Response to selection is fundamental to plant breeding. To gain insight into the genetic basis of response to selection, we propose a new experimental genetic framework allowing for the identification of trait-specific genomic loci underlying population improvement and the characterization of allelic frequency responses at those loci. This is achieved by employing a sampling scheme for recurrently selected populations that allows for the simultaneous application of genetic association mapping and analysis of allelic frequency change across generations of selection. The combined method unites advantages of the two approaches, permitting the estimation of trait-specific allelic effects by association mapping and the detection of rare favourable alleles by their significant enrichment over generations of selection. Our aim is to develop a framework applicable for many crop species in order to gain a broader and deeper understanding of the genetic architecture of response to artificial selection.