Abstract
Children’s skills in telling narratives are positively related to language (e.g., Cooper et al., 1992; Mallan, 1991), to reading and writing skills (e.g., Dickinson & Tabors, 2002; Reese et al., 2010), and to children’s overall school performance (Snow et al., 1998). Moreover, quality of narratives told at 5 years of age (Griffin et al., 2004), or retold at 6 years of age (Reese et al., 2010), has been shown to contribute to reading fluency and comprehension at 8 years of age.
Although children are capable of explaining events and behavior in their everyday interactions (Hickling & Wellman, 2001; Veneziano, 2001) and may succeed in first order false belief tasks at 4 to 5 years (Wellman et al., 2001), explicit reference to mental or epistemic states in self-constructed narratives is rarely observed before 6 to 7 years (Peterson & Slaughter, 2006). Until 6 to 7 years of age, children provide few explanations and evaluations of events, and it is not until 8 to 9 years that the majority of children express these elements (e.g., Berman, 2004, 2009; Berman & Slobin, 1994). From the age of 8 to 9 years, children begin to spontaneously refer to the characters’ internal states to explain their behaviors and the story events (Bamberg, 1994), and it is even later that they express false beliefs that require the characters’ beliefs to be clearly differentiated from the state of the world or to talk about the different points of view of the characters about the same event (Kielar-Turska, 1999; Küntay & Nakamura, 2004; Veneziano & Hudelot, 2009).
Given the centrality of narrative skills in children’s development, many studies have focused on promoting children’s narrative skills using different methodologies, settings, and durations of the intervention. In this study, the authors used an intervention they titled short conversational intervention (SCI). The SCI was chosen because, besides being short and simple to administer, it has three features likely to promote the inferential content of children’s narratives. First, the SCI focuses children’s attention on the causal structure of the story, stimulating them to go beyond the perceptual details present in the pictures. Second, by focusing on pictures one at the time, the SCI lets children segment the story structure into its components, thereby diminishing the cognitive load required to process and integrate all the various elements at the same time. Finally, by just asking questions without providing answers or referring to the characters’ mental states, the SCI is minimally intrusive so that it allows children to find new interpretations and attain new understandings by themselves.
Purpose of Study
This study had three aims:
The main aim of this study was to explore whether the causal and mind-oriented content of child-constructed monological narratives is strengthened after a short conversational intervention that focuses children’s attention on the causes of the events and of the characters’ behaviors
Second, this study aimed to determine whether a short conversational intervention could influence the ability of 5- to 8-year-old children to include more clausal links and references to characters’ mental states in their stories.
The third aim was to determine whether improvements obtained immediately after the SCI persisted over time when the children retold the same story and when they told a different story.
Finally, the researchers also wanted to determine whether any improvements obtained were specific to the SCI or could also be obtained by the retelling of the story and/or by the child becoming more at ease with the experimenter and the situation. To this effect, children in the SCI condition were compared with a control group who, after the initially produced narrative, played a memory game with the experimenter using cards that included the story pictures instead of participating in the SCI procedure. Moreover, in this study the authors also analyzed two aspects of the narratives’ verbal expression: the production of causal markers and the length of the narratives in words
One hundred fourteen typically developing French-speaking children between 5 years 6 months and 8 years 8 months participated in the study. Children attended kindergarten, first grade, and second grade, with 38 children per grade or age group. In each grade/age group, 28 children participated in the SCI group and 10 children in the control group (CTR).
Procedures
Procedures for the Experimental Group
The “Stone” story consists of a sequence of five wordless pictures meant to represent the story of a mis-understanding between two characters. The story was chosen for its potential to elicit expression of the false belief of one of the characters and of the two characters’ different representations of the same event. The first picture shows two boys (referred to hereafter as P1 and P2) waving hello to each other from a distance (the “greeting”). The second and third pictures represent the “complicating event”: P1 accidently stumbles on a stone, leading to the first push of P2 by P1, followed by P2 pushing back P1. The fourth picture represents the “attempt at resolution,” showing P1 crying and pointing toward the stone. The fifth picture represents “the resolution,” showing P2 helping P1 to get up. A likely interpretation of the sequence of pictures is that P2 interprets the physically caused push by P1 (determined by his stumbling on the stone) as being intentional—the source of the misunderstanding by P2 of P1’s behavior—and reacts by pushing P1 back. P1, who understands that P2 has misunderstood his behavior, tells him the real reason for the initial push and thus, after the clarification, the two characters make peace.
After the child told the story, the experimenter conducted a SCI about the causes in the story, asking the child questions that focused his or her attention on the reasons for the characters’ behaviors relating to each of the four key events in the story: “the first push,” “the second push,” “pointing towards the stone,” and “the reconciliation.” The experimenter asked, “Have you understood why: P1 pushes P2? [the researcher referred to the characters in the story in the same way as the child had done]; P2 pushes P1?; P1 shows the stone?; and P2 helps P1 to get up?” If the child did not respond or said she or he did not know, the experimenter first asked the question again and, in the case of no response, later added, for example, for the first push, something like, “Do you push your friend when you first see him?” The experimenter did not provide answers or relevant information that could help the child find the answer or make any reference to the characters’ internal states. If, in the first narrative, the child had provided an explanation for the event questioned, the experimenter acknowledged that the child had already provided an explanation and asked the child to tell the reason again.
After the SCI, the experimenter asked the child to tell the story once again. This is the second narrative (NAR2). One week later, children were again presented with the pictures of the Stone story and, following the same procedure as that used for the first two narratives, were asked to tell the story. This is the third narrative (NAR3). Then, still following the same procedure, children were shown the five wordless pictures of the “Bicycle” story and asked to tell the story when ready. Like the Stone story, the Bicycle story consisted of five wordless pictures that could be interpreted as a story of a misunderstanding between two characters. Here, the “first push” has a two-step physical cause: The nails on the road puncture the tire of P1’s bicycle, leading P1 to lose control of the bicycle and hit P2. This was the fourth narrative (NAR4).
The Procedure for the Control Group
A control group of children followed the same overall procedure as that described above with two differences. Between the first and the second narratives, the children, instead of participating in the SCI, played a Memory game with nine pairs of pictures (the five pictures of the Stone story and four additional pairs of pictures constructed by some transformations of the original ones). In this way, the children could see the story pictures again (as did the children in the SCI group) without, however, being solicited to think about the causes of the story events. The second difference is that, 1 week later, these children were asked to tell only the Stone story. Because these children did not have the SCI, there was no point in testing what the children would do on a new story.
Assessment Measures
Measures of Inferential Content
Explanation of events were coded when
it was explicitly marked by a causal connector or expression such as because, therefore, to, so as to;
the relationship was not explicitly marked but the explanatory interpretation dominated over an interpretation in purely temporal terms, for example; the relationship was presented retroactively, from the event to its cause (e.g., “He fell, there was a pebble, he hadn’t seen the pebble”); the relationship was presented proactively (from the cause to the event);
one of the elements of the relationship was a nonperceptible aspect introduced by the child (such as an internal state);
the elements of the relationship were meaningfully and inherently causally linked one to the other (for example, “He stumbles on a pebble and he pushes the girl”).
Measures of Attribution of Internal States
All the internal states attributed to the characters were identified and coded as “physical” (e.g., “He hadn’t seen the stone”), “emotional” (e.g., “The other [child] was angry”), “intentional” (e.g., “The other boy pushed him on purpose”), and “epistemic”—the children mention that a character knows, believes, thinks or understands something about a state of the world (first order belief, e.g., “He understood that there was the stone”) or about someone’s mental state (second order belief, e.g., “He believed he wanted to push him”). Internal states were also coded for whether they had an explanatory function (e.g., “He cries because he’s in pain”; “He pushes the other back because he had not understood why he had first pushed him”). Internal states used to explain behaviors are inferential both as internal states and as explanations and occur later in development than attributions.
The researchers identified two further epistemic states that play a central role in the comprehension of the story as a story of a misunderstanding and require a more complex expression than just mentioning that a character knows, believes, or thinks something.
The expression of the false belief (FB). The criteria for coding the false belief required the expression of both the physical and/or unintentional cause of the first push by P1 and the attribution to P2 of the belief that P1 had pushed P2 on purpose (e.g., “Someone fell on a stone and pushed his pal; the other believed he had done it on purpose”). As the belief of P2 about P1’s intentions is accompanied by the expression of the real cause of the first push, the belief attributed to P2 is a false belief.
The rectification of the false belief (RFB): The rectification of the false belief (RFB) was coded when children had P1 explain to P2 that his initial push was not intentional but was caused by the stone (e.g., “And afterwards he explained to him that he pushed him because of the stone”).
Language Measures
Markers of Causality
All causal markers present in the explanations were coded, for example, “because,” “so,” “then,” or expressions such as “because of,” “so that,” or “given that.”
Length of Narratives
Narrative length was measured by the number of words that children produced in narrating the story. The researchers did not count repetitions, false starts, metadiscursive comments (e.g., “I don’t remember what this is called”), comments that nullified what had been just said, or fillers, such as mmm, uh.
Results
Narrative produced immediately after the SCI (NAR2) contained more inferential and misunderstanding relevant content than the initially produced narrative (NAR1).
The narrative of the Stone story produced 1 week later (NAR3) and of the new story, also produced 1 week later (NAR4), also contained more inferential and misunderstanding relevant content, such as explanations, attributions of internal states, and expression of the false belief and of its rectification.
All the narratives produced after the SCI contained more causal markers and were longer than the initially produced narrative, a result that may be directly related to the higher inferential content and the greater elaboration of the plot of the narratives produced after the SCI.
First and second graders improved their narratives more than did kindergarten children.
The narratives of the Control group did not change with repetition.
Comments
Veneziano et al. (2020) have offered a very simple strategy to promote young elementary school children’s ability to explain how characters in stories are interpreting events simply by having them focus on the characters’ behaviors. Two features of the SCI may account for its success: story segmentation and focus on causality. The SCI segments the story into smaller components and focuses children’s attention on each event one at a time, thus reducing the cognitive load of treating the entire story all at once. The segmentation of the story provides children with greater opportunities to explore the pictures, fill in the “gaps” with inferred components, and possibly end up by having a better comprehension of the plot as a whole. The questions about the causes of the events and behaviors of the characters oriented children’s attention toward this aspect and, consequently, may have helped them discover the mental motivations for the characters’ behaviors. The deeper understanding of the plot leads children to tell a longer and more elaborate narrative and use clearer and more specific linguistic devices to mark causality links.
Children who improve their narratives after the SCI may have the underlying cognitive and linguistic competencies necessary to tell a causal and mind-oriented narrative but, without the facilitating support such as SCI, they may have difficulties in focusing on and integrating all the different aspects involved in the narrative task. The SCI orientation can easily be incorporated into narrative intervention conducted by speech-language pathologists and reading comprehension activities led by teachers.