Calculating p value for non-linear fit

Matlab documentation has the following on the calculation of the p statistics, implemented in the corrcoef function:

"The p-value is computed by transforming the correlation to create a t statistic having n-2 degrees of freedom, where n is the number of rows of X. The confidence bounds are based on an asymptotic normal distribution of 0.5*log((1+R)/(1-R)), with an approximate variance equal to 1/(n-3)."

I'm sorry, but the above text is Chinese for me. If I fit a set of data points [x y] to an arbitrary function, calculating the R squared is very easy. But how do I go about adapting the ttest so that I find the probability that the fit equation actually accounts (R squared), say, 0.9 (90%) of the variance in the data?

Thanks in advance!