Attention Deficit Hyperactivity Disorder

INTRODUCTION

Narratives, such as story telling or retelling, play a central role in daily social interactions and in the majority of children's academic activities throughout the school years (Trabasso & Stein, 1997). Examining the ability to comprehend and produce narratives provides one approach to understanding the cognitive processing skills of children, and may prove extremely helpful in clarifying children's academic difficulties. Researchers have examined narratives of atypically developing children in comparison to typically developing children to examine the differences in linguistic skills, social knowledge, and cognitive abilities. The atypically developing groups that have been studied include children with autism (Capps, Losh, & Thurber, 2000), Down's syndrome (Loveland, McEvoy, Tunali, & Kelley, 1990), language impairments (Wagner, Sahlen, & Nettelbladt, 1999), and pervasive developmental disorder (Tager-Flusberg & Sullivan, 1995). The present study continues this line of research by examining the on-line st ory representation of boys with attention deficit hyperactivity disorder (ADHD).

ADHD is characterized by developmentally inappropriate levels of inattention, hyperactivity, and impulsivity (American Psychiatric Association [APA], 1994) that impair functioning in major life domains, such as home, school, and peer relations. Specifically, children diagnosed with ADHD, in comparison with their nonreferred peers, face an increased risk of poor achievement in school (O'Neill & Douglas, 1991), retention in grade (Hinshaw, 1994), and school suspensions and expulsions (Barkley, 1990). Research has attempted to understand how the core symptoms of ADHD may relate to these children's academic problems, including deficient study skills (O'Neill & Douglas, 1991), disorganized work habits (Atkins, Pelham, & Licht, 1985), and inappropriate classroom behaviors (Abikoff, Gittelman-Klein, & Klein, 1977). However, few studies have focused on higher order cognitive processing skills of children with ADHD that may contribute to academic difficulties.

Story Comprehension

Effective story comprehension is an important contributor to early academic success and requires a variety of component skills (Low & Durkin, 1998). These include the strategic allocation of attention to plot-relevant information, the construction of a story representation that reflects the causal connections among events, the monitoring of comprehension, and the use of the story representation to direct retrieval of significant story events (Nezworski, Stein, & Trabasso, 1982; O'Brien & Myers, 1987; Trabasso, Secco, & van den Broek, 1984; van den Broek, 1997). As such, story comprehension is a powerful gauge of cognitive processing and cognitive development (Low & Durkin, 1998).

In the story comprehension literature, two major theoretical perspectives are story grammar theory (e.g., Stein & Glenn, 1979) and the causal network model of story representation (e.g., Trabasso & van den Broek, 1985). Story grammar theory specifies different types of story events (e.g., setting, the protagonist's goal, and attempts to attain the goal) and the episodic structure of a typical story. The network model builds on story grammar theory by representing different kinds of story events and the causal relations among them (Trabasso & van den Broek, 1985; Trabasso, van den Broek, & Suh, 1989; van den Broek, 1990).

A feature shared by both story grammar theory and the network model is a focus on characters' goals and motivations. According to story grammar theory, each episode within a story is structured around a goal (or subgoal necessary in attaining the overall goal). These goals arise from certain story events that in turn motivate other actions and outcomes (Mandler & Johnson, 1977; Stein & Glenn, 1979). According to the network model, a given goal may motivate a number of subsequent action sequences and thus is likely to have many causal connections to other story events. An event with many causal connections plays an important role in maintaining the coherence of the story. Thus, understanding a character's goals is an important influence on overall story comprehension.

Developmental research documents that goals and their causal connections are important influences on story comprehension and recall. Although young children show considerable difficulty in using goal information to enhance story comprehension, goals become a dominant feature guiding older children's story representation (Goldman & Varnhagen, 1986; Trabasso & Nickels, 1992). Further, children as young as 4 years of age recall events with many causal connections more often than events with few causal connections (van den Broek, Lorch, & Thurlow, 1996). In addition, as children develop, they become better able to identify causal relations in a story and to use them to guide recall (van den Broek, 1989).

Story Comprehension in Children With ADHD

Few studies have examined the story comprehension of children with ADHD, and these studies have focused primarily on story recall. A consistent finding in these studies is that children with ADHD reproduce fewer story events than nonreferred comparison children (Lorch, Diener, et al., 1999; Purvis & Tannock, 1997; Zentall, 1988). In addition, two studies suggest that the influence of story structure differs for children with ADHD and comparison children. Tannock, Purvis, and Schachar (1993) found that boys with ADHD retold less cohesive and less organized stories than their nonreferred peers. Lorch, Diener, et al. (1999) found that for both groups of children, recall increased as rated importance and the number of causal connections increased, but the effect of causal connections on recall was greater for nonreferred children than for children with ADHD.

Narration as On-Line Story Comprehension

Although these preliminary studies have been informative, they have focused entirely on the products of comprehension (i.e., recall). As such, they tell us very little about the on-line comprehension processes children employ as they build a story representation. As children encounter story information, effective comprehension requires the ability to identify characters and their relations to one another, and to link events as causes and consequences of other events. The construction of coherent story representations depends on the degree to which children can make such connections (Milch-Reich, Campbell, Pelham, Connelly, & Geva, 1999). Examining children's narrations of stories based on wordless picture books has proven to be a valuable method for understanding children's construction of story representations (Berman & Slobin, 1994; Trabasso & Stein, 1997).

Examination of children's on-line story narrations has been guided by the focus on the importance of characters' goals and motivations that is common to story grammar theory and the network model. Specifically, investigations of children's narrations have emphasized what Trabasso and Nickels (1992) called Goal-Attempt-Outcome sequences. In a typical story structure, Initiating Events occur that spark the protagonist to set up an Overall Goal, followed by Attempts to achieve the Goal. These Attempts enable or cause Outcomes that involve goal attainment (a Positive Outcome) or goal failure (a Negative Outcome; Trabasso & Stein, 1997). In the process of attempting to achieve the original Goal, a hierarchy of Subordinate Goals may be generated, particularly in response to Negative Outcomes. Although a story may contain events that are not part of these Goal-Attempt--Outcome sequences, the events encompassed within them are critical to an organized, coherent story representation.

Investigators have examined children's use of Goal-Attempt-Outcome sequences in their story narrations (Trabasso & Nickels, 1992; Trabasso, Stein, Rodkin, Munger, & Baughn, 1992). A consistent finding concerns developmental changes in children's use of goals in structuring story narrations. Typically, 3- and 4-year-old children primarily engage in identifying the contents of the pictures. By about age 5 children show an increase in representing events from Goal-Attempt-Outcome sequences, and a corresponding decrease in events outside of these sequences. With continued development, children become increasing likely to use goals to link events into a coherent representation, so that by age 9 children's stories are similar in structure to those of adults.

Narration in Children With ADHD

Very little research has compared on-line story representations in children with ADHD and their nonreferred peers. Similar to the procedures used by Trabasso, Stein, and colleagues, Tannock, 0mm, and Fine (1997) had these two groups of children narrate a story from a picture book. Tannock et al. found that children with ADHD encoded fewer goals and fewer attempts over the course of the story, and were less likely to spontaneously correct errors. However, because children with ADHD produced shorter narratives overall, it is unclear whether their difficulties are specific to the use of the goal structure per se or attributable to a general production deficit. In a second study examining the on-line processing of children with ADHD and comparison children (Much-Reich et al., 1999), children were asked to narrate a story from a set of seven pictures. Children with ADHD incorporated fewer causal relations and inferential or causal links between pictures than comparison children, although they included approximatel y the same number of events in their narrations. However, the Milch-Reich et al. (1999) study only included seven pictures for the narration task, limiting the complexity of possible story representations.

The present study builds on these two previous studies by examining the on-line processes that occur as a child tells a story, thereby providing a more detailed understanding of the story representation abilities of children with ADHD. Three questions guided this investigation. First, do the narrations of children with ADHD differ from those of nonreferred children in terms of specific story structure categories? In particular, the developmental literature on story narration points to the importance of goal-related events in structuring cohesive representations (Trabasso & Stein, 1997). Thus, in order to understand the story representation abilities of children with ADHD, the present study examines whether these children are particularly deficient in representing goal-related events.

Second, do the narrations of children with ADHD differ from those of nonreferred children in the errors they produce while telling their stories? If children are having difficulty building a cohesive story representation, one way this may be evident is in the errors they produce. Previous research on recall of stories by children with ADHD has shown reliably that these children make more errors in their recalls, but the types of errors may differ across studies (Lorch, Diener, et al., 1999; Tannock et al., 1993). Because little is known about group differences in errors during on-line story narrations, the present study specifically addresses this issue.

Finally, are there group differences in how children adjust their story narrations after having formed an initial representation of the story? Because on-line story narration is a dynamic process requiring modifications of the representation as new events are revealed, a second chance to tell the story should produce a more coherent representation. Whereas the first narration offers insight into the process of building an initial story representation, the second production provides information about how children use their existing representations to modify their narrations. Therefore, in the present study the children narrated the story a second time, allowing examination of group differences in specific changes in the story narrations.

METHOD

Participants

A total of66 boys, ages 9-11 years, participated in the study. Of the total sample, 22 boys had a confirmed diagnosis of ADHD/Combined type whereas the remaining 44 boys constituted the nonreferred comparison group. Boys with ADHID were referred from a university psychiatry clinic. This clinic diagnosis was made independent of the research study and merely generated the pool of eligible participants. As an additional confirmation of diagnostic status, a semistructured interview was conducted with a parent, usually the mother. The interviews were conducted by advanced graduate students trained in this interview procedure. The interview consisted of questions regarding the symptoms of ADHD taken verbatim from the DSM-IV (APA, 1994). In the interview, parents indicated whether each diagnostic criterion was true of their son, and, if so, they were asked to give behavioral examples. If a behavior was characteristic of the child, parents also indicated if that behavior seemed inappropriate for the child's age and w hether it impaired the child's functioning academically and/or socially. This interview procedure has been used successfully by this research group in previous studies, with interrater reliabilities for number of ADHD symptoms endorsed by the parent to be above 95% (e.g., Lorch, Sanchez, et al., 1999). Any boy referred as ADHD who did not meet criteria set forth in the DSM-IV was excluded. Boys who exhibited only attentional problems were not included in the study because of mounting evidence of differences between the predominantly inattentive and combined groups along important classification dimensions (e.g., demographics, family history, symptom presentation, associated features, comorbid disorders), indicating the inattentive group is a distinct disorder and not a subtype of ADHD (Barkley, 1997; Milich, Balentine, & Lynam, 2001).

Nonreferred boys were recruited through an ad in the local newspaper and flyers distributed in the community. During a recruitment phone call for nonreferred boys, parents were asked if the child had ever been referred for any behavioral or learning problems. Thus, comparison boys had no history of behavioral or learning problems, nor any current medical, behavioral, or attentional problems. Any nonreferred boy who met criteria for three or more symptoms of inattention, hyperactivity/impulsivity, or oppositional defiant disorder on the parent interview was excluded from the study.

Further confirmation of the diagnostic status of the boys with ADHD and the absence of behavioral problems for nonreferred boys was obtained by having a parent complete the Child Behavior Checklist (Achenbach, 1991). Consistent with their diagnostic status, the ADHD group scored significantly higher on the ADHD and oppositional defiant disorder (ODD) sections of the parent interview, as well as on the Externalizing scale of the CBCL (see Table I). Table I indicates that the groups did not differ in terms of age or the Block Design subtest scores from the Wechsler Intelligence Scale for Children-III (WISC-III; Wechsler, 1991). Comparison boys scored significantly higher on the Letter-Word Identification and Passage Comprehension subtests of the Woodcock-Johnson Tests of Achievement--Revised (Woodcock & Johnson, 1989), and on the Vocabulary subtest from the WISC-III. These subtests were the only ones administered and they were selected because of their excellent correlations with overall IQ and reading achievem ent scores from these tests (Sattler, 2001).

Boys in the ADHD group were medication free on the day of the study. Parents were reminded in a phone call the day before testing to discontinue medication for the following day. A reconfirmation of the suspension of the drug was obtained on test day. If parents indicated that the child took medication the morning of testing, the child did not participate in the study that day but instead was rescheduled.

The study was reviewed and approved by the university Institutional Review Board, and informed consent was obtained from the parents of all participants. The entire session lasted approximately 2.5 hr, and included measures additional to those reported here. Boys were paid $20 for their participation.

Materials

The wordless picture book, Frog, where are you? by Mayer (1969), was used for the story narration task because it is rich in opportunities for the coding of a narration. Previous investigators have used this picture book to examine developmental differences in story narration (Trabasso & Nickels, 1992; Trabasso & Rodkin, 1994), language usage (Wagner et al., 1999), and narrative style (Berman & Slobin, 1994). The book contains a total of 24 pictures that provide the narrator the opportunity to describe the actions of the protagonist in terms of a hierarchical goal plan. The first picture establishes that the boy has a dog and a frog as pets. While the boy is sleeping, the frog escapes from his jar, and the boy finds the jar empty in the morning. This establishes the superordinate goal of finding the lost frog and bringing it home. The superordinate goal of finding the lost frog and bringing it home motivates attempts to search for the frog, beginning in the boy's room. Failed outcomes from these attempts resu lt in several subordinate goals to search for the frog in various outdoor locations. Finally, the superordinate goal is achieved when the lost frog is found and brought home by the boy.

Procedures

All boys were tested individually and were seated across from a research assistant blind to the clinical status of the boy. A binder containing the pictures from the book was placed in front of the boy. The research assistant gave the following instructions: "This is a book that tells a story. There are no words to this book, only pictures. You have to make up the words. I want you to go through the book and make up a part of the story for each picture." The research assistant made sure the boy understood the directions and then began audio taping the narration. The boy told the story starting with the first page and proceeding one page at a time until the completion of the book.

After the boy told the story, the research assistant turned back to the beginning of the picture book and gave the following instructions for the second narration: "I want you to go back through the book and tell the story one more time. You can tell me the exact same story you just told, or you can change part of the story, or you can tell me a completely different story." Once again the boy was audio taped while retelling the story.

It is important to note that, prior to the first telling, boys were given no opportunity to look through the picture book, as is more commonly the case in studies employing this methodology (Trabasso & Nickels, 1992). We instituted this change because a major purpose of the present study was to examine the on-line comprehension processes the two groups of boys employ as they encounter story information and build a story representation. In contrast, the second production allows assessment of how boys use information gained from exposure to the pictures to plan their second narrations based on modifications of their story representations.

After completion of the two story narrations, all boys were administered the Block Design and Vocabulary subtests from the WISC-1II (Wechsler, 1991) and the Letter-Word Identification and Passage Comprehension subtests of the Woodcock-Johnson Tests of Achievement--Revised (Woodcock & Johnson, 1989). As indicated earlier, these subtests were administered in order to obtain estimates of intelligence and reading achievement.

Measures

The narrations were transcribed verbatim from the audio tapes. Each transcribed narration was parsed into units, each of which conveyed a single idea. Coding of units followed the procedures described by Trabasso and Nickels (1992), which are based on story grammar theory (Stein & Glenn, 1979) and the causal network model (Trabasso et al., 1989). According to story grammar theory, the episodic structure defining a typical story is: Setting [right arrow] (Initiating Event [right arrow] Internal Response [right arrow] Goal [right arrow] Attempt [right arrow] Outcome; Trabasso & Stein, 1997). This structure indicates specific categories of events and implies the relations among them.

Story Opening

Settings provide information about the characters in the story, and the time and place the story takes place. For example, if a child says, "There was a boy who had a dog and a frog and they are in the boy's bedroom," this statement contains information about each of three characters and their location. Once the setting is established, an episode begins with an Initiating Event that affects a main character. For example, "The boy sees that the frog is gone from his jar." This initiating event leads to some Internal Response (4) by the main character; for example, "The boy gets upset."

Goal-Based Categories

The initiating event and the character's responses to it give rise to an Overall Goal, for example, "The boy wants to bring the frog back home." Typically, the overall goal motivates a goal plan that contains subgoals, a series of attempts to achieve each goal, and the outcome of each attempt. In the case of "Frog, where are you?" a negative outcome (i.e., a failure to find the frog) reinstates the overall goal and motivates new subgoals (i.e., a search for the frog in a new location). Thus, the story is comprised of a number of Goal-Attempt-Outcome sequences until the overall goal is achieved.

The centrality of Goal-Attempt-Outcome sequences in a coherent representation of the story necessitates more detailed coding of the elements of these sequences. For example, children may make reference to different levels of goals, vary in how they link particular attempts to goals, and differ in whether they state explicit outcomes of attempts. Specifically, some attempts are linked with the subgoal of finding the frog (Attempts Linked to Goal; e.g., "The boy was calling for his frog"). Other attempts are linked to a more specific subgoal. For example, attempts by the protagonist to find the frog in specific locations (Attempts Linked to Goal and Location; e.g., "The boy looked inside the tree for the frog"). Outcomes result from attempts to find the missing frog. Failed attempts to find the frog result in Negative Outcomes (e.g., "The boy looks in his boot but the frog's not there"). A Positive Outcome can occur only once in the story, when the protagonist finds the frog. The resolution of the Overall Goal is achieved only when the frog is brought back home.

Other Story Grammar Categories

The story grammar categories already described capture the central storyline of losing the frog, searching for the frog, and finding the frog and bringing it home. In addition, other story grammar categories occur routinely in the story but are not part of the central storyline. Events are actions that are represented in the pictures but are not directly involved in the central storyline (e.g., "The deer picked up the boy with his antlers"), Unlinked Attempts are not tied explicitly to goals (e.g., "The boy was calling"), and Neutral Outcomes occur at the end of the story as a result of the protagonist finding the frog (e.g., "The two frogs had made babies").

Errors and Adjustments

In addition to the different categories of story units, three types of errors, adapted from Capps et al. (2000), were coded. Within-Clause Errors occurred when part of the clause was unclear or ambiguous (e.g., "They looked in the hole in the ground" when the picture shows only the boy looking in the hole). Whole-Clause Errors occurred when the child said something that did not happen (e.g., "The dog ate the frog"). Repetition Errors occurred when the child repeated a previously stated idea that was not represented in the current picture. In addition, one specific variable was coded to examine how boys adjust their existing representations once they have seen all of the pictures in the story book. In one picture, the protagonist is holding on to what seems to be the branches of a tree. The next picture reveals that these tree branches are actually the antlers of a deer. The boys were given credit for this item (Antler/Branch Adjustment) only if they mentioned both the protagonist's misunderstanding and correc tion of this false belief.

Coding

Each narrative protocol was independently coded by a graduate student and an undergraduate honors student. The graduate student was not blind to group status but the undergraduate student was unaware that boys with ADHD were participants in the study. The two coders were trained by coding practice protocols from pilot participants and reviewing and clarifying the coding categories with the investigators. Interrater reliability (i.e., Pearson correlations), calculated on the independently coded study protocols, averaged .80 across all coding categories.

RESULTS

As noted earlier, comparison boys scored significantly higher than the boys with ADHD on both subtests of the Woodcock-Johnson, as well as on the Vocabulary subtest from the WISC-III. Because the two subtests from the Woodcock-Johnson were highly correlated (r = .70, p < .01), a mean score was calculated for each child and used as a measure of comprehension ability. When examined separately for each telling and group, only one of the dependent variables was significantly correlated with the composite Woodcock-Johnson score at p < .01. Similarly, only one of the dependent variables was correlated significantly with the WISC-III Vocabulary subtest score. Therefore, the scores on the Woodcock-Johnson and on the Vocabulary subtest were not included in subsequent analyses.

Primary analyses were conducted using a 2 (Group: comparison, ADHD) x 2 (Telling: first vs. second) mixed analysis of variance (ANOVA). (5) Before examining production of individual story grammar categories, the total number of idea units produced by the boys was compared. There was no significant difference between the comparison boys (M = 45.73) and the boys with ADHD (M = 42.53) in the number of idea units produced, F(1, 64) = 1.03, p > .05, nor was there a main effect for telling, F(1, 64) 3.37, p > .05, or a telling by group interaction, F(1, 64) < 1. Because the groups did not differ in the total number of units produced, subsequent analyses utilized simple frequencies of occurrence. Means and standard deviations for each category as a function of group and telling are presented in Table II.

Story Opening

There were no main effects for group or telling, or a group x telling interaction for either Settings or Initiating Events.

Goal-Based Categories

One purpose of the present study was to examine whether children with ADHD are particularly deficient in representing goal-related events. In a coherent story representation, one of the most important events is completion of the overall goal. There was a significant group difference for this variable, F(l, 64) = 13.06, p < .001, such that 94.4% of comparison boys but only 65.9% of boys with ADHD included the completion of the overall goal across both tellings. More boys included the completion of the overall goal in the second telling (88%) than in the first telling (81%), F(l, 64) = 4.27, p < .05, but the group x telling interaction was not significant, F(1, 64)= 2.l8, p > .05.

Given that a coherent story representation is structured around the goal of finding the frog and bringing it home, it is important to examine the boys' use of goal-based attempts in their narrations. Two types of goal-based attempts were coded, attempts linked to the goal of finding the frog and attempts linked to the goal and specific search locations. For Attempts Linked to Goal, there was no significant group difference or group x telling interaction, F(1, 64) < 1, but boys made more of these statements in the first (M = 1.47) than in the second (M = 1.00) narration, F(1, 64) = 6.45, p < .05. In contrast, for the more specific Attempts Linked to Goal and Location, comparison boys (M = 5.56) mentioned more of these attempts than the boys with ADHD (M = 4.23), F(1, 64) = 12.72, p <.001. Although there was no main effect of telling, F(1, 64) = 1.11, p > .05, there was a marginally significant telling x group interaction, F(1, 64) = 3.44, p = .07, with the difference between groups tending to be more pronounce d for the first telling than for the second.

There were no main effects for group or telling, or a group x telling interaction for either Positive or Negative Outcomes.

Other Story Grammar Categories

There was a main effect of group for the number of Neutral Outcomes included in narrations, F(1, 64) = 4.96, p < .05, with comparison boys (M = 1.64) including more of these outcomes than boys with ADHD (M = 1.27). There were no main effects for group or group x telling interactions for the variables of Events or Unlinked Attempts. Boys included more Unlinked Attempts in the first telling (M = 3.68) than in the second telling (M = 2.97), F(1, 64) = 6.84, p < .01, but there were no other significant main effects for telling.

Errors and Adjustments

A second purpose of the present study was to examine whether boys with ADHD are more likely to make specific types of errors that interfere with the coherence of their story narrations. Misinterpretation of parts of the story, or an inability to recognize the importance of verbalizing events in the story, can lead to errors that create a less cohesive narration. There was a main effect for group in the number of Within-Clause Errors (which includes reference errors, and unclear or ambiguous parts of a clause), F(1, 64) = 19.8, p < .001. Boys with ADHD (M = 2.55) committed more of these errors than did comparison boys (M = 0.90). There was no main effect for telling, F(1, 64) = 2.15, nor was there a telling x group interaction, F(1, 64) < 1. There were no effects for Whole-Clause Errors (all Fs < 1).

Another type of error was a Repetition Error, in which boys repeated events that were not represented in the current picture. Boys with ADHD made more Repetition Errors than comparison boys, F(1, 64) = 4.47, p < .05, and all boys tended to make more of these errors in the first telling than in the second telling, F(1, 64) = 3.89, p < .10. These effects were qualified by a group x telling interaction, F(1, 64) = 4.99, p < .05. Boys with ADHD (M 0.50) made more Repetition Errors than comparison boys (M = 0.11) in the first telling, F(1, 64) = 7.03, p < .01. However, boys with ADHD corrected this type of error during the second telling (M = 0.14) and performed at the same level as comparison boys (M = 0.14), F(1, 64) < 1.

The final variable included in the analyses examined how boys adjusted their representations after building the initial story representation. This variable (Antler/Branch Adjustment) reflected whether the boys mentioned both the protaganist's initial misunderstanding and subsequent corrected belief regarding the deer antlers. There were no significant main effects, but there was a significant telling x group interaction, F(1, 64) 8.67, p < .01. In the first telling, 13.6% of the comparison boys and 18.2% of the boys with ADHD mentioned this discrepancy in their narration. During the second telling, 41.9% of the comparison boys mentioned the discrepancy whereas only 9.0% of the boys with ADHD did so.

DISCUSSION

The findings of the present study add to the burgeoning literature on story comprehension in ADHD by identifying the difficulties these children experience both in building an on-line story representation and in modifying an existing representation to create a more cohesive narration. There were several noteworthy results. First, the narrations of boys with ADHD differed from those of comparison boys in important ways that pertained to the goal structure of the story. Fewer boys with ADHD included the attainment of the overall goal in their narrations, the sine qua non of a complete story representation. In addition, boys with ADHD included fewer attempts linked to the goal and specific locations than did comparison boys, indicating a less sustained use of the goal plan throughout their narrations. Second, the coherence of the narrations of boys with ADUD was marred by a higher rate of errors than for comparison boys. Third, for one specific measure, boys with ADHD were less likely to adjust their story repre sentations based on information acquired during the first narration. Finally, the differences between the two groups reflect specific difficulties in representing story structure among the boys with ADHD and not a deficit in their overall story production.

As noted above, the stories produced by the boys with ADHD were deficient in two essential story grammar categories, completion of the overall goal and attempts linked to the goal and specific locations. Stories that do not include the completion of the overall goal of finding the frog and bringing it back home fail to establish a critical link between the protagonist's initial purpose and the final outcome of the story. Similarly, narrations that provide fewer attempts linked to the goal and specific locations lack cohesion because they do not specify connections between the protagonist's actions and the goal plan directing these actions. These group differences become more important when developmental trends for the use of goals in story narrations are examined. Developmental trends indicate that very young children (e.g., 4-year-olds) describe the concrete actions and objects in the pictures but do not link actions with goals or consequences. By 7-8 years of age children include some goal-based information in their narrations but do not systematically link the actions to the overall goal plan. By 9-11 years of age, children are able to produce coherent story representations by consistently reinstating the goal plan to tie together actions and events in the story (Trabasso & Stein, 1997). In the present study, the nonreferred boys produced stories with the expected emphasis on goals and the goal structure, whereas the boys with ADHD resembled younger children in their less consistent use of the goal plan. This difference was especially evident in that nearly all of the nonreferred boys included the completion of the overall goal in both of their narrations, whereas one third of the boys with ADHD failed to do so.

What might account for these group differences in the use of goals in story representations? First, it may be that children with ADHD are less likely to understand that the goal structure is the most appropriate organizing framework for connecting events into a coherent story. Goals are, by definition, internal constructs and so are more abstract than concretely depicted actions. This is a major reason why we see developmental trends in both the frequency and effectiveness with which children use goals to structure their narrations (Trabasso & Stein, 1997). The 9-11-year-old boys with ADHD in the present study produced narrations suggestive of a developmental delay. Like typical 7-8-year-olds, they connected some of the actions to the goal plan but they did not do so in a systematic or consistent fashion. Similarly, their more frequent failure to complete the goal plan suggests a lack of understanding of the importance of goal completion to a coherent story. In contrast, as expected by 9-11 years of age, the nonreferred boys clearly understood that a coherent story requires the completion of the overall goal.

Working memory deficits and problems in executive functioning are two features associated with ADHD that may contribute to the difficulties these children have in using a sustained goal plan. First, recent studies have documented working memory deficits among children with ADHD (Barkley, 1997). Such deficits may help account for failures to report completion of the overall goal of finding the frog and bringing it home, and for less consistent connections between the protagonist's actions and the goal plan. The boys with ADHD may have difficulty maintaining activation of the overall goal in working memory as they encounter new information, making it less likely that they will include completion of the overall goal in their narrations. This also may help explain why the boys with ADHD included fewer attempts linked to the goal and specific locations in their narrations; they may have difficulty maintaining simultaneous activation of both the overall goal and the connection between a recent failed attempt and th e protagonist's initiation of a new attempt.

Working memory difficulties in the boys with ADHD also may help to explain their higher rate of repetition errors during the first telling. The first time the boys tell the story they are unaware of upcoming events and so are in the process of forming their story representations. If the boys with ADHD have particular difficulty maintaining the goal plan in working memory during this first telling, they may repeat events from previous pictures to try to establish links with new information. One possible mechanism accounting for this pattern may stem from the inhibitory problems observed in boys with ADHD (Barkley, 1997). Their inhibitory problems may impair their on-line ability to suppress the activation of information that is no longer relevant for establishing a connection, thereby increasing the demands on working memory (Gernsbacher, 1997). In contrast, boys with ADHD did not make more repetition errors than comparison boys during the second telling. Having been through the story once, their knowledge of story events may enable the boys with ADHD to rely less on repetition to remind them of connections among events.

Problems in executive functioning also may contribute to the difficulties boys with ADHD have in the sustained use of a goal plan during story narration. Executive functions are self-regulatory processes that are responsible for the planning and organizing of goal-directed behavior, including mobilizing attention, monitoring incoming information, inhibiting responding, and modifying behavior as necessary. Children with ADHD have been found to show deficits in all of these components of executive functioning (Barkley, 1997). These documented deficits in executive functioning suggest that children with ADHD will have difficulty using goals to organize and guide their behavior sequences in many of their activities. The present results indicate that this difficulty generalizes to the use of goal plans in building a story representation. What is unique about the present results is that the boys' explicit statements of goal-based information were specifically measured. In contrast, for many executive function tasks , the use of goals must be inferred from performance (e.g., the number of blind alleys entered on a maze task).

Deficits in executive functioning also may account for the increased rate of within-clause errors among boys with ADHD. These errors include ambiguities or misinterpretations in the description of story events (e.g., the child says, "they looked for the frog in the tree" when in fact only the boy was looking). These errors may be due to problems in response inhibition, in that the boys with ADHD do not reflect carefully on all of the details of each picture before continuing their narrations. Another possibility is that the boys with ADHD have difficulty changing response sets during their narrations. For example, if a boy had been talking about "the boy and his dog," he may continue to use the pronoun "they" even though the picture indicates that only one character is performing a particular action. Problems in response inhibition and response perseveration are well documented in studies of children with ADHD (Barkley, 1997; Douglas, 1999). Either or both of these problems may be impairing the boys' ability to incorporate important details into an accurate and coherent narration. These errors clearly impair the boys' ability to communicate the story to a listener, but it is unclear whether the errors reflect deficits in the boys' understanding of the story.

Effective story comprehension requires the ability to use new information to modify the developing representation (Zwaan, Magliano, & Graesser, 1995). In the Frog, where are you? story there is only one instance where misleading information in one picture (i.e., apparent tree branches) makes it necessary for the boys to modify their representations when they encounter disambiguating information in a subsequent picture (i.e., the "branches" are revealed to be deer antlers). Very few boys with ADHD (i.e., 9%) used the information acquired during the first narration to adjust their story representations during the second telling. Comparison boys also did not include this detail on the first telling given the ambiguity of the picture, but once they had gained the knowledge of this event from subsequent pictures, many of them (i.e., 42%) used this knowledge in their second narration to tie the events of the story together. The failure of the boys with ADHD to revise the story representation to accommodate new info rmation may be due to working memory deficits. In order to accomplish such a revision, children must be capable of holding and manipulating multiple pieces of information in working memory. In the present case, children must hold onto the information that the boy grabbed something that appeared to be tree branches, and then must revise their representations to make the connection that the branches actually are deer antlers. It is possible that boys who did not mention this connection nevertheless did see the link. If so, however, their failure to incorporate the connection into their narrations reduced the coherence of these narrations.

It is important to note that the group differences obtained in this study reflect specific difficulties in representing story structure among the boys with ADHD and not a general deficit in story production. Both groups produced approximately the same number of total units, and similar numbers of Setting statements, Unlinked Attempts, and Event statements. This indicates that boys with ADHD have the general skills necessary to produce parts of the story that are appropriate to each picture, but their specific deficits interfere with their representation and use of the goal structure to narrate a coherent story.

Limitations and Future Directions

The present study examined the on-line narrative abilities of a specific group, boys with the combined type of ADHD, noted to have significant academic difficulties. Given that other groups of children have significant problems in school performance (e.g., learning disabled, the ADHD/predominantly inattentive group), valuable information about the common and unique cognitive processing difficulties among these groups can be gained by analyzing and comparing the narrations produced by these different groups. Similarly, given evidence that there are significant information processing differences between boys and girls with these disorders (e.g., Gaub & Carlson, 1997), future studies can investigate whether group differences in narrative abilities are moderated by gender.

The wordless picture book used in the present study contained 24 pictures that represent a logically sequenced hierarchical goal plan and is structured so that a narrator cannot stray far from the story's intended interpretation. Using a more complex story with episodes that contain actions and events that are off the causal chain would increase the demands on the narrator, and may provide a stronger test of children's ability to sustain the goal plan in their narrations. A complex story structure also would increase working memory demands and may tax other executive processes, including self-directed effort, planning, and organization.

As noted above, a limitation of the story narration methodology is the difficulty in differentiating deficiencies in communication from errors in understanding. One strategy for clarifying the specific difficulties demonstrated by children with ADHD would be to include questions designed to test comprehension of story events and connections among them. Such an approach would help distinguish how children represent the story from what they communicate to a listener. Given that it is unlikely the boys in the study perceived the listener as naive to the story, it is possible that the two groups differed in how expectations about the listener affected their narrations. For example, because the listener was perceived as informed, boys with ADHD may have expended less effort than comparison boys in supplying sufficient details. However, given that the lengths of the narrations produced by the two groups were comparable, this issue is unlikely to have had a major impact on group differences.

In discussing our findings we have speculated on the underlying mechanisms that may account for the difficulties that boys with ADHD experienced when building an on-line story representation. Future studies could include measures of executive functioning, working memory, and inhibitory control to examine what role they may play in accounting for specific story narration deficits shown by children with ADHD.

As noted earlier, there is a substantial line of research examining developmental changes in children's online story representation abilities. Although we previously speculated that children with ADHD may exhibit a developmental delay in their use of a goal plan, few studies have compared developmental changes in on-line story representation for nonreferred children and children with ADHD. It is necessary to investigate both the developmental precursors to the abilities shown by boys with ADHD in the present study, as well as the long-term implications of their deficits. A thorough developmental investigation of how children build story representations may further our understanding of the academic problems experienced by children with ADHD. As children advance through the school years, academic tasks increasingly require both the production and appreciation of story structure. If we are fully to understand and ameliorate the academic difficulties experienced by children with ADHD, we need to continue investig ating the development of their story comprehension abilities.

ACKNOWLEDGMENT

This research was supported by National Institute of Mental Health Grant MH47386 to the second author.

Received March 14, 2002; revision received August 21, 2002; accepted August 29, 2002

(4.) Two categories, one dealing with the protagonist's internal response to the initiating event and one with the internal responses to other events, were initially coded but failed to reach acceptable levels of interrater reliability due to low occurrence. Therefore, these two categories are not discussed further.

(5.) Although a few of the dependent variables were dichotomous, analyses of variance were used throughout because they are robust to violations of normality given sample sizes of the current study and they facilitate interpretation of interactions involving the repeated-measures variable (Myers & Well, 1991).

REFERENCES

Abikoff, H., Gittelman-Klien, R., & Klein, D. (1977). Validation of a classroom observation code for hyperactive children. Journal of Consulting and Clinical Psychology, 45, 772-783.

Achenbach, T. M. (1991). Manual for the Child Behavior Checklist/4-18 and 1991 Profile. Burlington, VT: University of Vermont, Department of Psychiatry.

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.

Atkins, M. S., Pelham, W. E., & Licht, M. H. (1985). A comparison of objective classroom measures and teacher ratings of attention deficit disorder. Journal of Abnormal Child Psychology, 13, 155-167.

Barkley, R. A. (1990). Attention deficit hyperactivity disorder: A handbook for diagnosis and treatment. New York: Guilford Press.

Barkley, R. A. (1997). ADHD and the nature of self-control. New York: Guilford Press.

Berman, R. A., & Slobin, D. I. (1994). Narrative structure. In D. I. Slobin & R. A. Berman (Eds.), Different ways of relating events in a narrative: A cross-linguistic study (pp. 39-84). Hillsdale, NJ: Erlbaum.

Capps, L., Losh, M., & Thurber, C. (2000). "The frog ate the bug and made his mouth sad": Narrative competence in children with autism. Journal of Abnormal Child Psychology, 28, 193-204.

Douglas, V. I. (1999). Cognitive control processes in attention-deficit/hyperactivity disorder. In H. C. Quay & A. E. Hogan (Eds.), Handbook of disruptive behavior disorders (pp. 105-138). New York: Kluwer Academic/Plenum.

Gaub, M., & Carlson, C. L. (1997). Gender differences in ADHD: A meta-analysis and critical review. Journal of the American Academy of Child and Adolescent Psychiatry, 36, 1036-1045.

Gemsbacher, M. A. (1997). Two decades of structure building. Discourse Processes, 23, 265-304.

Goldman, S., & Varnhagen, C. (1986). Memory for embedded and sequential story structures. Journal of Memory and Language, 25, 401-418.

Hinshaw, S. P. (1994). Attention deficits and hyperactivity in children. Thousand Oaks, CA: Sage.

Lorch, E. P., Diener, M. B., Sanchez, R. P., Milich, R., Welsh, R., & van den Broek, P. (1999). The effects of story structure on the recall of children with attention deficit hyperactivity disorder. Journal of Educational Psychology, 91, 273-283.

Lorch, E. P., Sanchez, R. P., van den Broek, P. W., Milich, R., Murphy, E. L., Lorch, R. F., Jr., et al. (1999). The relation of story structure properties to recall of television stories in young children with attention deficit hyperactivity disorder and nonreferred peers. Journal of Abnormal Child Psychology, 27, 293-309.

Loveland, K., McEvoy, R., Tunali, B., & Kelley, M. L. (1990). Narrative story-telling in autism and Down syndrome. British Journal of Developmental Psychology, 8, 9-23.

Low, J., & Durkin, K. (1998). Structure and causal connections in children's online television narratives: What develops? Cognitive Development, 70, 413-431.

Mandler, J. M., & Johnson, N. S. (1977). Remembrance of things parsed: Story structure and recall. Cognitive Psychology, 9, 111-151.

Mayer, M. (1969). Frog, where are you? New York: Dial.

Milch-Reich, S., Campbell, S. B., Pelham, W. E., Jr., Connelly, L. M., & Geva, D. (1999). Developmental and individual differences in children's on-line representations of dynamic social events. Child Development, 70, 413-431.

Milich, R., Balentine, A., & Lynam, D. R. (2001). ADHD combined type and inattentive type are distinct and unrelated disorders. Clinical Psychology: Science and Practice, 8, 463-488.

Myers, J. L., & Well, A. D. (1991). Research design and statistical analysis. New York: HarperCollins.

Nezworski, R., Stein, N. L., & Trabasso, T. (1982). Story structure vs. content in children's recall. Journal of Verbal Learning and Verbal Behavior, 21, 196-206.

O'Brien, E. J., & Myers, J. L. (1987). The role of causal connections in the retrieval of text. Memory and Cognition, 15, 419-427.

O'Neill, M. E., & Douglas, V. I. (1991). Study strategies and story recall in attention deficit disorder and reading disability. Journal of Abnormal Child Psychology, 19, 671-692.

Purvis, K. L., & Tannock, R. (1997). Language abilities in children with attention deficit hyperactivity disorder, reading disabilities, and normal controls. Journal of Abnormal Child Psychology, 25, 133-144.

Sattler, J. M. (2001). Assessment of children: Cognitive applications. San Diego, CA: Jerome M. Sattler Publisher.

Stein, N. L., & Glenn, C. G. (1979). An analysis of story comprehension in elementary school children. In R. D. Freedle (Ed.), Advances in discourse processes: Vol. 2. New directions in discourse processing (pp. 53-119). Norwood, NJ: Albex.

Tager-Flusberg, H., & Sullivan, K. (1995). Predicting and explaining behavior: A comparison of autistic, mentally retarded, and normal children. Journal of Child Psychology and Psychiatry, 35, 1059-1075.

Tannock, R., Oram, J., & Fine, J. (1997). Tell me a story: Narratives of children with ADHD with and without concurrent language impairment. Poster presented at the annual meeting of the International Society for Research in Child and Adolescent Psychopathology, Paris.

Tannock, R., Purvis, K. L., & Schachar, R. J. (1993). Narrative abilities if children with attention hyperactivity disorder and normal peers. Journal of Abnormal Child Psychology, 21, 103-117.

Trabasso, T., & Nickels, M. (1992). The development of goal plans of action in the narration of a picture story. Discourse Processes, 15, 249-275.

Trabasso, T., & Rodkin, P. C. (1994). Knowledge of goal/plans: A conceptual basis for narrating Frog, Where are You? In R. A. Berman & D. I. Slobin (Eds.), Relating events in narrative: A cross-linguistic developmental study (pp. 85-106). Hillsdale, NJ: Erlbaum.

Trabasso, T., Secco, T., & van den Broek, P. W. (1984). Causal cohesion and story coherence. In H. Mandler, N. L. Stein, & T. Trabasso (Eds.), Learning and comprehension of text (pp. 83-111). Hillsdale, NJ: Erlbaum.

Trabasso, T., & Stein, N. L. (1997). Narrating, representing and remembering event sequences. In P. W. van den Broek, P. J. Bauer, & T. Bourg (Eds.), Developmental spans in event comprehension and representation: Bridging fictional and actual events. (pp. 237-270). Mahwah, NJ: Erlbaum.

Trabasso, T., Stein, N. L., Rodkin, P. C., Munger, G. P., & Baughn, C. (1992). Knowledge of goals and plans in the on-line narration of events. Cognitive Development, 7, 133-170.

Trabasso, T., & van den Broek, P. (1985). Causal thinking and the representation of narrative events. Journal of Memory and Language, 24, 612-630.

Trabasso, T., van den Broek, P., & Suh, S. Y. (1989). Logical necessity and transitivity of causal relations in stories. Discourse Processes, 21, 1-25.

van den Broek, P. W. (1989). Causal reasoning and inference making in judging the importance of story statements. Child Development, 60, 286-297.

van den Broek, P. W. (1990). Causal inferences and the comprehension of narrative texts. The Psychology of Learning and Motivation, 25, 175-196.

van den Broek, P. W. (1997). Discovering the cement of the universe: The development of event comprehension from childhood to adulthood. In P. W. van den Broek, P. J. Bauer, & T. Bourg (Eds.), Developmental spans in event comprehension and representation: Bridging fictional and actual events (pp. 321-342). Mahwah, NJ: Erlbaum.

van den Broek, P. W., Lorch, E. P., & Thurlow, R. (1996). Children's and adults' memory for television stories: The role of causal factors, story-grammar categories and hierarchical level. Child Development, 67, 3010-3028.

Wagner, C. R., Sahlen, B., & Nettelbladt, U. (1999). What's the story? Narrative and comprehension in Swedish preschool children with language impairment. Child Language, Teaching, and Therapy, 15, 113-137.

Wechsler, D. (1991). Wechsler intelligence scale for children (3rd ed.). San Antonio, TX: Psychological Corporation.

Woodcock, R. W., & Johnson, M. B. (1989). Woodcock-Johnson Psycho-Educational Battery--Revised. Chicago: Riverside.

Zentall, S. S. (1988). Production deficiencies in elicited language but not in the spontaneous verbalization of hyperactive children. Journal of Abnormal Child Psychology, 16, 657-673.

Zwaan, R. A., Magliano, J. P., & Graesser, A. C. (1995). Dimensions of situation model construction in narrative comprehension. Journal of Experimental Psychology: Learning, Memory, and Cognition, 21, 386-397.

Kelly Renz (1), Elizabeth Pugzles Lorch (1,3), Richard Milich (1), Clarese Lemberger (1), Anna Bodner (1), and Richard Welsh (2)

(1.) Department of Psychology, University of Kentucky, Lexington, Kentucky.

(2.) Department of Psychiatry, University of Kentucky, Lexington, Kentucky.

(3.) Address all correspondence to Dr. Elizabeth Pugzles Lorch, Department of Psychology, University of Kentucky, Lexington, Kentucky 40506-0044; e-mail: elorch@uky.edu.

COPYRIGHT 2003 Plenum Publishing Corporation
COPYRIGHT 2003 Gale Group