ANALYSIS OF EYE TRACKING IN HIGH SCHOOL MATHEMATICS VIDEO CLASSES
The results of the International Student Assessment Program (Pisa) carried out in 2018 indicate that about 68.1% of Brazilian students have mathematical mastery below the minimum considered for the exercise of their citizenship. Considering the below-expected performance in Mathematics, a reflection on the planning of educational conditions in the context of secondary education from an inclusive perspective was proposed. Thus, the objective of the research was to carry out a double analysis, evaluating a mathematics teaching program for high school level content in front of participants with varied behavioral repertoires, while evaluating the viability of eye tracking as an implicit auxiliary measure in this process. Having as specific objectives to characterize the eye tracking of students with varied repertoires during the performance of a mathematics teaching procedure and to analyze the correlation between the eye tracking metrics with mathematical learning from the content displayed multimodally in the video lessons. To this end, we use an exploratory research methodology with a multiple baseline design between video lesson/mathematical concept, taking as a trigger, the mathematics video lessons produced by the research group Surdos e Libras (SueLi) of the Federal University of ABC (UFABC) . The collection of ocular measurements was performed using the Tobii 4C device. To this end, we divided our research into five stages, namely: 1) literature review; 2) development of the collection and delimitation protocol of our available equipment; 3) selection of 11 participants for characterization, using the Reasoning Evidence Battery (BPR-5) and questionnaires; 4) teaching procedure consisting of an initial survey in the form of fixation exercises and contextualized problems, video lessons on the concepts of multiplication, distributive, greatest common divisor and least common divisor, finally, the reapplication of the initial exercises; 5) treatment and discussion of the results, considering the successes, errors, characterization, fixation time, saccadic movements, number of fixations and Pisa parameters. We then found that the proposed procedure contributed to the participants' learning, revealing a positive impact of the intervention, which lasted an average of two hours (about two to three school classes) given the presentation of an acquisition of concepts, even if initially. We can also identify students' curricular discrepancies in relation to the mathematical content evaluated and taught, as well as highlighting the contribution of teaching planning within a systematized educational practice, in the context of high school