The organization of molecules into macromolecular (nanometer scale), supramolecular complexes (submicron-to-micron scale), and within subcellular domains, is an important architectural principle of cellular biology and biochemistry. Determining the precise nature and distribution of complexes within the cellular milieu is a challenging biophysical problem. Time-series analysis of laser scanning confocal microscopy images by image correlation spectroscopy (ICS) or fluctuation moments methods provides information on aggregation, flow, and dynamics of fluorescently tagged macromolecules. All the methods to date require a brightness standard to relate the experimental data to absolute aggregation. In this article, we show that ICS as a function of gradual photobleaching is a sensitive indicator of aggregation distribution on the submicron scale. Specifically, in photobleaching ICS, the extent of nonlinearity of the apparent cluster density as a function of bleaching is related to the size of clusters. The analysis is tested using computer simulations on model aggregate systems and then applied to an experimental determination of Aβ peptide aggregation on nerve cells. The analysis reveals time-dependent increases in Aβ1-42 peptide aggregation. Globally, the datasets could be described by a monomer-dimer-tetramer-hexamer or a monomer-dimer-trimer-pentamer model. The results demonstrate the utility of photobleaching with ICS for determining aggregation states on the supramolecular scale in intact cells without the requirement for a brightness standard.