Eur J Psychol Educ DOI 10.1007/s10212-013-0197-y
Comparing fictional, personal, and hypothetical narratives in primary school: story grammar and mental state language Emiddia Longobardi & Pietro Spataro & Marialuisa Renna & Clelia Rossi-Arnaud
Received: 12 December 2012 / Revised: 5 June 2013 / Accepted: 27 June 2013 # Instituto Superior de Psicologia Aplicada, Lisboa, Portugal and Springer Science+Business Media Dordrecht 2013
Abstract The present study examined the use of narrative categories and mental state words in the fictional, personal, and hypothetical stories written by 150 children in the third, fourth, and fifth grades of primary school. There were three main results. First, children were better able to write fictional and hypothetical than personal stories, when considering the total number of narrative categories and the percentages of stories including at least one complete episode. Second, there was clear evidence of differentiation between the three tasks, both in terms of narrative categories and mental state language. Third, the use of mental state words correlated with the frequency of subordinate propositions and the number of narrative categories included in the stories. These findings support the hypothesis of a bidirectional interaction between lexical and syntactic development and suggest that narrative writing involves metalinguistic abilities directly related to the spontaneous use of psychological lexicon. Educational implications are discussed. Keywords Narrative . Writing . Story grammar . Mental state language . Complex syntax
Introduction Children of all cultures are exposed to a wide variety of narrative genres, first in oral and then in written language. By definition, narrative genres represent “socially constructed language practices serving specific social purposes” (Beers and Nagy 2011, p.186). Previous studies showed that children are able to produce different types of stories from an early age (as young as 2 years: Allen et al. 1994; Berman 1995; Hudson and Shapiro 1991; Peterson and McCabe 1983). To do so, they need to manage at least four types of knowledge (Hudson and Shapiro 1991): (a) content knowledge (i.e., generalized representations of common social events or E. Longobardi (*) : M. Renna Department of Dynamic and Clinical Psychology, University Sapienza of Rome, Italy, Via degli Apuli 1, 00185 Rome, Italy e-mail:
[email protected] P. Spataro : C. Rossi-Arnaud Department of Psychology, University Sapienza of Rome, Italy, Via dei Marsi 78, 00185 Rome, Italy
E. Longobardi et al.
memories of specific episodes); (b) structural knowledge (i.e., knowledge of the structural components that characterize different types of narratives); (c) micro-linguistic knowledge (i.e., knowledge about the phonological, morphological and syntactic rules of language); and (d) contextual knowledge (i.e., knowledge about the pragmatic functions of narratives in specific social contexts: Boscolo 1991). In addition, they are also required to master those linguistic expressions that allow them to represent thoughts, emotions, and beliefs (Fox 1991; Shiro 2003). The present study investigates the development of narrative competence in a sample of school-age children involved in three different types of written tasks (fictional, hypothetical, and personal stories), by analyzing the use of the structural categories proposed by Stein and Glenn (1979) and the frequencies of mental state words (MSW; Bartsch and Wellman 1995). Using the above-mentioned criteria, it is possible to delineate several differences between personal and fictional stories. Personal narratives are accounts of autobiographical events and are typically reported using the past tense, from the narrator’s perspective (Hudson and Shapiro 1991). These stories serve the function of disclosing intimate information to other people and contribute to the establishment of interpersonal relationships. The basic elements of personal narratives are already present in 24- to 36-month-old children (Berman 1995; Eisenberg 1985; Fivush et al. 1987), although, in this period, there is a strong reliance upon parents’ cues and prompts, and the narrated stories resemble a chronological list of events. During the preschool and elementary school years, the coherence of personal stories improves, and children report more elaborated information and references to internal mental states (Allen et al. 1994; Hudson and Shapiro 1991; Peterson and McCabe 1983). On the other hand, fictional stories are narratives deliberately created by people (Schank 1990), on the basis of general event knowledge and, less frequently, memory of other stories (Allen et al. 1994). Thus, relating a good fictional story implies a satisfactory knowledge of social roles, personality types, and social interactions (Hudson and Shapiro 1991). In school, fictional narratives are widely used as a vehicle to develop language and literacy skills (Ukrainetz et al. 2005) and to explain, teach, and transmit cultural and moral knowledge (Stein 1988). The general consensus is that the development of fictional stories lags somewhat behind that of personal narratives, since preschoolers and kindergartners have often difficulty in generating complete fictional episodes (Allen et al. 1994; Hudson and Shapiro 1991; McCabe et al. 2008), and it is not until the years of 6 to 8 that children incorporate in their pretend stories all the key components of this genre, including references to intentional plans and psychological motives. Additionally, we specifically developed for the purpose of the present study the format of the hypothetical story. This narrative task can be conceived of as a middle way between personal and fictional stories, since children were required to invent a fictional episode in which they themselves were the main actors, starting from a brief stem describing an unconventional situation. This genre has the interesting property of stimulating children’s thought about hypothetical reference (Kuczaj and Daly 1979). Researchers found that children are able to reason from premises that did not conform to their everyday knowledge by the age of 4 years (Dias and Harris 1990) and that references to hypothetical future events appear earlier than references to past events (Kuczaj 1981). The present study aimed at analyzing developmental changes in the use of story grammar categories and mental state language. The episodic story grammar analysis proposed by Stein and Glenn (1979) assumes that well-organized stories should follow a standard scheme in which motives (external events or internal states) cause goal-directed actions, which in turn produce observable consequences. More specifically, this coding system allows the examination of several narrative categories, including an initiating problem, one or more attempts to achieve the goal, and the ultimate story outcome (Barnes and Baron-Cohen 2012; Freedman 1987; Norbury and Bishop 2003). Using this model, several studies have sought to compare the characteristics of orally produced personal and fictional stories, to understand if these two
Narrative writing in school-age children
genres follow distinct developmental paths. As mentioned above, the common finding is that children are “better in many ways at producing personal than fictional stories” (McCabe et al. 2008, p.196). Berman (1995); see also Kaderavek and Sulzby 2000), for instance, found that preschool children were capable of constructing well-formed personal episodes, whereas they often engaged in isolated event description when telling fictional stories. Evidence for an advantage of personal narratives has also been reported with older children between 4 and 8 years of age by Allen et al. (1994), who showed that personal event narratives contained more reactive sequences and complete episodes, while fictional stories contained higher proportions of action sequences and multiple episodes. On the other hand, Freedman (1987) reported that, for children of grades 5 and 8, a significantly higher percentage of invented than personal stories satisfied the minimal criteria for an “ideal” structure (i.e., including an initiating event, an action, and a resolution). Another crucial aspect of oral and written narratives is the use of MSW, that is, the set of terms employed to describe the internal states of oneself and others (Grazzani and Ornaghi 2012). The development of the ability to use MSW in spontaneous conversations follows a well-known trajectory, with 2- to 3-year-olds using volitional, perceptual, and emotional terms, and older children between 3 and 5 years of age producing more complex mental-state language referring to cognitive and moral states (Dunn et al. 1991; Hughes and Dunn 1998). A similar progression has also been observed with older children, when analyzing written essays. For instance, Camaioni et al. (1998) found that fifth-grade children employed a higher number of cognitive terms and fewer volitional and perceptual words, relative to third-graders. Although many researchers have independently examined the occurrence of MSW in personal and fictional stories (Bamberg and Damrad-Frye 1991; Berman and Slobin 1994; Meins et al. 2006; Özyildirim 2009; Ukrainetz et al. 2005), only few studies have addressed the question of the differences between these two narrative genres, with results being controversial. In a seminal article, Fox (1991) analyzed the personal and fictional stories written by 135 children aged 9, 11, and 13 years, to assess their ability to represent the actors’ inner world and make inferences about the psychological significance of their behaviors. Results indicated significant age and gender differences, with scores being higher for older than for younger children and higher for girls than for boys but no differences between the two narrative genres. Shiro (2003) examined 444 personal and fictional narratives orally produced by 113 first- and fourth-grade children between 6 and 10 years of age. She found that the overall frequency of evaluative expressions was significantly greater in fictional than in personal stories and that age and socioeconomic status affected the use of evaluations in fictional narratives. In particular, older and high socio-economic status children employed cognitive terms more often than did younger and low socio-economic status children. However, these findings were not replicated by Losh and Capps (2003), who reported that children between 8 and 14 years of age produced more emotional and cognitive terms in personal than in fictional narratives. On the basis of this evidence, the first aim of the present study was to ascertain whether the advantage of personal stories, typically observed during the preschool years in the oral modality (McCabe et al. 2008), would persist in the written essays of older children attending the third, fourth, and fifth grades of primary school. The rationale is that narratives are intimately connected to the social activity structures in which they arise (Hicks 1990). Personal stories may be better developed in the preschool years because this type of narrative discourse is prevalent in the social interactions and freely occurring conversations of kindergartners (Preece 1987). However, this state of affairs is likely to change as soon as children begin to spend many hours in highly structured classroom activities, in which the production of fictional stories “is frequently used in the teaching of reading and writing skills” (Shiro 2003, p.176). A common assumption in the literature is that narrative competence develops through direct experience
E. Longobardi et al.
with texts featuring the specific characteristics of a given genre (Duke 2000). In agreement, it has been found that first-grade children who began primary school with little knowledge of fictional narrative language rapidly acquired it after regular in-school experience with makebelieve stories (Purcell-Gates et al. 1995). Moreover, a study aimed at providing descriptive information about text experiences offered to U.S. children in 20 first-grade classrooms showed that fictional narratives were the most common genre, whereas activities involving informational texts were rare (Duke 2000). Taken together, these results suggest that school-age children between the third and fifth grades might be more proficient in writing fictional than personal stories: Therefore, we expected that a higher number of narrative categories would be included in the former rather than in the latter type of stories and that a higher percentage of fictional narratives would satisfy the minimal criteria for a complete episode (Freedman 1987). In addition, we also expected to find more narrative categories and a higher percentage of complete episodes in hypothetical than in personal stories. This is because, according to Berman (2008), the highest level of maturity in discourse construction requires the creative synthesis between different narrative genres. Berman (2008) observed a shift from a rigid dichotomy between personal and expository essays in children until high-school age, to the emergence of more diversified structures in adolescence. In fact, only adolescents and adults were able to include in their narratives genre-atypical forms of linguistic expression (e.g., references to autobiographical events in expository texts or generalized propositions in personal stories). Therefore, hypothetical stories should promote the use of more advanced narrative skills, compared with personal stories. Besides examining the overall story structure, we also analyzed the frequency of the eight narrative categories identified by Stein and Glenn (1979). Based on previous findings, we expected that: (1) the category “Setting” would be used more often in personal than in fictional and hypothetical stories (Allen et al. 1994; Hudson and Shapiro 1991); (2) the categories “Introduction,” “Internal Response,” and “Resolution” would occur at a greater frequency in fictional and personal narratives than in hypothetical stories, since the former genres are more likely to have underwent the process of narrativization, whereby children progress from a sequential statement of single events to a complete story structure (Allen et al. 1994; Hudson and Shapiro 1991); and (3) the category “Attempt” would be employed with the highest frequency in hypothetical stories, because children tend to focus more on actions when the requirements of the narrative task do not conform to their everyday knowledge (Shapiro and Hudson 1991). The second aim of the present study was to compare the children’s ability to produce MSW in the three types of stories. We expected to find significant differences in the use of specific categories of MSW, with emotional terms occurring more frequently in personal than in fictional and hypothetical narratives (Losh and Capps 2003; Özyildirim 2009) and volitional and cognitive terms occurring more often in fictional stories (Shiro 2003). It is conceivable to suppose that these differences would combine one another, leading to a substantial equivalence between fictional and personal stories in the total frequency of MSW (Fox 1991). Lastly, the third aim was to evaluate the syntactic properties of written stories and how they relate to the use of narrative categories and MSW. We employed the approach of propositional analysis (Coelho et al. 2005; Van Dijk and Kintsch 1983), since we aimed at examining the children’s ability to use subordinate propositions (e.g., “there was a horse in the forest who didn’t want to be seen”; Gutierrez-Clellen and Hofstetter 1994). Although few studies have assessed the relationship between complex syntax and the occurrence of MSW in school-age children, a positive correlation between the two variables was expected. Using the picture-book “Frog, Where are You?”, Manhardt and Rescorla (2002) elicited narratives from a sample of 31 late talkers and 23 typically developing children between 8 and 9 years of age. A factor analysis
Narrative writing in school-age children
revealed that syntactic structure use (a variable assessing the total number of complex propositions, like relative clauses and sentential complements: de Villiers and Pyers 2002; Smith et al. 2003) and the mean subordinate clause length loaded on the same factor as evaluative information (a variable assessing the use of MSW referring to characters’ emotions and cognitions). According to Manhardt and Rescorla (2002), this result pointed to a high degree of intercorrelation among syntactic and evaluative measures. In a similar vein, Sun and Nippold (2012) found that, in three groups of typically developing children aged 11, 14, and 17 years, the use of abstract nouns and cognitive verbs was significantly associated with the production of complex syntax and subordinate clauses. Additionally, we also predicted a significant correlation between the number of narrative categories and the frequency of MSW. The latter expectation is supported by experimental findings showing that the ability of 6- to 8-year-old children to realize an ideal story structure is significantly related to the frequency of MSW in orally elicited story narratives (Charman and Shmueli-Goetz 1998). Here, we were interested in ascertaining whether this association extended to older children (8- to 10-year-olds) involved in written narrative tasks. In sum, the present study was designed to address the following questions: 1. Do school-age children between the third and fifth grades use more narrative categories and produce more complete episodes in fictional and hypothetical stories than in personal narratives? 2. Are there differences between the three types of stories in the frequency of narrative categories? 3. Are there differences between the three types of stories in the total frequency of MSW and the use of mental-state language categories? 4. Does the use of MSW correlate with the frequency of subordinate propositions and the number of narrative categories included in the stories?
Method Participants One hundred fifty children from two primary schools in the urban area of Rome (Italy) equally divided between the third, fourth, and fifth grades, participated in the study. There were 77 boys and 73 girls, with a mean age of 8.5 (SD=0.3, range, 7.9–9.5), 9.7 (SD=0.3, range, 8.9–10.5) and 10.8 (SD=0.5, range, 10.2–12.6) years, respectively. Their socio-economic status was medium and medium–high, as defined on the basis of mothers’ education (e.g., graduate or high school) and fathers’ occupations (e.g., clerical worker/official, teacher, businessman, professional). The three groups performed within age-appropriate limits in the Italian version of the Peabody Picture Vocabulary Test (Stella et al. 2000), with standardized scores being 101.5 (SD=14.2), 109.8 (SD=11.6) and 106.1 (SD=11.8), respectively. No child had known or suspected sensory, intellectual, speech, language, or learning deficits, based on teachers’ and parents’ reports. Procedure Two trained assistants visited each class before task administration, to obtain the collaboration of teachers and the informed consents from the children’s parents to participate in a study generally described as “an inquiry into the writing abilities of primary school children”. Following Fox (1991), all writings were carried out collectively in three separate sessions and were embedded
E. Longobardi et al.
into the usual classroom activities. Both the researchers and teachers were instructed to introduce the narrative tasks in a standardized way and to provide no help during the composition. Discussions between children were not allowed throughout the period of task administration. All participants were asked to write three different types of stories. In the task “Invent a story” (Camaioni et al. 1998), children were given a neutral keyword and were required to create a fictional story based on that stimulus. The keywords could be Animate (e.g., astronaut, robot, whale, policeman, parrot), Artefact (e.g., glue, scissor, snack, rucksack), or Natural (e.g., water, wind, sun, puddle, darkness). Across the three grades, the use of the 42 keywords was balanced across participants, so that an equal number of children produced fictional stories starting from the keywords of each category. For personal stories, participants were told to write about an autobiographical event, choosing among six routine situations selected to stimulate private memories (“Tell what happened when you went: to grandparents’ home, to the park, to a friend’s home, to the restaurant, to the doctor, to the supermarket”). Finally, in the hypothetical task, children were requested to write a story in which they were the main actors, by completing the following situation: “Imagine that a day you and [a selected partner] are in [a selected place] and hear a noise behind your back. You turn around and …” Coding Written texts were coded for several variables. First, the total number of words and propositions contained in each story was tallied and averaged. Propositions were defined as a verb plus its arguments (Coelho et al. 2005; Van Dijk and Kintsch 1983). Following Reilly et al. (2004), complex propositions were represented by: (a) coordinate propositions, in which two sentences having the same rank are linearly ordered to form a composite event by means of coordinating connectives and (b) subordinate propositions, in which two sentences are hierarchically ordered (i.e., one fact is taken as a specification of the other) through the use of relative pronouns (“The girl saw the man that killed the frog”; Gutierrez-Clellen and Hofstetter 1994). For the purpose of analysis, the raw frequencies of coordinate and subordinate propositions were divided by the total number of propositions produced in each story (Losh and Capps 2003; Reilly et al. 2004), yielding two proportional measures, termed Frequency of Coordinate Propositions and Frequency of Subordinate Propositions. Narrative structure was examined using the eight categories proposed by Stein and Glenn (1979): (1) Introduction (“There was once a 12-year-old girl named Sabrina”, YF third grade); (2) Setting (i.e., a description of the story context: “The treasure was enclosed in a gold chest located in a volcano”, BB fifth grade); (3) Initiating Event (i.e., environmental changes or things that happen to the main characters: “One day Zach ran away from home and went into a river”, FA fourth grade); (4) Internal Response (i.e., the characters’ thoughts and emotions in response to the initiating event: “The hunter saw the horse and got scared”, MB third grade); (5) Planning (i.e., the characters’ goals and plans: “It was too hot, so we decided to stop in a bar”, SA fifth grade); (6) Attempt (i.e., the characters’ actions to reach the goals: “Goku and Vegeta took the radar to locate Freezer”, AA third grade); (7) Direct Consequences (i.e., the result of the characters’ actions: “[I fall off the bicycle] but I was lucky because I did not break my arm”, DD fifth grade); and (8) Resolution (i.e., the end of the story: “And from that day I understood I cannot run too much fast”, DD fifth grade). From this analysis, a measure was obtained indicating the number of different categories (range, 1–8) produced in the three types of stories. The frequencies of the eight categories were computed as a ratio of the total number of categories coded in each story. Lastly, MSW were classified in four categories (Lecce et al. 2010; Ornaghi et al. 2011): Emotional (e.g., happy, proud, angry, to hate, to fall in love), Volitional (e.g., to intend, to
Narrative writing in school-age children
decide, able, skillful), Cognitive (e.g., to understand, to believe, to remember, to be sure, to think), and Moral (bad, nasty, to sacrifice, to regret, to forgive). References to MSW were not counted if they appeared within idiomatic expressions, like “you must know that” or “and they lived happily ever after,” or were used to denote stereotyped personality traits (e.g., “Giovanni is clever”), without no clear evidence of genuine psychological reference (Bartsch and Wellman 1995). To control for individual differences in verbosity, the overall frequency of MSW (collapsed across the four categories) was computed as a ratio of the total number of words produced in each story. Likewise, to control for individual differences in the tendency to spontaneously use mental-state language, the frequency of individual categories was computed as a ratio of the total number of MSW reported in each story; it is worth noting, however, that the main results did not change when the latter values were computed as a ratio of the total number of words. Reliability was evaluated by having a second trained assistant coding 30 % of the narratives. The mean inter-rater agreements were 95 % (Cohen’s K=0.94) for propositional coding, 91 % (Cohen’s K=0.82) for narrative categories, and 92 % (Cohen’s K=0.89) for MSW.
Results Preliminary analyses Preliminary analyses were firstly conducted to ascertain whether the choice of the type of keyword in fictional stories and the type of event in personal narratives affected story measures: However, no significant differences were found across the three grades (for fictional stories: Fs(2, 147)<1.52, p>0.29; for personal stories: Fs(5, 144)<1.88, p>0.11). Secondly, Spearman’s correlations verified that the PPVT scores were not significantly associated with the total number of words (rs<0.14, p>0.09) and propositions (rs<0.13, p>0.12) produced in the stories, as well as with the frequency of subordinate propositions (rs<0.11, p>0.19) and the total number of narrative categories (rs<0.12, p>0.15). Thus, the PPVT performance was not included as a covariate in the subsequent analyses. Tables 1 and 2 illustrate the mean number of words and propositions produced by children and the mean frequency of coordinate and subordinate propositions, as a function of grade and story type. The last columns report the results of the univariate between-subjects ANOVAs testing the effects of Grade. As can be noted, significant differences between the three grades were obtained on all variables, except for the frequency of coordinate propositions in personal stories. Overall, the total number of words and propositions and the frequency of subordinate propositions increased across the three grades, whereas the frequency of subordinate propositions decreased. Regarding the effects of Story Type, children wrote significantly more words and propositions in fictional and hypothetical stories than in personal narratives [for words: F(2, 298)=10.23, MSE=3124.69, p<0.01, η 2 =0.06; for propositions: F(2, 298)=19.90, MSE=104.01, p<0.001, η2=0.12], whereas no differences between the three types of stories were found on the frequency of coordinate and subordinate propositions [F(2, 298)=2.38, p=0.12; and F(2, 298)=0.04, p=0.96]. Total number of narrative categories To investigate the first question of the present study (Do school-age children between the third and fifth grades use more narrative categories in fictional and hypothetical stories than in personal narratives?), the total number of narrative categories produced in the stories was
E. Longobardi et al. Table 1 Mean number of words and propositions, as a Function of Story Type, and Grade III Grade
IV Grade
V Grade
Grade effect
Fictional story
100.3 (50.3)
193.7 (86.2)
163.4 (75.4)
F(2,147)=21.69**
Personal story
96.4 (48.3)
147.7 (75.7)
152.8 (92.3)
F(2,147)=8.78**
119.3 (67.9)
159.4 (62.1)
203.3 (94.0)
F(2,147)=15.28**
Fictional story
18.1 (8.7)
34.8 (15.7)
27.7 (12.8)
F(2,147)=21.22**
Personal story
16.5 (9.4)
23.9 (12.5)
23.6 (15.6)
F(2,147)=5.35*
Hypothetical story
20.0 (10.5)
29.4 (12.6)
35.8 (16.8)
F(2,147)=17.09**
Number of words
Hypothetical story Number of propositions
Standard deviations in parentheses *p<0.01, **p<0.001
submitted to a mixed 3(Grade: third, fourth, and fifth grade)×3 (Story Type: fictional, hypothetical, and personal) ANOVA. Results showed a significant main effect of Story Type [F(2, 294)=16.44, MSE=1.47, p<0.001, η2=0.10]: Post hoc comparisons, with the Bonferroni adjustment, indicated that children produced more narrative categories in fictional (M=4.79) and hypothetical stories (M=4.54) than in personal narratives (M=4.01; p<0.05; see Fig. 1). The main effect of Grade did not reach the significance level [F(2, 147)=2.25, p=0.11], but the interaction between the two factors did [F(4, 294)=2.48, MSE=1.47, p<0.05, η2=0.03]. A follow-up analysis of simple effects revealed significant differences between the three grades for fictional stories [F(2, 147)=5.27, MSE=1.88, p<0.01, η 2=0.07], with fourth- and fifth-graders outperforming third-graders (both ps<0.05) but not for personal and hypothetical stories [F(2, 147)<1.40, p>0.25]. The same analysis revealed significant main effects of Story Type for fourth- and fifth-grade children [F(2, 146)=9.86, p<0.001, η2=0.12, and F(2, 146)=15.08, p<0.001, η2=0.17] but not for third-grade children [F(2, 146)=1.16, p=0.32]. Fourth-grade children produced significantly more narrative categories in fictional than in personal stories (M=5.02 versus M=4.06, p<0.001) and marginally more categories in fictional than in hypothetical stories (M=5.02
Table 2 Frequency of coordinate and subordinate propositions, as a Function of Story Type and Grade III Grade
IV Grade
V Grade
Grade effect
Coordinate propositions Fictional story
0.42 (0.19)
0.32 (0.13)
0.33 (0.14)
F(2,147)=5.98*
Personal story
0.38 (0.13)
0.36 (0.12)
0.34 (0.11)
F(2,147)=1.14
Hypothetical story
0.51 (0.15)
0.28 (0.10)
0.27 (0.07)
F(2,147)=64.29**
Fictional story Personal story
0.19 (0.15) 0.19 (0.11)
0.28 (0.10) 0.24 (0.13)
0.31 (0.13) 0.27 (0.10)
F(2,147)=9.99** F(2, 147)=5.57*
Hypothetical story
0.13 (0.10)
0.26 (0.11)
0.32 (0.10)
F(2, 147)=44.18**
Subordinate propositions
Standard deviations in parenthesis *p<0.01, **p<0.001
Narrative writing in school-age children Fig. 1 Mean number of narrative categories produced in fictional, personal, and hypothetical stories, as a function of Grade. Bars represent standard deviations
versus M=4.44, p=0.07). Similarly, fifth-grade children used more narrative categories in fictional and hypothetical stories, compared with personal narratives (M=5.08 and M=4.78 versus M=3.92, both ps<0.01). To ascertain whether children produced more complete episodes in fictional and hypothetical stories (relative to personal narratives), the percentages of stories satisfying the minimal criteria for an ideal structure were also analyzed. An episode was considered to be complete if it contained at least some reference to: (a) an initiating event; (b) an attempt (i.e., a goal-directed action); and (c) a resolution, indicating the attainment or non-attainment of the goal. As expected, a chi square for presence versus story type was significant [χ2(2)=18.87, p<0.001], indicating that a higher percentage of fictional stories (55 %) satisfied the minimal criteria for a complete episode structure, compared with hypothetical (41 %) and personal narratives (30 %). Frequency of narrative categories To assess the second question (Are there differences between the three types of stories in the frequency of individual narrative categories?), a series of univariate analyses were conducted, considering Story Type as the within-subjects factor (see Table 3). Significant differences between the three stories were observed for the categories Introduction [F(2, 298)=10.49, MSE=0.006, p<0.001, η2=0.07], Setting [F(2, 298)=15.63, MSE=0.018, p<0.001, η2=0.09], Internal Response [F(2, 298)=10.25, MSE=0.005, p<0.001, η2=0.06], Attempt [F(2, 298)=34.55, MSE=0.033, p<0.001, η2=0.18], Direct Consequences [F(2, 298)=12.10, MSE=0.003, p<0.001, η 2 =0.08], and Resolution [F(2, 298)=9.27, MSE=0.002, p<0.001, η2=0.06]. Post hoc comparisons, with the Bonferroni adjustment, indicated that: (a) the categories Introduction, Internal Response, and Resolution were used more often in fictional and personal narratives than in hypothetical stories (all ps≤0.01); (b) the category Direct Consequences was produced more often in fictional than in personal and hypothetical stories (all ps≤0.01); (c) the category Setting occurred more frequently in personal narratives than in fictional and hypothetical stories (all ps≤0.001); and (d) the category Attempt occurred more often in hypothetical stories than in personal and fictional narratives (all ps<0.001).
E. Longobardi et al.
Table 3 Mean proportions of narrative categories in fictional, personal, and hypothetical stories
Standard deviations in parenthesis Proportions were computed as ratios of the total number of categories included in each story
Narrative categories
Story type Fictional
Personal
Hypothetical
Major setting
0.07 (0.10)
0.07 (0.08)
0.03 (0.04)
Minor setting
0.07 (0.09)
0.15 (0.19)
0.08 (0.08)
Initiating event
0.28 (0.14)
0.26 (0.15)
0.24 (0.10)
Internal response
0.06 (0.07)
0.05 (0.09)
0.02 (0.04)
Planning Attempt
0.01 (0.03) 0.38 (0.18)
0.01 (0.02) 0.37 (0.22)
0.01 (0.02) 0.53 (0.15)
Direct consequence
0.05 (0.08)
0.02 (0.05)
0.03 (0.04)
Resolution
0.05 (0.05)
0.04 (0.06)
0.02 (0.03)
Mental state words Question 3a aimed at investigating differences between the three types of stories in the occurrence of MSW. To this purpose, the total proportions of MSW, collapsed across all categories (see Fig. 2), were submitted to a mixed 3(Grade: third, fourth, and fifth grade)×3 (Story Type: fictional, hypothetical, and personal) ANOVA. Results revealed a significant main effect of Story Type [F(2, 294)=4.02, MSE=0.0004, p≤0.05, η2=0.03] and a marginal main effect of Grade [F(2, 147)=2.88, MSE=0.0005, p=0.059, η2=0.04]. Post hoc comparisons, using the Bonferroni adjustment, showed that MSW occurred more frequently in fictional and personal narratives (M=0.25 and M=0.26) than in hypothetical stories (M=0.20; p≤0.05 for both comparisons) and that fifth-graders tended to produce more MSW than third-graders (M=0.27 and M=0.20; p=0.056). However, these effects were qualified by a significant interaction between Grade and Story Type [F(4, 294)=2.89, MSE=0.0004, p<0.05, η2=0.04]. A followup analysis of simple effects found that the effect of Grade was significant for hypothetical
Fig. 2 Mean frequency of mental state words in fictional, personal, and hypothetical stories, as a function of Grade. Bars represent standard deviations
Narrative writing in school-age children
stories [F(2, 147)=12.88, MSE=0.0002, p<0.001, η2=0.14], with fourth- and fifth-graders outperforming third-graders (all ps<0.001) but not for fictional or personal stories [F(2, 147)<1.87, p>0.15]. The same analysis indicated that the effect of Story Type (i.e., the advantage of fictional and personal narratives over hypothetical stories) was significant for third-grade children [F(2, 146)=11.16, p<0.001, η2=0.13] but not for fourth- and fifth-grade children [F(2, 146)<0.58, p>0.56]. Therefore, the latter two groups produced the same amount of MSW in all types of stories (Fig. 2). Regarding question 3b (Are there differences between the three types of stories in the frequency of individual categories of mental-state language?), the mean proportions of emotional, volitional, cognitive, and moral terms (see Table 4) were analyzed with a repeated-measures MANOVA, considering Story Type as the within-subjects factor. As expected, the main effect of Story Type was significant [F(8, 592)=10.81, p<0.001, η2=0.13]. Follow-up univariate analyses revealed significant differences: (1) for emotional terms [F(2, 298)=31.40, MSE=0.045, p<0.001, η2=0.17], with frequency being greater in personal than in fictional stories and greater in fictional than in hypothetical stories (p<0.01 for both comparisons); and (2) for volitional terms [F(2, 298)=3.70, MSE=0.046, p<0.05, η2=0.02], with frequency being greater in fictional than in hypothetical stories (p≤0.01). No significant differences were found for cognitive and moral terms [F(2, 298)=0.62, p=0.53 and F(2, 298)=2.25, p=0.11]. Correlations The last aim of the present study was to address the relationships between syntactic, narrative, and lexical measures (question 4). Spearman’s correlations between the frequency of subordinate propositions, the total number of narrative categories, and the frequency of MSW are shown in Table 5. As predicted, the correlations between the frequency of subordinate propositions and MSW (Manhardt and Rescorla 2002; Sun and Nippold 2012) were significant in fictional and hypothetical stories and marginally significant in personal narratives. Similarly, the hypothesized correlations between the total number of narrative categories and the frequency of MSW (Charman and Shmueli-Goetz 1998) were significant in all three types of stories. In addition, the use of subordinate propositions was significantly associated with the total number of narrative categories produced in fictional stories. Importantly, all of the above correlations remained significant after applying the Bonferroni correction—for each story, the adjusted α was equal to 0.016 (0.05/3), whereas the minimum observed α was 0.007. The only exception was the marginal correlation between the frequency of subordinate propositions and MSW in personal narratives, which fell below the significance level.
Table 4 Mean proportions of emotional, volitional, cognitive, and moral terms, as a function of story type Standard deviations in parenthesis Proportions were computed as ratios of the total number of mental state words contained in each story
Mental-state words
Story type Fictional
Personal
Hypothetical
Emotional
0.18 (0.24)
0.28 (0.31)
0.08 (0.10)
Volitional
0.20 (0.24)
0.14 (0.23)
0.14 (0.19)
Cognitive
0.15 (0.21)
0.13 (0.22)
0.13 (0.17)
Moral
0.08 (0.16)
0.10 (0.21)
0.06 (0.12)
E. Longobardi et al.
Table 5 Spearman’s correlations between the frequency of coordinated and subordinated propositions, the number of narrative categories, and the frequency of mental state words
1
2
3
Fictional story Subordinated propositions Number of narrative categories
1.00 0.22**
1.00
Mental state words
0.30*
0.22**
1.00
Personal story Subordinated propositions
1.00
Number of narrative categories
0.07
1.00
Mental state words
0.16†
0.22**
1.00
Hypothetical story
*p<0.001, **p<0.01, †0.05
Subordinated propositions Number of narrative categories
1.00 0.14
1.00
Mental state words
0.40*
0.23**
1.00
Discussion The present cross-sectional study examined the use of narrative categories and MSW in three types of stories (fictional, personal, and hypothetical) written by a sample of 150 Italian children between the third and fifth grades of primary school. The results showed that children were more proficient at writing fictional and hypothetical than personal stories and were capable of making fine-grained distinctions between different narrative genres, both in terms of narrative categories and mental state language. In addition, there were reciprocal correlations between the frequency of subordinate propositions, the production of MSW, and the total number of narrative categories. Our first prediction, stating that children would demonstrate a more advanced level of narrative competence when required to write fictional and hypothetical stories, was confirmed. Indeed, we found that children produced a greater number of narrative categories in fictional and hypothetical stories than in personal ones. Furthermore, replicating previous findings reported by Freedman (1987), a significantly higher percentage of fictional and hypothetical narratives satisfied the minimal criteria for a complete episode (Barnes and Baron-Cohen 2012), relative to personal stories. According to our proposal, these results may be explained by the enhanced exposure to the fictional genre, which is typically associated with the process of acquisition of writing skills in primary school (Duke 2000; Purcell-Gates et al. 1995; Shiro 2003). Two lines of evidence support the above conclusion. First, a significant increase in the number of narrative categories between the three grades was observed in fictional but not in personal or hypothetical stories, suggesting that the text experiences offered to school-age children (for which previous studies documented a strong prevalence of the fictional genre: Duke 2000) did not lead to a substantial improvement in the structural organization of the latter two types of narratives. Second, significant differences in the number of narrative categories were observed in fourth- and fifth-grade children (favoring fictional and hypothetical stories) but not in third-grade children, indicating that narrative competence developed at different rates in the three genres, as far as children became more familiar with storytelling activities and the language of schooling (Schleppegrell 2001). In addition to school experiences, a second reason why children may be less able to conform to a prototypical story schema when writing personal than fictional narratives may reside in the fact that “much that happens in the real world does not include the goal-oriented behaviors that
Narrative writing in school-age children
Stein and Glenn characterize as central to a complete episode” (Freedman 1987, p. 160). In agreement, Peterson and McCabe (1983) reported that, for personal stories, reactive sequences (i.e., sequences in which a set of changes automatically cause other changes, without any form of causality) remained a potential alternate form of story structure throughout middle childhood, with their frequency being constant in children between 5 and 9 years of age (40 %–50 %). Likewise, Freedman (1987) found that actions sequences (i.e., lists of actions chronologically ordered, with no planning) were employed twice as often in personal as invented stories in Grade 5 and three to four times as often in Grades 8 and 12. As concerns the second question of the present study (Are there differences between the three types of stories in the frequency of individual narrative categories?), our data confirm and extend previous evidence indicating a clear differentiation between the three types of stories (Allen et al. 1994; Hudson and Shapiro 1991; Peterson and McCabe 1983). To recapitulate, we found that the categories Introduction, Internal Response, and Resolution appeared more often in fictional and personal narratives than in hypothetical stories, whereas the categories Setting and Attempt showed the highest frequency in personal and hypothetical stories, respectively. In line with our results, Allen et al. (1994) reported that significantly more descriptive sequences (i.e., sequences describing characters, setting and habitual actions, with no causal relationship) were contained in personal than in fictional narratives; similarly, the children examined by Hudson and Shapiro (1991) produced more descriptive propositions in personal narratives than in scripts and fictional stories. Hence, one distinctive feature of personal stories in primary school is the tendency to focus on the description of the characters and the environment in which the main episodes take place. On the other hand, the finding that the frequency of the categories Introduction and Resolution did not differ between fictional and personal stories may be related to emergent process of narrativization (Hudson and Shapiro 1991), whereby children attempt to apply to autobiographical events the typical structure of well-formed fictional stories. In doing so, they begin to recognize the crucial function of introductory and conclusive comments in enhancing the coherence of written narratives. Finally, the increased occurrence of the category Attempt in hypothetical stories might be ascribed to the specific properties of this narrative task, which was based on the description of an uncommon event and thus required participants to adopt new and creative solutions (compared with the highly practiced formats of fictional and personal narratives). Shapiro and Hudson (1991) found that preschoolers and first-graders focused on actions, used less complex syntactic structures, and included fewer references to the characters’ internal states when requested to retell sequences of unrelated events. Similarly, it appears from our data that children were more reliant on action sequences when the instructions made it difficult for them to adopt the standard story schema of fictional narratives and thus imposed higher cognitive demands on their limited event knowledge. The third aim of the present study was to compare the use of MSW in different narrative genres. Again, the results revealed a clear differentiation between the three types of stories, with volitional and emotional terms showing the greatest frequency in fictional and personal stories, respectively. The higher occurrence of emotional evaluations in personal narratives was expected, since the same pattern has been reported in previous studies. For instance, Losh and Capps (2003) showed that typically developing children between 8 and 14 years of age employed nearly twice as much emotional expressions in personal than in storybook narratives, and Özyildirim (2009) found that emotional terms were prevalent in the personal experience narratives of Turkish university students. Likewise, the higher frequency of volitional terms in fictional stories is consistent with the findings reported by Shiro (2003): In that study, intention and cognitive terms were used more often in fictional than in personal narratives, whereas the reverse occurred for perceptual and emotional terms. It is
E. Longobardi et al.
also interesting to note that a significant increase in the production of MSW was only observed in hypothetical stories, suggesting that this narrative task may be particularly appropriate for examining developmental changes in the ability of school-age children to use mental state language. The last aim of the present study was to assess the concurrent relationships between syntactic, lexical, and narrative measures. In line with our predictions, the frequency of MSW was positively correlated with the use of subordinate propositions in hypothetical and fictional stories. Such a result is consistent with other findings demonstrating that, for school-age children, the use of evaluative information and cognitive verbs is significantly associated with the production of complex syntax (Manhardt and Rescorla 2002; Sun and Nippold 2012) and provides further evidence in support of the so-called lexicon–syntax interface, namely the hypothesis of a rich interaction between lexical and syntactic development in the context of narrative writing (Ravid 2004). In addition, the frequency of MSW was also correlated with the number of narrative categories included in the stories, suggesting that individual differences in the tendency to spontaneously use one’s own skills of representing mental states play a crucial role in the production of complex and coherent narratives. This is because “writing is by its very nature a metalinguistic activity” (Olson 1991, p. 260), in which the syntactic and semantic aspects of language become an object of thought. Charman and Shmueli-Goetz (1998) found that a composite measure of story structure, which assessed the children’s capability of producing the components of formal opening, main theme and formal ending, was significantly related to the frequency of MSW in oral narratives. In commenting their results, the authors wrote that: “Story structure devices might reflect a meta-linguistic skill related to a mentalistic stance that takes account of the listener’s perspective” (Charman and Shmueli-Goetz 1998, p. 263). The close relationship between narrative competence and MSW is confirmed by the fact that children with autism, who are significantly impaired in the understanding of false beliefs (Baron-Cohen et al. 1985), produce impoverished narratives when compared with typically developing children, especially with reference to the paucity of mental state language (Tager-Flusberg 1995). The present data have a number of educational implications. First, on the basis of the results obtained with typically developing and language-impaired preschoolers, McCabe et al. (2008) proposed that personal narratives should be recommended as the appropriate genre for early writing by educators, and they should be preferred for the purposes of assessment and intervention in the field of speech–language pathology, because they are better formed than fictional narratives. The present findings suggest that there may be important cultural and agerelated differences in the type of stories which are likely to elicit the application of the children’s most advanced narrative abilities and that fictional stories may be equally or even more appropriate than personal ones for the assessment of school-age children. As a consequence, educators and clinicians must be aware of these discrepancies and select the narrative genres most suitable for use with primary school children. A second implication directly follows from the above considerations, as it concerns the opportunity to enhance the quality of personal narratives. As acknowledged by Gelati (2012), primary school teachers often worry about the orthographic and morphological aspects of written stories and do not spend sufficient time in explaining how to integrate the referential and evaluative components of autobiographical narratives (i.e., the description of external events and internal emotions, respectively). The consequence is that the personal stories written by school-age children often resemble lists of events and facts, without personal comments or information about subjective states. Our results substantially confirm this view, since the use of MSW was very limited in all three types of stories. Regarding this issue, a training study by Gelati and Boscolo (2009; see also Boscolo et al. 2012) demonstrated that an instructional intervention focused on the comprehension and application of a number of evaluative strategies
Narrative writing in school-age children
significantly improved the children’s ability to coherently narrate events and express feelings and personal thoughts. Along this line of research, the combination of fictional and autobiographical cues in the hypothetical stories appeared to improve the structural quality of children’s narratives (relative to personal stories), as indicated by significant increases in the number of narrative categories and the percentages of stories containing complete episodes. Furthermore, only the hypothetical stories showed a significant increase in the frequency of MSW across the three grades. These findings suggest that the introduction of fictional elements into personal stories may represent an effective way to trigger the use of more advanced lexical, narrative, and syntactic skills. One last note concerns the observed correlations between the number of narrative categories, the frequency of subordinate propositions and the use of MSW. Our suggestion that these findings may be due to the metalinguistic nature of story writing potentially links the present data to a productive line of research examining the relationships between narrative abilities, theory of mind, and metacognitive skills (Doherty and Perner 1998; Lecce et al. 2010; Peskin and Astington 2004). Previous studies with preschoolers found that training interventions focused on the active use of mental state lexicon and sentential complements significantly improved performance in theory-of-mind tasks (Hale and Tager-Flusberg 2003; Lohmann and Tomasello 2003; Ornaghi et al. 2011), although the passive exposure to cognitive language in story books does not necessarily result in a better understanding of false belief (Peskin and Astington 2004). In future studies, it would be helpful to ascertain whether an early training in theory of mind and metalinguistic knowledge may result in a later improvement of narrative competence and a more frequent use of MSW in older children attending primary school.
References Allen, M., Kertoy, M., Sherblom, J. C., & Pettit, J. M. (1994). Children narrative productions: A comparison of personal event and fictional stories. Applied PsychoLinguistics, 15, 149–176. Bamberg, M., & Damrad-Frye, R. (1991). On the ability to provide evaluative comments: Further explorations of children's narrative competencies. Journal of Child Language, 18, 689–710. Barnes, J. L., & Baron-Cohen, S. (2012). The big picture: Storytelling ability in adults with autism spectrum conditions. Journal of Autism and Developmental Disorders, 42, 1557–1565. Baron-Cohen, S., Leslie, A. M., & Frith, U. (1985). Does the autistic child have a ‘theory of mind’? Cognition, 21, 37–46. Bartsch, K., & Wellman, H. M. (1995). Children talk about the mind. New York: Oxford University Press. Beers, S. F., & Nagy, W. E. (2011). Writing development in four genres from grades three to seven: Syntactic complexity and genre differentiation. Reading and Writing, 24, 183–202. Berman, R. A. (1995). Narrative competence and storytelling performance: How children tell stories in different contexts. Journal of Narrative and Life History, 5, 285–313. Berman, R. A. (2008). The psycholinguistics of developing text construction. Journal of Child Language, 35, 735–771. Berman, R. A., & Slobin, D. I. (1994). Relating Events in Narratives: A Crosslinguistic Developmental Study. Hillsdale, NJ: Lawrence Erlbaum Associates. Boscolo, P. (1991). Contexts for writing, writing in context. European Journal of Psychology of Education, 6, 167–174. Boscolo, P., Gelati, C., & Galvan, N. (2012). Teaching elementary school students to play with meanings and genre. Reading & Writing Quarterly, 28, 29–50. Camaioni, L., Longobardi, E., & Bellagamba, F. (1998). Evoluzione dei termini di stati mentali nelle storie di fantasia scritte da bambini in età scolare [The development of mental state terms in fantasy stories written by school-aged children]. Età Evolutiva, 60, 20–29. Charman, T., & Shmueli-Goetz, Y. (1998). The relationship between theory of mind, language, and narrative discourses: an experimental study. Cahiers de Psychologie Cognitive/Current Psychology of Cognition, 17, 245–271.
E. Longobardi et al. Coelho, C. A., Grela, B., Corso, M., Gamble, A., & Feinn, R. (2005). Microlinguistic deficits in the narrative discourse of adults with traumatic brain injury. Brain Injury, 19, 1139–1145. de Villiers, J., & Pyers, J. (2002). Complements to cognition: A longitudinal study of the relationship between complex syntax and false-belief understanding. Cognitive Development, 17, 1037–1060. Dias, M. G., & Harris, P. L. (1990). The influence of the imagination on reasoning by young children. British Journal of Developmental Psychology, 8, 305–318. Doherty, M. J., & Perner, J. (1998). Metalinguistic awareness and theory of mind: Just two words for the same thing? Cognitive Development, 13, 279–305. Duke, N. K. (2000). 3.6 minutes per day: The scarcity of informational texts in first grade. Reading Research Quarterly, 35, 202–224. Dunn, J., Brown, J., Slomkowski, C., Tesla, C., & Youngblade, L. (1991). Young children’s understanding of other people’s feelings and beliefs: Individual differences and their antecedents. Child Development, 62, 1352–1366. Eisenberg, A. R. (1985). Learning to describe past experiences in conversation. Discourse Processes, 8, 177–204. Fivush, R., Gray, J. T., & Fromhoff, F. A. (1987). Two-year-olds talk about the past. Cognitive Development, 2, 393–410. Fox, R. (1991). Developing awareness of mind reflected in children’s narrative writing. British Journal of Developmental Psychology, 9, 281–298. Freedman, A. (1987). Development in story writing. Applied PsychoLinguistics, 8, 153–170. Gelati, C. (2012). Writing complexity: Effective interventions to improve narrative writing. In C. Gelati, B. Arfé, & L. Mason (Eds.), Issues in writing research. In honour of Piero Boscolo (pp. 147–161). Padova: Cleup. Gelati, C., & Boscolo, P. (2009). Improving the quality of primary school children's narration of personal events: An intervention study on the use of evaluation strategies. L1 –Educational Studies in Language and Literature, 9, 1–28. Grazzani, I., & Ornaghi, V. (2012). How do use and comprehension of mental-state language relate to theory of mind in middle childhood? Cognitive Development, 27, 99–111. Gutierrez-Clellen, V. F., & Hofstetter, R. (1994). Syntactic complexity in Spanish narratives: A developmental study. Journal of Speech and Hearing Research, 37, 645–654. Hale, C. M., & Tager-Flusberg, H. (2003). The influence of language on theory of mind: A training study. Developmental Science, 6, 346–359. Hicks, D. (1990). Narrative skills and genre knowledge: Ways of telling in the primary school grades. Applied PsychoLinguistics, 11, 83–104. Hudson, J. A., & Shapiro, L. R. (1991). From knowing to telling: Children’s scripts, stories, and personal narratives. In A. McCabe & C. Peterson (Eds.), Developing Narrative Structure. Hillsdale, NJ: Erlbaum. Hughes, C., & Dunn, J. (1998). Understanding mind and emotions: Longitudinal associations with mentalstate talk between young friends. Developmental Psychology, 34, 1026–1037. Kaderavek, J. N., & Sulzby, E. (2000). Narrative production by children with and without specific language impairment: Oral narratives and emergent readings. Journal of Speech, Language, and Hearing Research, 43, 34–49. Kuczaj, S. A., II. (1981). Factors influencing children’s hypothetical reference. Journal of Child Language, 8, 131–138. Kuczaj, S. A., II, & Daly, M. J. (1979). The development of hypothetical reference in the speech of young children. Journal of Child Language, 6, 563–579. Lecce, S., Zocchi, S., Pagnin, A., Palladino, P., & Taumoepeau, M. (2010). Reading minds: The relation between children’s mental state knowledge and their metaknowledge about reading. Child Development, 81, 1876–1893. Lohmann, H., & Tomasello, M. (2003). The role of language in the development of false belief understanding: A training study. Child Development, 74, 1130–1144. Losh, M., & Capps, L. (2003). Narrative ability in high-functioning children with autism or Asperger’s syndrome. Journal of Autism and Developmental Disorders, 33, 239–251. Manhardt, J., & Rescorla, L. (2002). Oral narrative skills of late talkers at ages 8 and 9. Applied PsychoLinguistics, 23, 1–21. McCabe, A., Bliss, L., Barra, G., & Bennett, M. (2008). Comparison of personal versus fictional narratives of children with language impairment. American Journal of Speech-Language Pathology, 17, 194–206. Meins, E., Fernyhough, C., Johnson, F., & Lidstone, J. (2006). Mind-mindedness in children: Individual differences in internal-state talk in middle childhood. British Journal of Developmental Psychology, 24, 181–196. Norbury, C. F., & Bishop, D. V. M. (2003). Narrative skills of children with communication impairments. International Journal of Language & Communication Disorders, 38, 287–313. Olson, D. R. (1991). Literacy as metalinguistic activity. In D. R. Olson & N. Torrance (Eds.), Literacy and Orality (pp. 251–270). New York: Cambridge University Press.
Narrative writing in school-age children Ornaghi, V., Brockmeier, J., & Grazzani Gavazzi, I. (2011). The role of language games in children’s understanding of mental states: A training study. Journal of Cognition and Development, 12, 239–259. Özyildirim, I. (2009). Narrative analysis: An analysis of oral and written strategies in personal experience narratives. Journal of Pragmatics, 41, 1209–1222. Peskin, J., & Astington, J. W. (2004). The effects of adding metacognitive language to story texts. Cognitive Development, 19, 253–273. Peterson, C., & McCabe, A. (1983). Developmental psycholinguistics: Three ways of looking at a child’s narrative. New York, NY: Plenum. Preece, A. (1987). The range of narrative forms conversationally produced by young children. Journal of Child Language, 14, 353–373. Purcell-Gates, V., McIntyre, E., & Freppon, P. (1995). Learning written storybook language in school: A comparison of low-SES children in skills-based and whole language classrooms. American Educational Research Journal, 32, 659–685. Ravid, D. (2004). Derivational morphology revisited: Later lexical development in Hebrew. In R. A. Berman (Ed.), Language Development Across Childhood and Adolescence (pp. 53–81). Amsterdam, The Netherlands: John Benjamins. Reilly, J., Losh, M., Bellugi, U., & Wulfeck, B. (2004). Frog, where are you? Narratives in children with specific language impairment, early focal brain injury and Williams syndrome. Brain and Language, 88, 229–247. Schank, R. C. (1990). Tell Me a Story: Narrative and Intelligence. Evanston, Illinois: Northwestern University Press. Schleppegrell, M. (2001). Linguistic features of the language of schooling. Linguistics and Education, 12, 431–459. Shapiro, L. R., & Hudson, J. A. (1991). Tell me a make-believe story: Coherence and cohesion in young children’s picture-elicited narratives. Developmental Psychology, 27, 960–974. Shiro, M. (2003). Genre and evaluation in narrative development. Journal of Child Language, 30, 165–195. Smith, M., Apperly, I. A., & White, V. (2003). False belief reasoning and the acquisition of relative clause sentences. Child Development, 74, 1709–1719. Stein, N. L. (1988). The development of children’s storytelling skill. In M. B. Franklin & S. Barten (Eds.), Child Language: A Book of Readings. New York: Oxford University Press. Stein, N. L., & Glenn, C. G. (1979). An analysis of story comprehension in elementary school children. In R. O. Freedle (Ed.), New Directions in Discourse Processing (Vol. 2, pp. 53–120). Hillsdale: Lawrence Erlbaum. Stella, G., Pizzoli, C., & Tressoldi, P. E. (2000). Il Peabody Test. Test di vocabolario recettivo. Torino: Omega. Sun, L., & Nippold, M. A. (2012). Narrative writing in children and adolescents: Examining the literate lexicon. Language, Speech, and Hearing Services in Schools, 43, 2–13. Tager-Flusberg, H. (1995). “Once upon a rabbit”: Stories narrated by autistic children. British Journal of Developmental Pyschology, 13, 45–59. Ukrainetz, T. A., Justice, L. M., Kaderavek, J. N., Eisenberg, S. L., Gillam, R. B., & Harm, H. M. (2005). The development of expressive elaboration in fictional narratives. Journal of Speech, Language, and Hearing Research, 48, 1363–1377. Van Dijk, T. A., & Kintsch, W. (1983). Strategies of Discourse Comprehension. New York: Academic Press.
Emiddia Longobardi. University Sapienza of Rome, Department of Dynamic and Clinical Psychology, Via degli Apuli 1, 00185, Rome (Italy). Current themes of research: Language development. Gestural communication. Pragmatic abilities. Adult-child interaction. Text composition. Language assessment. Theory of mind. Most relevant publications in the field of Psychology of Education: Iverson, J. M., Longobardi, E., Spampinato, K., Caselli, M. C. (2006). Gesture and speech in maternal input to children with Down Syndrome. International Journal of Language and Communication Disorders, 3, 235–251.
E. Longobardi et al. Longobardi, E., Rossi-Arnaud, C., Spataro, P. A. (2011). Longitudinal examination of early communicative development: evidence from a parent report questionnaire. British Journal of Developmental Psychology, 29 (3), 572–592. Longobardi, E., Rossi-Arnaud, C., Spataro, P. (2012). Individual differences in the prevalence of words and gestures in the second year of life: developmental trends in Italian children. Infant Behavior & Development, 35, 847–859.
Pietro Spataro. University Sapienza of Rome, Department of Psychology, Via dei Marsi 78, 00185, Rome (Italy). Current themes of research: Implicit memory. Working memory. Attention. Language development. Most relevant publications in the field of Psychology of Education: Spataro, P., Mulligan, N. W., Rossi-Arnaud, C. (2013). Divided attention can enhance memory encoding: the attentional boost effect in implicit memory. Journal of Experimental Psychology: Learning, Memory, and Cognition (in press). Spataro, P., Longobardi, E., Saraulli, D., Rossi-Arnaud, C. (2013). Interactive effects of age-of-acquisition and repetition priming in the lexical decision task: a multiple loci account. Experimental Psychology, 60, 235– 242. Spataro, P., Mulligan, N. W., Longobardi, E., Rossi-Arnaud, C. (2012). Effects of age of acquisition in the word fragment completion task: evidence for an orthographiclocus in implicit memory. Experimental Psychology, 59, 22–29.
Marialuisa Renna. University Sapienza of Rome, Department of Dynamic and Clinical Psychology, Via degli Apuli 1, 00185, Rome (Italy). Current themes of research: Language development. Risk conditions for language development. Text composition. Most relevant publications in the field of Psychology of Education: Longobardi, E., Renna, M., Rinaldi, P., Devescovi, A. (2010). Uno studio longitudinale sullo sviluppo linguistico dibambini gemellies in golinatinel secondo anno divita [A longitudinal examination of language development in twin and singlet on children during the second year of life]. Et à Evolutiva, 3, 65–79. Longobardi, E., Renna, M. (2010). Riferimenti a stati interni nell’ input rivolto a bambin icon ritardo mentale e con sviluppo tipico [Mental state talk addressed to children with mental delay and with typical development]. Psichiatria dell ’Infanziae dell ’Adolescenza, 77, 87–97.
Clelia Rossi-Arnaud. University Sapienza of Rome, Department of Psychology, Via dei Marsi 78, 00185, Rome (Italy). Current themes of research: Working memory. Gestures. Binding. Suggestibility. Theory of mind.
Narrative writing in school-age children Most relevant publications in the field of Psychology of Education: Hünefeldt, T., Rossi-Arnaud, C., Furia, A. (2009). Effects of in formation type on children’s interrogative suggestibility: is theory-of-mind involved? Cognitive processing, 10, 199–207. Rossi-Arnaud, C., Pieroni, L., Spataro, P., Baddeley, A. (2012). Working memory and individual differences in the encoding of vertical, horizontal and diagonal symmetry. Acta Psychologica, 141, 122–132. Rossi-Arnaud, C., Spataro, P., Longobardi, E. (2012). Effects of pointing on the recall of simultaneous and sequential visuospatial arrays: a role for retrieval strategies? Psychological Research, 76, 699–712.