Date of Award

Spring 2023

Document Type

Open Access Dissertation



First Advisor

Paul Malovrh


The strand of feedback research within the field of Instructed Second Language Acquisition (ISLA) examines the effects of feedback on the development of second language (L2) knowledge and learning behaviors. While findings often make claims about the relationship between feedback and language processing, much of this research has not measured learners’ concurrent responses to feedback. By and large, the effects of feedback have been inferred from test scores collected after feedback is provided, with few studies analyzing how feedback affects L2 learning behaviors in real time (i.e., during a task). Consequently, we know little about how feedback affects language processing.

Moreover, researchers have traditionally investigated the effects of oral or written classroom feedback in learners of single instructional level. However, the use of computer-assisted language learning (CALL) technology over the past twenty years has changed language curricula and the ways in which many students are instructed in the L2, e.g., online/hybrid courses, language learning applications, etc. (Leow & Cerezo, 2016). Research on CALL has found that computer-delivered instruction and feedback are not only effective at teaching complex grammatical forms but proffer unique advantages related to the focused and individualized nature of instruction (Cerezo, 2012; Heift, 2010). Moreover, CALL offers an analytic framework that allows us to isolate the effects of feedback in real time (Leow & Cerezo, 2016).

The present study extends the CALL line of research and investigates the effects of computer-delivered feedback on on-line processing and accuracy in the acquisition of the French causative, or, faire causatif. A cross section of L2 leaners (N = 80) with different levels of prior L2 knowledge received instruction on the faire causative and then practiced with the form using a self-paced reading (SPR) task, during which they received different types of feedback. They then completed a picture matching recognition test and a controlled production test immediately following instruction and four weeks later. Mixed-effects models were run to investigate the effects of feedback, prior knowledge, and transitivity of the verb phrase on processing and accuracy. Simple-pairwise comparisons and non-parametric tests were also conducted within prior knowledge and feedback groups to investigate simple effects within groups.

Results revealed that feedback had a limited effect on the processing (p < .05) and delayed test scores (p < .05) of participants with intermediate prior L2 knowledge but did not impact processing and accuracy across all participants. Results also found main effects for prior knowledge wherein those with intermediate prior knowledge had significantly longer processing (p <.001) and significantly more retention (p < .05) of the causative compared to participants with low and high prior knowledge. Finally, the present study found that learners with high prior knowledge processed transitive verb phrases more slowly than intransitive (p < .05) and that there was overall processing cost for transitive over intransitive items across all leaners (p < .001).


© 2023, Lesley Erin Smith

Included in

Linguistics Commons