This study examined the effects of task conditions on L2 listening performance. According to the Performance System of the Language Alignment Model, performance is the result of the degree of alignment between environmental demands (e.g., text and task characteristics) and individual abilities (e.g., proficiency level), with better comprehension reflecting closer alignment. From this perspective, listening support facilitates comprehension because it lowers the environmental demand of the task to better align with listener abilities. However, inconsistent findings in the empirical literature suggest that listening support may reduce, increase, or have no effect on demand. To address this inconsistency, 77 Chinese senior high school EFL learners were randomly assigned to two groups—average-speed group (N = 33) and slow-speed group (N = 44). Both groups completed listening tasks across four conditions—repeated input, vocabulary pre-teaching, combination repeated input and vocabulary pre-teaching, and listen-once. Results from within-groups analysis of variance showed that performance was significantly better under the repeated input condition than any other condition at both speeds. The combined condition performance was equal to the baseline listen-once condition performance and better than the vocabulary pre-teaching condition for both speeds. The vocabulary pre-teaching condition performance was weaker than the other methods and equal to the listen-once condition only for the slow-speed text. Results further revealed no differences in performance between the speed groups for any condition. These results indicate that repeated input reduced environmental demand, pre-teaching vocabulary increased demand, and the combined condition offset the demand to match the baseline.