Chain dynamics in macromolecules that occur with slow characteristic exchange frequencies (1-100 Hz) influence the bulk mechanical properties of polymers. In this contribution, we systematically evaluate the change in conformational reorientations of individual polymer backbones once they form a miscible binary blend, using the solid-state CODEX NMR experiment over a 100 K temperature range. The temperature range encompasses the glass-transition points of each pure polymer. The high molecular weight polyisobutylene (PIB)/head-to-head polypropylene (hhPP) polymer blend is known to form an intimate mixture at the 50:50 wt % composition used in this study. Detailed site-specific measurements and analysis of slow polymer backbone dynamics in the pure and blend state reveal that the two blended components converge to a common averaged temperature where slow chain motion is maximized in each chain type with equal correlation times, but unique exchange intensity distributions and activation barriers are preserved for each polymer. Most interestingly, the value of this composition-weighted average temperature is 5-7 deg lower than predicted using the Fox equation, quantitatively confirming our previous assignments of configurational entropy as a miscibility driver in polyolefin blends. Quantitative estimates of the log-Gaussian correlation time distribution models and their temperature dependence are discussed with respect to the activation energies E-a for slow backbone reorientation in pure amorphous PIB, pure amorphous hhPP, and each in the miscible blend; emphasis is directed toward changes in the models induced by formation of the miscible blend relative to the pure components. By direct comparison of data obtained using identical experimental parameters, chain dynamics are relatively more perturbed for hhPP, the high-T-g blend component, than for PIB (-30 K vs +15 K, respectively, for the temperature of maximum slow exchange density relative to the pure material), in agreement with previously published conclusions. Using an Adams-Gibbs model as the most reasonable physical model suggested by the data, we calculate an increase in the configurational entropy S-c of the miscible blend relative to the unmixed components ranging from 11 to 17%. Importantly, all data are specifically extracted from individual signals arising from each polymer in the solid blend, rather than bulk-averaged responses, and without any form of sample modification.