Arctic modes of atmospheric variability, including the North Atlantic Oscillation (NAO), East Atlantic Pattern (EA), and Scandinavian Pattern (SCA), explain a large fraction of interannual climate variability in the Northern Hemisphere. These modes are also hypothesized to play a role as drivers of ocean heat transport variability. Here we explore the temporal variability associated with the NAO, EA, and SCA in pre-industrial control and future scenario simulations archived as part of CMIP6. We characterize the three modes using empirical orthogonal functions (EOF) for overlapping 100-year periods of time within simulation output for several CMIP6 models. The NAO, EA, and SCA can be estimated as the first, second, and third EOFs of sea level pressure, respectively, over the North Atlantic region. We compare these conventional EOF calculations to using Common Basis Functions (CBF), which avoid some common issues associated with the conventional EOF approach, such as arbitrarily assigned and inconsistent sign, or poorly separated EOFs which complicate comparison with observations. We evaluate fluctuations in the spatial patterns of the EOFs by tracking movements of the centers of activity for each EOF over time. We find much larger spatial variability when conventional EOF analysis used compared to the use of CBF. We also examine variability in principal component (PC) time series by comparing the intensity of variance in PC.