Cognitive load researchers have used varying subjective techniques based on rating scales to quantify experienced cognitive load. Although it is generally assumed that subjects can introspect on their cognitive processes and have no difficulty in assigning numerical values to the imposed cognitive load, little is known about how visual characteristics of the rating scales influence the validity of the cognitive load measure. In this study we look at validity of four subjective rating scales (within groups) differing in visual appearance by participants rating perceived difficulty and invested mental effort in response to working on simple and complex weekday problems. We used two numerical scales (the nine-point Likert scale most often used in Cognitive load theory research and a Visual Analogue Scale ranging between 0–100%) and two pictorial scales (a scale consisting of emoticons ranging from a relaxed blue-colored face to a stressed red-colored face and an “embodied” scale picturing nine depicted weights from 1–9 kg). Results suggest that numerical scales better reflect cognitive processes underlying complex problem solving while pictorial scales Underlying simple problem solving. This study adds to the discussion on the challenges to quantify cognitive load through various measurement methods and whether subtleties in measurements could influence research findings.