Analysis of attention and analogical reasoning in children of poverty

Applied Developmental Psychology 27 (2006) 125 – 135

Analysis of attention and analogical reasoning in children of poverty Tara N. Weatherholt a , Ruby C. Harris a,⁎, Barbara M. Burns a , Catherine Clement b a

Department of Psychological and Brain Sciences, University of Louisville, Louisville, KY 40292, United States b Psychology Department, Eastern Kentucky University, United States Available online 27 January 2006

Abstract This study examined the relationship between specific attentional aspects of processing capacity and analogical reasoning in children from low-income families. 77 children aged 48–77 (M = 56.7) months were assessed on an analogical reasoning task (matrices subtest of the Kaufman Brief Intelligence Test) and on computerized attention tasks designed to assess orienting, vigilance, and executive attention abilities [Posner, M.I., and Petersen, S.E. (1990). The attention system of the human brain. Annual Review of Neuroscience, 13, 25–42]. Results showed that analogical reasoning abilities were associated with the executive attention network abilities, suggesting that skills associated with this network, such as the resolution of conflicts between competing demands on attention, may be particularly important for relational mapping. This was evident in girls only. Implications for understanding how attentional components of processing capacity can affect children's academic success in impoverished environments are discussed. © 2006 Elsevier Inc. All rights reserved. Keywords: Analogical reasoning; Attention networks; Orienting; Vigilance; Executive attention; Relational mapping; Children; Poverty; Sex differences

1. Introduction Analogical reasoning is an important cognitive skill involved in abstract mental processes such as creating metaphors, constructing explanations, and solving complex problems (Goswami, 2001). Researchers describe analogical reasoning as achieved when similarity judgments shift from simple perceptual feature comparisons to more complex reasoning based on common relational structures (Gentner, 1989; Goswami & Brown, 1990). Researchers have begun to study the mechanisms underlying the development of analogical reasoning and the age at which the shift from similarity judgments based on perceptual similarity to relational mapping appears (Goswami & Brown, 1990; Halford, 1989; Kotovsky & Gentner, 1996; Markman & Gentner, 1993). Research shows that infants as young as eleven months old make inferences between objects based on perceptual similarity. ⁎ Corresponding author. Tel.: +1 502 852 2348. E-mail address: [email protected] (R.C. Harris). 0193-3973/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.appdev.2005.12.010

126

T.N. Weatherholt et al. / Applied Developmental Psychology 27 (2006) 125–135

For example, an infant might manipulate a red apple in a manner similar to a previously encountered red ball because they share similar features like color and roundness (Baldwin, Markman, & Melartin, 1993). Additionally, Brown (1989) found that one- and two-year olds could solve a transfer problem that required them to obtain an out of reach toy by recognizing that perceptually different tools such as a cane and rake could be used to achieve the same outcome because of their similar enabling qualities. It is believed that the major accomplishment of relational mapping between different structures arises from such early inferences based on similarity while irrelevant object features are ignored (Gentner, 1989). While evidence has shown that infants can make similarity judgments and 2 year olds are capable of recognizing functional relational similarities, full comprehension of more sophisticated abstract relational mapping is not achieved until the preschool years. For example, Holyoak and Thagard (1995) concluded that a “mental leap” occurs between the ages of 3 and 5 years as reasoning formerly based on object similarity shifts to more advanced processes using relational mapping. To make successful relational mappings, children must selectively pay attention to the relevant relations between objects while ignoring distracting mappings that may be perceptually similar (Goswami & Brown, 1990; Ratterman & Gentner, 1998). That is, the child must sort through possible matches between a source and a target, many of which may involve irrelevant perceptual similarities, before finding a match based on relational similarity. For example, a relational mapping question might present a child with pictures of a loaf of bread (A) paired with a slice of bread (B); then a picture of a whole lemon (C) is presented. The child's task is to select the correct picture (D) from an array to pair with (C) that corresponds to the first A:B relationship, which in this case would be a slice of lemon (Goswami & Brown, 1990). In the problem the child must ignore distracting (D) choices that are perceptually similar, such as a lemon cut in half or another whole lemon. According to Gentner (1989), younger children must overcome a preference for the concrete similarities of objects to be mapped. Several studies indicate that success at ignoring the concrete similarities and at finding the relational similarities increases significantly in the preschool years (Gentner & Rattermann, 1991; Kotovsky & Gentner, 1996; Ratterman & Gentner, 1998). Theoretical accounts that explain the underlying mechanisms and variability in the age of children's shift to relational mapping strategies generally represent two positions. First, there are theoretical approaches that stress the importance of domain knowledge to the ability to make relational mappings in a particular domain (Brown, 1989; Crisafi & Brown, 1986; Gentner, 1989; Gentner & Rattermann, 1991; Kotovsky & Gentner, 1996; Vosniadou, 1989). Other theoretical accounts argue for the contribution of maturation and a global change in children's cognitive processing capacity (Halford, Wilson, & Phillips, 1998). For example, Halford and colleagues state that higher order relational mappings, such as those used in analogies, require a high number of propositional arguments, placing processing demands on working memory. Halford and colleagues argue that analogical reasoning develops with changes in strategies and individuals' capacity for parallel processing of complex relations with multiple arguments (e.g., Halford & McCredden, 1998; Halford et al., 1998). 1.1. Analogical reasoning and attention skills The present study explored an aspect of processing capacity in the development of analogical mapping that had not been previously examined. Specific attentional requirements of analogical mapping were explored. Particular components of attention may be differentially important for relational mapping. Recently attentional processes have been examined in terms of three networks of attention: the orienting, alerting, and executive networks (Posner & Petersen, 1990). The orienting network is involved in focusing, disengaging, and shifting spatial attention. The alerting network is related to the maintenance of an alert state and sustaining attention. The executive network controls executive functions, such as goal-directed behavior, target detection, error detection, resolving conflict among responses, and inhibition of an automatic response. The executive network also appears to be required during tasks that require mental effort (Jones, Rothbart, & Posner, 2003; Posner & Rothbart, 1998). All three components are assumed to play a role in an analogy task but the executive network is expected to be especially important. The orienting network may help direct attention to, and shift attention within, task relevant information (at least with spatial problems). The alerting network may help maintain focus on the task over time. But the executive network is expected to be particularly relevant because analogical reasoning requires the inhibition of attention to irrelevant perceptual features of stimuli and the direction of attention to relational information. This conflict between salient perceptual features and more hidden relational matches is precisely what makes analogical reasoning

T.N. Weatherholt et al. / Applied Developmental Psychology 27 (2006) 125–135

127

challenging. Analogy research indicates that concrete perceptual similarities are often noticed automatically, whereas relational similarities, even for adults, are detected only with cognitive effort (Clement, Mawby, & Giles, 1994; Genter, Rattemann, & Forbus, 1993; Waltz, Lau, Grewal, & Holyoak, 2000). Thus, the functions of the executive network to inhibit automatic responses, to resolve conflicting demands on attention, and to support tasks requiring mental effort, would appear to be key components of successful analogical mapping. Consistent with these proposals, Waltz et al. (2000) demonstrated that placing many requirements on the executive component of adult working memory specifically interfered with relational but not object matches between analogous visual stimuli. It should be noted that the executive function of working memory is presumed to overlap with the executive component discussed in the attentional research (Jones et al., 2003). The relevance of the executive network for analogical reasoning has also been supported by neuroimaging studies. Neuroimagery technology has demonstrated the neurological components of the three attention networks as well as those that may be similarly important for analogical reasoning tasks. Recent PET and fMRI research findings have shown that analogical reasoning tasks activate neural systems of working memory, and particularly appear to involve areas related to the central executive (Luo et al., 2003; Wharton et al., 2000). Selective attention activities of the executive network activate neurological pathways of the left frontal lobe and anterior cingulate cortex (Duque & Posner, 2001). These pathways have also been implicated in analogical reasoning (Luo et al., 2003). Given that certain neurological components may underlie both executive attentional skills and analogical reasoning, it is important to understand the relation between specific aspects of attention and analogical transfer. 1.2. Analogical reasoning and attention in children The contribution of each attention network to analogical reasoning in children has not been examined previously. Information about the relation between measures of the three attention networks and analogical reasoning performance can help establish the relative importance of these attention skills in successful analogical reasoning. These attention abilities develop rapidly during the preschool years, which may be important given that this is also the time that children are beginning to use relational mapping to acquire new knowledge. A comprehensive understanding of how attention relates to analogical reasoning requires a framework that examines each network independently at an age when the greatest variability could be seen. A better understanding of this relationship has important implications for early intervention research and implementation. In particular, understanding this relationship will be important for helping children who are at risk for attention and reasoning difficulties. 1.3. Disadvantaged environments, attention, and analogical reasoning Previous research has shown that children from low SES backgrounds have poorer sustained attention as compared to children from middle SES backgrounds (Levy & Hobbes, 1979; Norman & Breznitz, 1992). Moreover, children from low-income families show lower literacy skills and academic achievement (Bryant, Burchinal, & Lau, 1994; Eamon, 2002), and lower scores on intelligence tests (Klebanov, Brooks-Gunn, McCarton, & McCormick, 1998). We know of no research relating children's attentional skills to analogical reasoning in low or middle SES populations. This is surprising given the importance of analogical reasoning for cognitive and academic development (Logie & Baddeley, 1987). It is particularly important to understand the extent to which attention skills in children from low-income backgrounds affects analogical reasoning abilities in order to address SES differences in performance that might be derived from attentional differences. 1.4. The current study In the current study, individual differences in children's skills on the three attention networks (see Posner & Petersen, 1990) were studied in order to understand the relationship between attention and analogical reasoning. The executive attention network was expected to have the strongest relation to analogical reasoning performance given the functional skills and neurological components it shares with analogical reasoning. In addition, this relation was investigated in children from low SES families to better characterize the negative effects of poverty on individual differences in children's analogical reasoning.

128

T.N. Weatherholt et al. / Applied Developmental Psychology 27 (2006) 125–135

2. Methods 2.1. Participants Participants were 77 children (40 males and 37 females) between the 48 and 71 months old (M = 56.7, SD = 5.78). Participants were recruited from several Head Start sites or a medical clinic serving low-income families. The ethnic group composition of the sample was 92% African American, 5% Caucasian, and 3% other. Parents were informed of the study through fliers at their children's school and their medical clinic. Interested parents contacted researchers for further information. All families qualified for enrollment at each site based on their income levels under Federal Poverty Guidelines. Table 1 provides more detailed information regarding income level, family structure, and maternal education of the sample of 77 families. Following the completion of the study, the parent received a monetary stipend and the child received a small toy and a t-shirt. 2.2. Measures 2.2.1. Computerized attention tasks Three computerized attention tasks designed to tap the three attention networks identified by Posner and Petersen (1990) were administered to the children. Berger, Jones, Rothbart, & Posner (2000) designed these games based on RT paradigms used in previous cognitive and neuropsychological studies examining various aspects of attention. An orienting task, vigilance task, and executive task (Berger et al., 2000) were employed on a standard Pentium II computer with a 38.1 cm touch screen monitor. For each task, reaction times (RTs) and accuracy were automatically recorded. The dependent variables of median reaction time (MRT) and percent correct accuracy were calculated from these data. For each task there were five practice trials and 32 test trials. 2.2.1.1. Orienting task. Two glass fishbowls appeared to the left and right of a fixation point. A fish appeared randomly inside one of the fishbowls. The child was told to touch the fish as quickly as possible when it appeared in one of the fishbowls. Before each trial, a central fixation point appeared. Following the fixation stimulus, a cue appeared consisting of a color change in one of the fishbowls from light blue to dark blue to light blue. Duration of the cue was 500 ms and appeared at both fishbowls with equal chance. 2.2.1.2. Vigilance task. The vigilance task was a measure of the child's arousal system following an auditory warning signal. Animals (targets) appeared on the center of the screen one at a time. The child was told that in order to catch an animal, he/she had to touch it. Following completion of the task, a picture appeared on the screen with all of the animals back on the farm. 2.2.1.3. Executive task. The executive, or spatial conflict, task was designed to tap the executive network in which the child has to resolve a cognitive conflict between the location of the stimulus and the location of the response. Children were asked to attend to a stimulus in one spatial location, but they may have to respond (touch) to the opposite location. At the start of each trial, two houses appeared on the left and right bottom corners of the screen with a picture in each house. A central fixation was used in order to maintain attention before each trial began. A picture then appeared on either the left upper or right upper corner. A compatible trial consisted of an ipsilateral configuration between the picture and the inside-the-house picture whereas an incompatible trial consisted of a contralateral configuration. 2.2.2. Assessment of analogical reasoning The Kaufman Brief Intelligence Test (K-BIT; Kaufman & Kaufman, 1990) is a brief, individually administered measure of verbal and nonverbal intelligence that has been standardized for children and adults from 4 to 90 years of age. The test consists of 2 subscales, Vocabulary and Matrices, and yields an overall score known as the K-BIT IQ Composite. Split-half reliability coefficients for the Vocabulary, Matrices, and IQ Composite scores for 5-year-old children were .92, .74, and .88, respectively. Test–retest reliability for children 5–12 years for the Vocabulary, Matrices, and IQ Composite scores were .86, .83, and .92, respectively. In studies to establish construct validity in children age 4– 6, K-BIT IQ Composite scores were significantly correlated with the Mental Processing Composite (r = .58) and the Achievement test of the Kaufman Assessment Battery for Children (K-ABC; r = .74). To establish concurrent validity,

T.N. Weatherholt et al. / Applied Developmental Psychology 27 (2006) 125–135

129

Table 1 Percentage of families with various types of family structure, income and maternal education levels (N = 77) Variable Income level $0–8859 8860–11,939 11,940–15,019 15,020–18,099 18,100–21,179 21,180–24,259 24,260–27,339 27,340–30,419 30,420–49,999 Family structure 1 Adult N 1 Adult 1 Child 2 Children 3 Children 4 Children 5 Children N 5 Children Maternal education Completed high school (HS) Some education beyond HS Did not complete HS

Percentage 45 12 15 5 7 3 3 3 8 58 42 25 33 25 12 4 1 25 64 21

the K-BIT IQ Composite was compared to another brief test of intelligence, the Slossen Intelligence Test (r = .76). Sampling techniques were employed to obtain a standardization sample representative of the U.S. population on sex, ethnicity, geographic region, and SES. In addition, items were eliminated if they demonstrated item bias when comparing scores between males and females and groups based on ethnicity. Performance on the matrices subtest was taken as the primary measure of analogical reasoning. This subtest measures nonverbal skills and the ability to solve new problems by assessing the child's ability to perceive visual relationships and to complete analogies. It is composed of meaningful visual stimuli, such as objects and people, and abstract visual stimuli, such as designs and symbols. The easiest items require the child to select one picture (out of five), which goes best with a stimulus picture (such as a bone goes with a dog). The next set of items involves choosing one of six or eight pictures that best completes a visual analogy (e.g., a hat goes with a head just as a shoe goes with a foot). The last set of matrices problems includes abstract stimuli and requires the child to complete a pattern of dots or to solve a 2 × 2 or 3 × 3 matrix. Raw scores are first calculated by subtracting the total number of items the child gets incorrect from the highest item administered (ceiling) on the matrices subtest. Raw scores may range from 0 to 48. These raw scores were then converted to standard scores used in subsequent analyses. 2.2.3. Verbal intelligence The expressive vocabulary subtest (Kaufman & Kaufman, 1990) of the Kaufman Brief Intelligence test is a measure of verbal IQ that was used as a control variable in subsequent analyses. This subtest measures expressive vocabulary skill by requiring the child to verbally express the name of visually presented objects. An example of an easy item may be a picture of a “bed” or “window” whereas a more difficult item may show a picture of “binoculars” or an “anchor.” Raw scores are first calculated by subtracting the total number of items the child gets incorrect from the highest item administered (ceiling) on the expressive vocabulary subtest. Raw scores may range from 0 to 45. These raw scores are then converted to standard scores used in subsequent analyses. 3. Procedure The child was seated at a table with a touch screen monitor 15.24 cm away from the edge of the table. The child was first asked which hand and finger he or she would like to use to play the computer games. The child was then asked to

130

T.N. Weatherholt et al. / Applied Developmental Psychology 27 (2006) 125–135

Table 2 Mean (and SD) reaction time (MRT) and accuracy scores on the three attention tasks for the current sample and for a middle-income comparison sample Measures

Orienting MRT Orienting accuracy Vigilance MRT Vigilance accuracy Executive MRT Executive accuracy a

Middle-income study a

Current study sample Females (n = 37)

Males (n = 40)

(N = 68)

1226.69 (672.42) 98.3 (3.6) 893.47 (167.39) 98.1 (4.2) 1881.66 (462.94) 93.2 (12.5)

1126.99 (251.22) 98.4 (3.0) 941.48 (188.47) 98.9 (2.2) 1671.66 (435.23) 92.7 (9.8)

1124.46 (225.17) 97.9 (3.0) 878.38 (138.90) 98.8 (2.4) 1864.67 (399.44) 93.3 (8.5)

The average median reaction times and accuracies reported here were obtained from Snyder et al. (submitted for publication).

place that finger on a sticker, which was located right at the edge of the table in front of the touch screen. Between each trial for each task, the child was asked to return his or her finger to the sticker, so that each child was starting the trials from a standard point. The tasks were presented in the following order: orienting, vigilance, and executive. Berger et al. (2000) did not suggest that the tasks be presented in a standard order however; this order was chosen for the current study so as to increase task demands. Attention tasks were recorded on a digital camera and subsequently reviewed for missed trials and/or trials in which reaction time was affected by extraneous variables. In general the orienting task took 5 min, the alert task took 4 min, and the executive task took 7 min, for an average total of 15 min per child. Children were individually given both the vocabulary and matrices subtests of the K-BIT following the completion of the three attention tasks. 4. Results 4.1. Preliminary analyses The average (and SD) median reaction times and accuracy scores for the three attention tasks are presented in Table 2 separately for female and male children. Collapsed across sex of child, the average (and SD) median reaction times for the orienting, alerting, and executive attention tasks were 1174.89 (499.07), 918.41 (179.12), and 1772.57 (458.12), respectively. Additionally, the mean (and SD) scores on the analogical reasoning task (KBIT matrices) and expressive vocabulary subtest scores were 97.53 (12.18) and 88.49 (8.77), respectively. Few studies have employed the attention network tasks with at-risk samples and no standardized scores are available. However, for comparison, average median reaction times reported in similar studies from our lab are also shown in Table 2 (Snyder, Davis, Burns, & Robinson, submitted for publication). This comparison sample consisted of 68 children with a mean (SD) age of 58.08 (6.51) months. The sample was primarily middle class with an average household income between $24,260 and $27,339 and was composed of 48.5% Caucasian, 45.6% African American, and 5.9% other ethnic group children. Inspection of the average median reaction times for the middle class sample and for the current sample on three attention tasks reveals no wide disparities.

Table 3 Intercorrelations among child sex, age, attention task MRTs, verbal IQ and K-BIT matrices scores Measure 1. Age 2. Sex 3. K-BIT verbal IQ 4. Matrices 5. Executive MRT 6. Orient MRT 7. Vigilance MRT ⁎p b .05.

2

3

4

5

6

7

− .10 –

− .02 − .03 –

.07 .01 − .34 ⁎ –

.00 .23 ⁎ −.10 −.29 ⁎ –

− .22 .10 − .07 − .04 − .25 ⁎ –

− .13 − .14 .03 − .10 .37 ⁎ − .29 ⁎ –

T.N. Weatherholt et al. / Applied Developmental Psychology 27 (2006) 125–135

131

Table 4 Summary of results of hierarchical regression analyses predicting K-BIT matrices scores from orienting and alert attention (vigilance) MRT R2Δ

Variable

Step 1 Verbal IQ Chronological age Step 2 Attention task MRT

β

B

Orienting

Vigilance

Orienting

Vigilance

.12

.12

.48 .16

.48 .13

.00

.01

.00

−.00

Orienting

Vigilance

.34 .08

.35 .06

− .001

− .10

In the sample of the present study, no significant differences were found between males and females on median reaction time (MRT) averaged across all three attention tasks, t(76) = − .956, p N .05, or on the analogical reasoning KBIT matrices score, t(76) = − .147, p N .05. However, when analyses were conducted for each attention network considered separately, a significant sex difference was found for the executive attention task t(75) = −.052, p b .05 but not the orienting, t(75) = − .874, p N .05, or vigilance, t(75) = 1.178, p N .05, tasks. Additionally, a significant correlation was found between sex of the child and the executive attention network task, r = .23, p b .05. Correlations among all measures are shown in Table 3. As indicated in Table 3, sex of the child was related to executive attention MRT, as expected given the mean comparisons. The MRTs of the three aspects of the attention network were intercorrelated. In addition, the overall MRT and overall accuracy (averaged across each attention task for each participant) were significantly negatively correlated, r = − .41, b .01. Thus, no evidence for speed–accuracy trade-offs existed. In other words, the speed at which the children performed the task did not compromise their accuracy. 4.2. Regression analyses The purpose of the present study was to examine whether children's analogical reasoning ability was related to the networks of attention, specifically the executive attention task. To determine this, hierarchical regressions were performed with each attention network task MRT to predict K-BIT matrices performance scores. In subsequent regression analyses, the child's age and verbal IQ were controlled for in the first block to account for the contribution of verbal experience to analogical performance. In the second block of each regression the median reaction time (MRT) for each attention network was entered. For the first regression, verbal IQ and chronological age were entered in the first block and orienting attention MRT was entered in the second block (see Table 4, left columns). For the second regression, verbal IQ and chronological age were again controlled for in the first block and vigilance attention MRT was entered in the second block (see Table 4, right columns). Neither of these regression analyses produced significant findings. Next, separate regression analyses were performed for females and males with analogical reasoning scores (K-BIT matrices) as the dependent variable and executive attention MRT as the predictor variable because prior analyses had revealed sex differences in executive attention MRTs. For each of these analyses, verbal IQ and chronological age were again entered in the first block. The results, summarized in Table 5, showed that executive attention MRT significantly

Table 5 Summary of results of hierarchical regression analyses predicting female and male children's K-BIT matrices performance from executive attention MRT R2Δ

Variable

Step 1 Verbal IQ Chronological age Step 2 Executive MRT ⁎p b .05.

β

B

Females

Males

Females

Males

Females

Males

.07

.23

.16 − .20

.54 .53

.10 − .10

.41 .23

.13⁎

.04

− .01

.00

− .38

.16

132

T.N. Weatherholt et al. / Applied Developmental Psychology 27 (2006) 125–135

predicted performance on the K-BIT matrices subtest for female children (left column), and accounted for 13.2% of the variance, F(1, 33) = 5.45, p b .05. The regression analysis was not significant for male children (see Table 5, right column). The skills associated with the executive attention network appeared to be related to analogical reasoning in females but not in males. Female children who had faster reaction times on the executive attention task had higher scores on the analogical reasoning task. In contrast, performance on the vigilance and orienting tasks did not account for a significant amount of variance to analogical reasoning performance and there were no sex differences observed when vigilance and orienting tasks were considered. 5. Discussion We found that skills tapped by the executive attention network were related to analogical reasoning skills in four- to six-year old children from low-SES families. There is some evidence that similar neurological components appear to be involved in both the executive attention network and analogical reasoning. Previous research has shown that tasks requiring executive attention skills activate neurological pathways of the midline frontal cortex including the anterior cingulate cortex (ACC) and prefrontal cortex (Berger & Posner, 2000; Duque & Posner, 2001). Similar neurological structures were also shown by Wharton et al. (2000) to be activated during analogy comparisons. Moreover, research has shown that particular cognitive aspects of the ACC relevant to analogical reasoning include attention regulation and response selection in the face of competing information demands (Bush, Luu, & Posner, 2000). Thus, the regulatory functions of the ACC used for resolving cognitive conflicts during the executive attention task may also be engaged when the child must ignore salient but irrelevant perceptual information, and must direct attention to more difficult-todetect relational matches. Given that previous research has found that similar neurological components are utilized when performing tasks of executive attention and analogical reasoning, the results of our study provide support that regulatory components of executive attention may be involved in relational mapping. This is not surprising given the similar role of spatial conflict resolution for executive attention and analogical reasoning. The present study found differences in the pattern of executive attention network and analogical reasoning performance for boys and girls. Faster reaction times were significantly related to higher analogical reasoning scores in female children. While the same trend appeared in male children it was not significant. Research examining the attention networks in children has yielded inconsistent results in regards to the child's sex (Chang & Burns, 2005; Mezzacappa, 2004). In the current study girls had significantly slower median reaction times (1881.66 ms) compared to males (1671.66 ms) on the executive attention task, which may have significantly affected their analogical reasoning performance. Previous studies have similarly found sex differences in spatial conflict tasks to vary with age (Rothbart, Ellis, Rueda, & Posner, 2003). Further research is needed to determine if the sex differences in executive attention and analogical reasoning found in the current study provide additional support for Rothbart et al. (2003) or are due to restricted age range. Future studies will be necessary to replicate this finding as well as to examine further possible interactions of sex of the child with other regulatory mechanisms as they relate to analogical reasoning performance. The results highlight the importance of understanding individual differences in attention skills when preparing children for the transition from preschool to kindergarten. Given that skills such as error detection, resolving conflict among responses, and inhibition of an automatic response are important for both executive attention and analogical reasoning, students could benefit from educational practices that emphasize the process of the task rather than the speed of the response (see Posner & Rothbart, 2005; Rothbart & Jones, 1998). 5.1. Limitations and implications Limitations of the current study suggest several directions for future research. First, the current study yielded a small effect size for the relation between executive attention and analogical reasoning in girls, which may make generalization of the findings difficult. The amount of variance accounted for by the executive attention task is not a large contribution, but it does highlight the need for future research in this direction. A second limitation is that the measure of analogical reasoning in the current study was only visual. It may be that different components of attention are involved when alternate forms of analogical reasoning, such as verbal analogies are performed. Other measures of analogical reasoning should be used to explore the role of the three attention networks in making relational mappings.

T.N. Weatherholt et al. / Applied Developmental Psychology 27 (2006) 125–135

133

Lastly, the current study did not investigate the relation of the attention networks to analogical reasoning over time. Future studies should employ longitudinal designs to determine the predictive power of attention for analogical reasoning. Given the variability and changes in attention skills of preschool children (Rothbart et al., 2003; Ruff & Rothbart, 1996) it would be useful to examine this relationship longitudinally to determine the impact of attention skills on later analogical success. The question remains as to whether executive attention skills are a precursor for successful analogical reasoning, which could begin to be addressed with longitudinal research. The findings from the current study have implications for understanding how attentional components of processing capacity impact children's academic success (see also Burns et al., in press). Researchers agree upon the importance of analogical reasoning for the development of higher-level human reasoning (Goswami, 1992; Halford, 1993; Kotovsky & Gentner, 1996). Children use analogical reasoning as a mechanism to acquire new knowledge and build new schemata (Vosniadou, 1989). Additionally, academic achievement and giftedness in children has been associated with performance on measures of analogical reasoning (Caropreso & White, 1994; Marr & Sternberg, 1986). The regulation of reasoning processes in working memory may be of special importance to academic achievement given its use for such tasks as counting and math (Baddeley & Hitch, 1974; Logie & Baddeley, 1987). Thus, the examination of the regulatory mechanisms of the three attention networks as they relate to analogical reasoning performance provides a more comprehensive understanding of the basic cognitive skills that may be required for higher level thinking. Additionally, the present study was unique in its examination of the extent to which specific aspects of the attention networks impair or support analogical reasoning performance, and how this relationship may be magnified in the context of a high-risk environment. It is well understood that impoverished environments expose children to a number of environmental stressors, which may interfere with regulatory processes such as working memory and attention (Bronfenbrenner, 1999; Matheny et al., 1995; Robinson et al., 2003). Still, it will be important for future studies to replicate the current study in various populations in order to gain a more comprehensive understanding of how impoverished environments impact attention and analogical reasoning. The detrimental effects of poverty on attention are theorized to result from the excess number of negative factors often associated with poverty including fewer educational resources, exposure to environmental toxins, poor parenting strategies, health problems, and a disorganized home environment (Bronfenbrenner, 1999; Matheny et al., 1995; Mezzacappa, 2004). In summary, exposure to such risks in poverty has been associated with deficits in attention regulation, analogical reasoning, and negative academic outcomes (Bronfenbrenner, 1999; Dubow & Ippolitto, 1994; Huston et al., 1994; Lengua, 2002; Mezzacampa, 2004). The current study found that the executive attention network might be more engaged when making relational mappings than the orienting or vigilance networks. Promoting executive attention skills through interventions that focus on the executive attention components of inhibiting responses and thinking through choices more thoroughly may be important for academic settings given that these skills are useful for successful relational mapping. Interventions that focus on training parents to scaffold interactions with their children according to individual differences in executive attention skills may also be important. The role of parents may be especially salient given that they are the primary environmental influence on children's development. Recent research has shown that the amount of direction and structure that a mother provides at age 2 negatively impacted their children's later visual-motor and executive-processing skills (Assel, Landry, Swan, Smith, & Steelman, 2003). The authors suggest that skills important for executive functioning, such as block construction and puzzle solving, may be particularly impaired when mothers are overly directive in an activity rather than focused on the reasoning process. Thus, interventions aimed at helping parents identify the level at which their children need direction in visual–spatial activities that require executive functioning such as puzzle solving may also be successful at improving children's executive attention skills used in analogical reasoning. Furthermore, research suggests that it will be important to implement these interventions early in development to increase school readiness skills and the transition from preschool to kindergarten (Assel et al., 2003). Previous research has suggested that attentional processes may play a mediating role in the relationship between various social and cognitive outcomes associated with poverty (Chang & Burns, 2005). Therefore, it may be particularly important to increase executive attention skills in children from low-income backgrounds to improve their academic success through analogical reasoning performance. Given that low-SES children are at risk for lowered academic achievement, intervening to enhance executive attention skills may have implications for school success via their possible regulatory role for improved reasoning strategies in not only analogies but other academic domains as well. Finally, future studies should attempt to integrate the present findings with other proposals about the development of analogical reasoning. In particular, researchers should examine how executive attention skills relate to specific

134

T.N. Weatherholt et al. / Applied Developmental Psychology 27 (2006) 125–135

proposals about the capacity for processing relational structures of increasing complexity (see Halford, 1989; Halford et al., 1998). Given the importance of analogical reasoning for learning, future research on interventions that target executive attention skills may be a fruitful direction. Acknowledgement The authors would like to thank Dr. Sonia Franco, Director of the Child and Youth Project, Department of Pediatrics, University of Louisville for her support of this research. This study was funded by a grant to Dr. Burns from the Kentucky Science and Engineering Foundation, reference number KSEF-01-114-RDE-001. References Assel, M. A., Landry, S. H., Swank, P., Smith, K. E., & Steelman, L. M. (2003). Precursors to mathematical skills: Examining the roles of visual– spatial skills, executive processes, and parenting factors. Journal of Applied Developmental Science, 7, 27−38. Baddeley, A. D., & Hitch, A. D. (1974). In G. Bower (Ed.), The Psychology of Learning and Motivation, Vol. 8. (pp. 47−89). NY: Academic Press. Baldwin, D. A., Markman, E. M., & Melartin, R. L. (1993). Infants' ability to draw inferences about non-obvious object properties: Evidence from exploratory play. Child Development, 64, 711−728. Berger, A., Jones, L., Rothbart, M. K., & Posner, M. I. (2000). Computerized games to study the development of attention in childhood. Behavior, Research Methods, Instruments, and Computers, 32(2), 297−303. Berger, A., & Posner, M. I. (2000). Pathologies of brain attentional networks. Neuroscience and Biobehavioral Reviews, 24, 3−5. Bronfenbrenner, U. (1999). Environments in developmental perspective: Theoretical and operational models. In S. L. Friedman & T. D. Wachs (Eds.), Measuring environment across the life span: Emerging methods and concepts (pp. 3−28). Washington, DC: American Psychological Association. Brown, A. L. (1989). Analogical learning and transfer: What develops? In S. Vosniadou & A. Ortony (Eds.), Similarity and analogical reasoning (pp. 369−412). New York: Cambridge University Press. Bryant, D. M., Burchinal, M., & Lau, L. B. (1994). Family and classroom correlates of Head Start children's developmental outcomes. Early Childhood Research Quarterly, 9, 289−304. Burns, B., Chang, F., Snyder, E., Robinson, J. B., Davis, D. W., Weatherholt, T., et al. (in press). Characterizing attention regulation in children of poverty. In F. Columbus (Ed.), Child psychology: New research. NY: Nova. Bush, G., Luu, P., & Posner, M. (2000). Cognitive and emotional influences in anterior cingulated cortex. Trends in Cognitive Science, 4, 215−222. Caropreso, E. J., & White, C. S. (1994). Analogical reasoning and giftedness: A comparison between identified gifted and nonidentified children. Journal of Educational Research, 87(5), 271−278. Chang, F., & Burns, B. (2005). Attention in preschoolers: Associations with effortful control and motivation. Child Development, 76, 247−263. Clement, C. A., Mawby, R., & Giles, D. (1994). The effects of manifest relational similarity on analog retrieval. Journal of Memory and Language, 33, 396−420. Crisafi, M. A., & Brown, A. L. (1986). Analogical transfer in very young children: Combining two separately learned solutions to reach a goal. Child Development, 57, 953−968. Dubow, E. F., & Ippolito, M. F. (1994). Effects of poverty and quality of the home environment on changes in the academic and behavioral adjustment of elementary school-age children. Journal of Clinical Child Psychology, 23, 401−412. Duque, D. F., & Posner, M. (2001). Brain imaging of attentional networks in normal and pathological states. Journal of Clinical and Experimental Neuropsychology, 23(1), 74−93. Eamon, M. K. (2002). Effects of poverty on mathematics and reading achievement of young adolescents. Journal of Early Adolescents, 22, 49−74. Gentner, D. (1989). The mechanisms of analogical learning. In S. Vosniadou & A. Ortony (Eds.), Similarity and analogical reasoning (pp. 199−241). Cambridge, MA: Cambridge University Press. Gentner, D., & Ratterman, M. J. (1991). Language and the career of similarity. In S. A. Gelman & J. P. Brynes (Eds.), Perspectives on thought and language: Interrelations in development (pp. 225−277). New York: Cambridge University Press. Gentner, D., Rattermann, M. J., & Forbus, K. D. (1993). The roles of similarity in transfer: Separating retrievability from inferential soundness. Cognitive Psychology, 25, 524−575. Goswami, U. (1992). Analogical reasoning in children. Mahwah, NJ: Erlbaum. Goswami, U. (2001). Analogical reasoning in children. In D. Gentner, K. Holyoak, & B. Korkinov (Eds.), The analogical mind: Perspectives from cognitive science (pp. 437−470). Cambridge, MA: The MIT Press. Goswami, U., & Brown, A. L. (1990). Melting chocolate and melting snowmen: Analogical reasoning and causal relations. Cognition, 35, 69−95. Halford, G. (1989). Cognitive processing capacity and learning ability: An integration of two areas. Learning and Individual Differences, 1(1), 125−153. Halford, G. (1993). Children’s understanding: The development of mental models. Mahwah, NJ: Erlbaum. Halford, G., & McCredden, J. (1998). Cognitive science questions for cognitive development: The concepts of learning, analogy, and capacity. Learning and Instruction, 8, 289−308. Halford, G., Wilson, W., & Phillips, S. (1998). Processing capacity defined by relational complexity: Implications for comparative, developmental, and cognitive psychology. Behavioral and Brain Sciences, 21, 803−865. Holyoak, K., & Thagard, P. (1995). Mental leaps: Analogy in creative thought. Cambridge, MA: The MIT Press. Huston, A. C., McLoyd, V. C., & Garcia-Coll, C. (1994). Children and poverty: Issues in contemporary research. Child Development, 65, 275−282.

T.N. Weatherholt et al. / Applied Developmental Psychology 27 (2006) 125–135

135

Jones, L. B., Rothbart, M. K., & Posner, M. I. (2003). Development of executive attention in preschool children. Developmental Science, 6, 498−504. Kaufman, A. S., & Kaufman, N. L. (1990). Kaufman Brief Intelligence Test. Circle Pine, MN: American Guidance Service. Klebanov, P. K., Brooks-Gunn, J., McCarton, C., & McCormick, M. C. (1998). The contribution of neighborhood and family income to developmental test scores over the first three years of life. Child Development, 69(5), 1420−1436. Kotovsky, L., & Gentner, D. (1996). Comparison and categorization in the development of relational similarity. Child Development, 67, 2797−2822. Lengua, L. J. (2002). The contribution of emotionality and self-regulation to the understanding of children's response to multiple risk. Child Development, 73, 144−161. Levy, F., & Hobbes, G. (1979). The influences of social class and sex on sustained attention (vigilance) and motor inhibition in children. Australian and New Zealand Journal of Psychiatry, 13, 231−234. Logie, R. H., & Baddeley, A. D. (1987). Cognitive processes in counting. Journal of Experimental Psychology: Learning, Memory, and Cognition, 13(2), 310−326. Luo, Q., Perry, C., Peng, D., Jin, Z., Xu, D., Ding, G., et al. (2003). The neural substrate of analogical reasoning: An fMRI study. Cognitive Brain Research, 17, 527−534. Markman, A. B., & Gentner, D. (1993). Structural alignment during similarity comparisons. Cognitive Psychology, 25, 1−37. Marr, D. B., & Sternberg, R. J. (1986). Analogical reasoning with novel concepts: Differential attention of intellectually gifted and nongifted children to relevant and irrelevant novel stimuli. Cognitive Development, 1, 53−72. Matheny, A. P., Wachs, T. D., Ludwig, J. L., & Phillips, K. (1995). Bringing order out of chaos: Psychometric characteristics of the Confusion, Hubbub, and Order Scale. Journal of Applied Developmental Psychology, 16, 429−444. Mezzacappa, E. (2004). Alerting, orienting, and executive attention: Developmental properties and sociodemographic correlates in an epidemiological sample of young, urban children. Child Development, 75(5), 1373−1386. Norman, G., & Breznitz, Z. (1992). Difference in the ability to concentrate in first grade Israeli pupils of low and high socioeconomic status. Journal of Genetic Psychology, 153, 5−17. Posner, M. I., & Petersen, S. E. (1990). The attention system of the human brain. Annual Review of Neuroscience, 13, 25−42. Posner, M. I., & Rothbart, M. K. (1998). Summary and commentary: Developing attentional skills. In J. E. Richards (Ed.), Cognitive neuroscience of attention: A developmental perspective (pp. 317−323). Mahwah, NJ: Erlbaum. Posner, M. I., & Rothbart, M. K. (2005). Influencing brain networks: Implications for education. Trends in Cognitive Sciences, 9, 99−103. Ratterman, M. J., & Gentner, D. (1998). More evidence for a relational shift in the development of analogy: Children's performance on a causalmapping task. Cognitive Development, 13(4), 453−478. Robinson, J., Dossett, D., Chang, F., Snyder, L., Weatherholt, T., Davis, D., et al. (2003, April). Individual differences in attention regulation in children of poverty. Poster presented at the biannual meeting of the Society of Research on Child Development, Tampa, FL. Rothbart, M. K., Ellis, L. K., Rueda, M. R., & Posner, M. I. (2003). Developing mechanisms of temperamental effortful control. Journal of Personality, 71, 1113−1143. Rothbart, M. K., & Jones, L. B. (1998). Temperament, self-regulation, and education. School Psychology Review, 27, 479−491. Ruff, H. A., & Rothbart, M. K. (1996). Attention in early development. New York: Oxford. Snyder, E., Davis, D. W., Burns, B., & Robinson, J. (submitted for publication). Examining attention networks in preschool children born with VLBW. Vosniadou, S. (1989). Analogical reasoning as a mechanism in knowledge acquisition: A developmental perspective. In S. Vosniadou & A. Ortony (Eds.), Similarity and analogical reasoning (pp. 413−437). New York: Cambridge University Press. Waltz, J. A., Lau, A., Grewal, S. K., & Holyoak, K. J. (2000). The role of working memory in analogical mapping. Memory and Cognition, 28, 1205−1212. Wharton, C. M., Grafman, J., Flitman, S. S., Hansen, E. K., Brauner, J., Marks, A., et al. (2000). Toward neuroanatomical models of analogy: A positron emission tomography study of analogical mapping. Cognitive Psychology, 40, 173−197.