This study investigated the effectiveness of a usage-based grammar instruction approach compared to a rule-based approach for Iranian EFL learners, addressing a gap in research comparing these methods. Specifically, it examined whether Langacker's Cognitive Grammar, a usage-based model, yielded greater improvement in past perfect tense acquisition than the rule-based methods presented in Iranian high school EFL textbooks. Thirty-nine high school students participated in a quasi-experimental study using a cluster sampling method. Participants were assigned to either a Cognitive Grammar instruction group or a control group receiving rule-based instruction. Grammatical knowledge was assessed via pre- and post-tests. Results, analyzed using an independent samples t-test, indicated that both approaches improved learners' grammatical competence. However, the usage-based approach demonstrated a statistically significant advantage, with a large effect size (Cohen's d = 0.91). These findings suggest that Cognitive Grammar may be more effective for teaching grammar to EFL learners. While supporting the implementation of usage-based approaches, especially Cognitive Grammar, the study acknowledges limitations due to sample size.
