CDIO programs have tenets of self-directed learning and often use either problem or project based learning. The assessment questions usually model real world engineering scenarios using fairly complex questions which are located in the ‘zone of proximal development’ (ZPD) of the students. The efficacy of the CDIO approach is reported in many studies and the approach is emerging as an accepted best practice in the field of engineering education. However, the consequence of the CDIO pedagogical approach on the cognitive load induced in students is not understood. This study therefore aimed to ascertain the amount of cognitive load induced due to the central tenets of the CDIO approach namely, complex questions, zone of proximal development and self-directed learning. The study follows a quantitative research design and a positivist philosophy using a deductive research approach using a cross sectional questionnaire survey and non-probability sampling. Structural equation modelling was performed using IBM SPSS AMOS v25 while descriptive and reliability analysis were done using SPSS v25. The findings show that the use of complex questions yields significant levels of cognitive load and locating the questions in the zone of proximal development of students also induces some amount of cognitive load. Self-directed learning on the other hand does not subject students to significant levels of cognitive load. Several studies have established the detrimental impact of high levels of cognitive loading on learning. The findings therefore suggest that it is necessary and important to monitor and manage the levels of cognitive loading induced by the CDIO approach so that it does not begin to interfere with the learning process. Specifically, the complexity of the assessment problems used should be carefully planned to be appropriate to the knowledge level of the students and not located outside the zone of proximal development of the students.
