The Challenge In the age of online learning, what are the best instructional approaches to build learners' second language confidence? Does videoconferencing between peers offer the same benefits as conversing with native speakers? This study investigates the impact of the interlocutor in videoconferencing on learners' second language confidence development.

Now more than ever before, language learners can autonomously engage with the target culture beyond the classroom through international television shows, online forums, video clips, and study abroad. Still, much of the literature has deemed the target-language dominant speaker (TLDS) as a key source of cultural knowledge, and, thus, the principal aim here is to ascertain how paid TLDSs and student peers are perceived in terms of their cultural expertise and whether these perceptions affect learners’ confidence development over time. This research is guided by the functional model of second language confidence (FML2C), which integrates the contact space dimensions of richness and self-involvement, and novel to this study is cultural depth. In terms of procedure, 32 second language learners of Spanish carried out eight 30-minute videoconferences over 12 weeks, four with a peer and four with a TLDS on Talk Abroad. In terms of the FML2C, the results revealed that the TLDSs were considered to afford significantly more richness and cultural depth than peers. In spite of this, longitudinal gains in learner confidence were indistinguishable by interlocutor type, suggesting that videoconferencing over time is beneficial, no matter the interlocutor with whom it is realized.

Despite the growing number of heritage language (HL) learners in language courses, the interaction between second language (L2) and HL learners has not been fully explored in the research, and thus, it still remains unclear whether mixed (L2-HL) or matched (L2-L2, HL-HL) dyads are more beneficial for language learning. This study investigates whether the three possible dyad types differ in terms of the form-focused episodes (FFEs) that arose while learners worked together, as well as learning gains as a result of the interaction. Research questions were addressed by analyzing the audio-recorded interactions of eight L2-L2, eight HL-HL, and eight L2-HL dyads engaged in a collaborative writing task, as well as the texts produced by the learners. Learning gains were assessed through the incorporation of linguistic information from FFEs in immediate and two-week delayed post-treatment individual writing tasks. There were no differences regarding the total number of FFEs, but dyads differed with respect to the linguistic focus and resolution of FFEs. Moreover, learning gains were greater for matched dyads than for mixed dyads, suggesting that working with a peer of similar linguistic background and proficiency might be more beneficial for both L2 and HL learners.

Over the past few decades, extensive research has investigated the impact of technology in the second language classroom; however, studies are scarce on the role of digital tools for the teaching and learning of a heritage language (HL). The present pilot study aims to fill this gap by examining the effect of two interlocutor types in videoconferencing, a trained target-language speaker on Talk Abroad and an HL peer on Zoom, on HL learner confidence development over time. Addressing this question is critical given that motivating HL learners and validating their linguistic and ethnic identities is considered cornerstone for HL maintenance (Sánchez-Muñoz, 2016). Theoretically, the present research is guided by the functional model of second language confidence proposed by Sampasivam and Clément (2014). In this study, the participants were students enrolled in a fifth-semester Spanish conversation course, who, throughout the duration of a 16-week semester, completed eight 30-minute videoconferences, four with each of the two interlocutor types. The analysis of the questionnaire data revealed that both interlocutor types, but most notably the HL peer learners, facilitated confidence gains over time. These results are discussed with an emphasis on their pedagogical and theoretical implications.