Abstract:Background: Reading and listening comprehension are essential for accessing the school curriculum. Inference-making is integral to successful comprehension and involves integrating information between clauses (local coherence) and integrating information with background knowledge (global coherence). We require appropriate methods to assess comprehension and inference-making in order to identify areas of difficulty and provide appropriate support. Aims: Typically developing children's ability to generate local and global coherence inferences was assessed. The effect of text modality (reading and listening comprehension) and presentation format (stories presented in segmented and whole story format) was explored using two comprehension measures (question answering and story retell). The main aims were to determine whether there were advantages for reading or listening comprehension and for segmented or whole text presentation. Methods & procedures: Typically developing children in Year 3 (n = 33) and Year 5 (n = 40) either read or listened to short stories. Their ability to generate global and local coherence inferences was assessed in two ways: answers to inference-tapping questions and story retelling (scored for inclusion of necessary inferences). Stories were presented in either a whole format (all questions after the story) or a segmented format (questions asked at specific points during story presentation); the retelling was always after the complete story and questions had been presented. Outcomes & results: For both comprehension measures, there was developmental progression between age groups and a benefit for the reading modality. Scores were higher for global coherence than local coherence inferences, but the effect was significant only for the question-answering responses, not retells. For retells there was a benefit in presenting the text as a whole compared with the segmented format, but this effect was not present for the comprehension questions. There was a significant interaction between inference type and modality for both comprehension measures (question answering and story retell): for the local coherence inferences scores were significantly greater in the reading compared with the listening modality, but performance on the global coherence inferences did not differ significantly between modalities. Conclusions & implications: Clinicians, teachers and other professionals should consider the modality and presentation format for comprehension tasks to utilize areas of strength and support areas of difficulty. Oral presentation may result in poorer comprehension relative to written presentation in general, and may particularly affect local integrative processing. These findings have important implications for the development of appropriate assessments as well as for supporting children with comprehension difficulties. What this paper adds What is already known on the subject Reading and listening comprehension are critical for accessing the school curriculum and educational success. Inference-making is integral to successful comprehension and involves integrating information between clauses (local coherence) as well as integrating information with background knowledge (global coherence). Children have an awareness of the need to generate coherence inferences, but not all children will generate sufficient coherence inferences for adequate comprehension during text presentation. Existing assessment tools measure comprehension by asking questions after story presentation. This provides an overall indication of comprehension or inference-making ability and can identify children with comprehension or language and communication difficulties. What this paper adds to existing knowledge The study compared coherence inference-making in two presentation conditions: whole format (all questions after the story) or segmented format (questions asked at specific points during story presentation). Children (aged 7-10 years) were assessed in the reading or listening modality. Two comprehension measures were used (inference-tapping questions and story retell). There was developmental progression and a benefit for the reading modality for both comprehension measures. Scores were higher for global coherence than local coherence inferences for the comprehension questions. There was a benefit in presenting the text as a whole compared with the segmented format for story retells. What are the potential or actual clinical implications of this work? The results are important for clinicians and other professionals assessing and supporting comprehension skills. The results suggest that the modality and presentation format of comprehension tasks should be considered to utilize areas of strength and support areas of difficulty. The optimum form of input and structure may depend on a child's individual profile and the skill being assessed or supported. Targeted questions may identify a child's potential to generate an inference. This may assist identification of children who may require more targeted or specialist intervention. The reading modality may provide a means of support for development of verbal comprehension.

Interviews and Inference: Making Sense of Interview Data in Qualitative Research

Exploring inferences across media: an application of Kintsch's framework of inference processes

The interview reconsidered: context, genre, reflexivity and interpretation in sociological approaches to interviews in higher education research

Interpreting hidden meaning in qualitative research interview data: opportunities and challenges

Inferences and Modal Vocabulary

Being explicit about the implicit: inference generating techniques in visual narrative

Epistemology mediating conceptual knowledge: Constraints on data accessible from interviews

To link, to infer, to understand

Sensible Use(rs) and the Construction of Self-Identity in Research Interviews

Psycholinguistic Model of Abductive Reasoning in Fiction Narrative Discourse

Going Forth and Drawing Back: an Intensional Approach in Nonmonotonic Inference.

Linguistic inferences without words

Some Reflections on Drawing Causal Inference using Textual Data: Parallels Between Human Subjects and Organized Texts

Working through Challenges in Doing Interview Research

Inference From Visible Information And Background Knowledge

What Do We Mean When We Say "Insight"? A Formal Synthesis of Existing Theory

Assessment of inference-making in children using comprehension questions and story retelling: Effect of text modality and a story presentation format

Talking Space: inference from spatial linguistic meanings

Intentions and Information in Discourse

What is the Logic of Experimental Inquiry?

11 'Open, 'closed, and in-between: a worked example of and reflections on data-sharing practices in qualitative interview research