Resumen:

Libraries of stellar spectra are fundamental tools in the study of stellar populations, among other applications. Theoretical libraries -- i.e. model stars -- have been extensively used to cover gaps left by empirical libraries, such as regions of the Kiel diagram not populated by nearby stars. Yet, they suffer from modelling inaccuracies. So should we use or not use model stars in integrated stellar populations? To study this question, we build a synthetic stellar library mimicking the coverage of an empirical library, and stellar population models tailored to these investigations. We learned that uncertainties in stellar population colours are dominated by the lack of coverage in the Kiel diagram; the effects on spectral features indices are multiple and defy simple conclusions; galaxy ages are virtually unaffected by the choice of the library, but are underestimated when libraries with limited Kiel coverage are used; metallicities are underestimated when using synthetic libraries. With current models and observation of stellar libraries, there isn't a unique answer, and the right choice will depend on the science case.