The present study is based on a theoretical framework of cognitive load that distinguishes causal factors (learner characteristics affecting cognitive load e.g., self-concept; interest; perceived stress) and assessment factors (indicators of cognitive load e.g., mental load; mental effort; task performance) of cognitive load. Various assessment approaches have been used in empirical research to measure cognitive load during task performance. The most common methods are subjective self-reported questionnaires; only occasionally objective physiological measures such as heart rates are used. However, the convergence of subjective and objective approaches has not been extensively investigated yet, leaving unclear the meaning of each kind of measure and its validity. This study adds to this body of research by analyzing the relationship between these causal and assessment (subjective and objective) factors of cognitive load. The data come from three comparable studies in which high school students (N = 309) participated in a one-day out of school molecular biology project and completed different tasks about molecular biology structures and procedures. Heart rate variability (objective cognitive load) was measured via a chest belt. Subjective cognitive load (i.e., mental load and mental effort) and causal factors including self-concept, interest, and perceived stress were self-reported by participants on questionnaires. The findings show that a) objective heart rate measures of cognitive load are related to subjective measures of self-reported mental effort but not of mental load; b) self-reported mental effort and mental load are better predictors of task performance than objective heart rate measures of cognitive load; c) self-concept, interest and perceived stress are associated with self-reported measures of mental load and mental effort, and self-concept is associated with one of the objective heart rate measures. The findings are discussed based on the theoretical framework of cognitive load and implications for the validity of each measure are proposed.