Bridging the Modality Gap: How Digital Visualisation Supports Syntactic Scaffolding in English Reading Comprehension among Tanzanian Primary Learners

Abstract

Tanzanian primary school learners face persistent challenges in English reading comprehension, particularly in processing complex syntactic structures that differ substantially from their native Kiswahili. This syntactic bottleneck limits their ability to progress beyond literal comprehension to inferential and critical reading. Despite the promise of digital visualisation, the specific mechanisms through which it supports comprehension remain undertheorised, and existing research has not examined differential effects across comprehension levels in African educational contexts. This study investigates how dynamic digital visualisation supports English reading comprehension among these learners. It examines differential effects across literal, inferential, and critical comprehension levels and proposes a Visual-Syntactic Scaffolding (VSS) Model as an inferential framework. A mixed-methods case study was conducted in ten purposively selected primary schools in Morogoro Municipality, Tanzania, involving 100 Standard VI pupils and 90 English language teachers. Purposive sampling was adopted because the intervention required schools with reliable electricity, basic digital devices, and willingness to participate in a 20-week intervention. The 20-week Playcast intervention combined dynamic video, colour-coded syntactic formatting, interactive features, and embedded comprehension questions, with structured teacher training and ongoing support. Data were collected through pre- and post-intervention comprehension assessments, teacher surveys, classroom observations, and semi-structured interviews. Quantitative data were analysed using paired *t*-tests and effect sizes; qualitative data were analysed thematically. The intervention produced significant overall comprehension gains (*d* = 0.81), with strongest effects at literal (*d* = 0.87), moderate at inferential (*d* = 0.54), and weakest at critical (*d* = 0.22) levels. Teacher data indicate high intervention acceptability, pedagogical feasibility, and implementation fidelity. Lower proficiency students relied on visual cues for meaning construction; higher proficiency students used visualisation for verification. The study measures global comprehension outcomes rather than real-time psycholinguistic processing; the VSS Model is presented as an inferential framework requiring validation through experimental designs with eye-tracking and think-aloud protocols. The VSS Model explains how dynamic visualisation may support comprehension through three mechanisms: syntactic bootstrapping, cognitive load reduction, and external mental model construction.