York, 1988, pp. 167–183. [47] R.A. Schmidt, T.D. Lee, Motor Control and Learning: A behavioral emphasis, Human Kinetics, Champaign ILL, 1999. [48] M.A. Serlin, R.C. Seaman, Equivalence confidence intervals for twogroup comparisons of means, Psychol. Methods 3 (1996) 403–411. [49] R.W. Simmons, T. Wass, J.D. Thomas, E.P. Riley, Fractionated simple and choice reaction time in children with prenatal exposure to alcohol, Alcohol Clin. Exp. Res. 26 (2002) 1412–1419. [50] I.M. Smith, C.D. Coles, J. Lancaster, P.M. Fernhoff, The effect of volume and duration of prenatal ethanol exposure on neonatal physical and behavioral development, Neurobehav. Toxicol. Teratol. 8 (1986) 375–381. [51] I.M. Spigel, Lift reaction time and topographic compatibility of the S-R field, J. Gen. Psychol. 72 (1965) 165–172. [52] A.P. Streissguth, K. O'Malley, Neuropsychiatric implications and longterm consequences of fetal alcohol spectrum disorders, Semin. Clin. Neuropsychiatry 5 (2000) 177–190. [53] A.P. Streissguth, H.M. Barr, D.C. Martin, C.S. Herman, Effects of maternal alcohol, nicotine, and caffeine use during pregnancy on infant mental and motor development at eight months, Alcohol Clin. Exp. Res. 4 (1980) 152–164. [54] A.P. Streissguth, D.C. Martin, H.M. Barr, B.M. Sandman, G.L. Kirchner, B.L. Darby, Intrauterine alcohol and nicotine exposure: attention and reaction time in 4-year-old children, Dev. Psychol. 20 (1984) 533–541. [55] A.P. Streissguth, H.M. Barr, P.D. Sampson, J.C. Parrish-Johnson, G.L. Kirchner, D.C. Martin, Attention, distraction and reaction time at age 7 years and prenatal alcohol exposure, Neurobehav. Toxicol. Teratol. 8 (1986) 717–725. [56] A.N. Streissguth, F.L. Bookstein, P.D. Sampson, H.M. Barr, Attention: prenatal alcohol and continuities of vigilance and attentional problems from 4 through 14 years, Dev. Psychopathol. 7 (1995) 419–446. [57] W.H. Teichner, M.J. Krebs, Laws of visual choice reaction time, Psychol. Rev. 81 (1974) 75–98. [58] C. Vickrey, A. Neuringer, Pigeon reaction time, Hick's law, and intelligence, Psychon. Bull. Rev. 7 (2000) 284–291.
R.W. Simmons et al. / Neurotoxicology and Teratology 28 (2006) 278–285 [59] M.G. Wade, K.M. Newell, S.A. Wallace, Decision time and movement time as a function of response complexity in retarded persons, Am. J. Ment. Defic. 83 (1978) 135–144. [60] T.S. Wass, R.W. Simmons, J.D. Thomas, E.P. Riley, Timing accuracy and variability in children with prenatal exposure to alcohol, Alcohol Clin. Exp. Res. 26 (2002) 1887–1896. [61] K.F. Widaman, J.S. Carlson, Procedural effects on performance on the Hicks paradigm: bias in reaction time and movement time parameters, Intell 13 (1989) 63–85.
285
[62] K. Wisniewski, M. Dambska, J.H. Sher, Q. Qazi, A clinical neuropathological study of the fetal alcohol syndrome, Neuropediatrics 14 (1983) 197–201. [63] J. Yordanova, V. Kolev, J. Hohnsbein, M. Falkenstein, Sensorimotor slowing with ageing is mediated by a functional dysregulation of motorgeneration processes: evidence from high-resolution event-related potentials, Brain 127 (2004) 351–362.