Exploring non linear relationship between two variables I did a study measuring the relationship between metamemory and actual memory scores in a test. Basically what I find is that only those people who are good or bad in memory have a accurate metamemory. However, those in the middle of the distrution do not. I have analysed my results with simple correlations (weak but reliable correlation when all the sample is included) and dividing the sample in four groups according to their scores in the objective memory test. Is there any way to do explore this in a different way? 
Thanks! 
 A: So, we have a situation where the people in the middle are harder to assess on memory skills. 
One idea, teach them memory tricks, like constructing a nonsensical sentence/song. Some may have superior learning skills as will be witnessed in subsequent memory tests.
Also, change up the parts of the brain that are employed in recording an event. So, in place of a test on what they have read, play an audio only recording content, or display a short movie clip or photos, and then have them answer questions on the media content.
Have people described their state before testing. Examples include tired, hungry, headache, distracted due to an event, suffering from a cold, ..., some or all of which could impact mental test scores.
Finally, nutrients can improve memory, with a positive effect on some more than others. Before testing again, find out how rigorous they were in consuming the 'memory aids' and their age, which may impact any biological benefit achieved. This idea is akin to tuning up your car, and then assessing its engine performance.
