I had an idea for a nice little experiment that could be enlightening in several ways about language. The experiment essentially weaves together 3 different works I have been reading. On the one hand, in the study of sound in language, phonology, there is a distinction often made between a phonological process and a morphophonological rule. Those are the terms of the school called Natural Phonology, which my advisor happens to adhere to. Other theories make similar distinctions under different names. What they are talking about is systematic, constant changes in speech due to pressures of articulation and hearing versus memorized rules of sound change. So a classic example of a process is an /n/ becoming an [m] before another labial consonant. (Trust me, or ask me and I will explain.) Making this change makes the sequence of sounds much easier to get out. A classic rule, on the other hand, is opaque becoming opacity. Notice how the k sound at the end of opaque becomes an s sound in opacity. This is no easier to say. It is just a rule of English we have all memorized - at least those of us who say opacity. This distinction is item one. I was wondering all last semester if there was an experimental way to verify a neurological distinction between processes and rules.
Then I've been reading Joan Bybee's Phonology and Language Use, which I love and will purchase soon. She states several times that phonology should be thought of as procedural memory - especially phonological proceese, but perhaps not rules. She nicely even states that if she is right, then phonology should behave with similar characteristics to other bits of procedural memory.
The final step is that I have been glancing through Eichenbach and Cohen's From Conditioning to Conscious Reflection. It's kind of a cognitive neuroscience look at memory. They have a discussion of procedural memory and note on the first page that procedural memory has been known for decades now to reliably follow the Power Rule of Practice.
I want to have subjects practice and learn a phonological process. Do they exhibit the Power Law of Practice as they learn to say the sequence? If so, this is good evidence that procedural memory is in use. Do they do the same when acquiring a rule? If not, then this is good evidence that the distinction is real. If the experiment can be simple enough, it is potentially a very useful tool of phonological analysis.
I hope I can work on designing this, if it's a good idea and not done already, in the Spring.