A dull echo of bad practice in teaching...

There are many things in language teaching theory that are hotly debated, but there are some things that are universally accepted.  In theory.  In practice, they can be forgotten about.  I'm currently working through the Michel Thomas Polish Foundation course and one of these springs to mind:

The echo effect

The echo effect is quite simple: the last thing you hear stays in your mind longest.  The theory around this varies as our understanding of the human brain improves, and some people talk about "echoic memory", others about "feedback loops", others still "working memory".  But whatever the theoretical models people come up with, they all seek to model the same universally agreed observation: the last thing you hear stays in your mind longest.
The echo effect in practice

So that's the theory, but how does this work in practice?  The canonical example would be the listening exam.  A sentence or passage presented in a listening paper will be followed by silence -- all instructions come before the passage so that the internal echo is the actual material, not the instructions.  After all, if the instructions are clear, the student should understand and internalise them easily.

Failure to follow through to the classroom

However, you will find that some teachers don't consciously consider the echo effect in their day-to-day teaching.  Instead, they try to follow a natural order for language.  The reason this example is based on an MT course is that it's the nearest most lay people get to being able to observe a language class.

Let's look at a couple of quotes from Jolanta Cecula's MT course:

"I'm sorry, but I don't quite understand what you are saying"... talking to a man? (CD3 Track 2)
Notice here that the background information, "talking to a man" comes after the sentence to be translated.  This means that "talking to a man" is in echoic memory, rather than "I'm sorry, but I don't quite understand what you are saying".  This makes the task harder in a way that is of benefit to the learner.

Two tracks later, we get this:
"Can you help him", meaning to him, asking a woman? (CD3 Track 5)

Here we have two pieces of background information coming after the material that should really be in echoic memory.  The learner then has to expend effort on recalling the prompt, distracting from the task of producing the desired target language.

Hi-ho, hi-ho, it's off to work(ing memory) we go...

But the problem of prompt wording goes beyond the simple echo effect, and into bigger questions of language processing.  On a few occassions, the course has prompts of the form:

So what would "" be? / So "" would be...?

Here we make life harder for working memory by interrupting the simple prompt with the phrase for translation.  Processing the interrupted clause further distracts our working memory from the target translation, and makes the task unnecessarily difficult.

What I strive to do in class is to make sure the students know what is expected of them with the minimum of prompting.  In the case of teaching-by-translation, MT-style, I would start a session with some explicit prompting, but quickly move to just giving them the target phrase with no other prompting.

Let them concentrate on the language, not on the classroom