Heift, feedback in CALL, and 'uptake'

Heift, T. (2004). Corrective feedback and learner uptake in CALL. ReCALL, 16, 416-431.

Being that this study was conducted several years ago, it's aim at the time was to help fill the gap in (the dearth of) research on corrective feedback and CALL (computer-assisted language learning). Participants (177 beginning, high-beginning, and intermediate students of German at three Canadian universities) engaged in online grammar activities that supplemented their regular class sessions over the duration of one semester. The online exercises involved three types of feedback: metalinguistic, metalinguistic + highlighting (of the error), and repetition + highlighting. (This last category is a little vague, since 'repetition' is actually a broad category prompt, such as "grammar", which helps students identify the type of error they committed.)

The goal of the study is to discover which type of feedback leads to higher instances of uptake, here defined as any attempt by a student to correct his or her mistake. (Note that students always had the option to skip ahead to the next exercise without making any correction whatsoever.) Results show that the metalinguistic + highlighting is "most effective at eliciting learner uptake", though not in a statistically significant way. Additionally, the two learner variables of student gender and language proficiency did not have a significant effect on the results.

OK, now that that's out of the way. This was an interesting study to read, personally, as I'm also fiddling around in this very same area. The organization of the article was clear and the statistics and charts all very comprehendable. What raises my hackles, though, is the central question this article is asking. While there is value in showing that students prefer or attend to one type of feedback over another (and only three types of feedback were studied here), in the end I wind up asking myself, "So what?" — especially when the definition of "uptake" means merely attempting to correct a mistake when the computer is telling you, 'Hey, you made mistake.'

Personally, I wanted to see what kind of long-term uptake occurred, but that was not of immediate interest to the researcher. I also kept asking myself what the value of being told 'you made a mistake with the past participle' is when students aren't asked to do anything further with that mistake other than type in something else and have the computer check the answer. Sure, it beats what is possible in a workbook, but I'm skeptical of how much real uptake is happening here. I would have much preferred to see how this kind of explicit feedback stacks up against an implicit variety where students have to judge whether the meaning of what they've said/written is interpreted by a 'listener' as what they meant to say. Hey, wait a minute: that sounds an awful lot like what I've been tinkering with myself... I just have little faith that rewriting a word because you've been told the form is wrong leads to anything substantial in the way of SLA. I may be wrong.

Should we read this for class? Probably not. It was good for me and what I'm studying, but it doesn't have a lot of class-wide appeal, I'm guessing.