Ahura – Auto Assistant Tutor

Identifying lack of understanding of university students in online learning environment

‘Ahura’ is an auto assistant tutor who assists university students when they are confused and it can bring up information to them and help them during their online lecture. It also reports the educational state of the online students to teacher to help instructors adjust the speed of the class.

Challanges

This project has been conducted to learn about the deep meaning of educational emotions, and explore different ways to identify students’ lack of understanding in online learning environment. This project is significant for three reasons: firstly due to the global pandemic and forcing education online, secondly because of the lack of experience of students and teachers, and finally the nature of online teaching has brought limitations on communication between students and teachers. There is a high possibility of less engagement from both teachers and students in online class activities, and it leads to lack of support from teachers, and understanding of students (Arguel et al., 2016).

Emotion Detection Methods with Technology

In face-to-face class settings, dealing with confusion is much easier than online learning because instructors can observe physical and verbal cues for interventions to induce or to manage the confusion level (Goleman, 1995). HenceHowever, detection of confusion in remote learning is more difficult to assess confusionand to adjust the learning activity in order to maintain effective study situations (Calvo et al., 2014). For this reason, numerous researchers have targeted to developing methods to capture confusion in a learning environment.

Speculative Design

What if a solution to this actually exists? What is the impact of technology detecting and recording human emotions?

This study applies speculative design as an approach to feel, listen to, and taste different versions of the future which in this study is different ways of detecting educational emotion in online learning (Sterling, 2009, Dunne & Raby, 2013). The main focus of the speculative design is to force an aspect of the future into the present so it seeks responses (Tonkinwise, 2014). Speculative design is different from design thinking and social design. Design thinking is about solving problems and social design is concerned with complex human problems. The speculative design strives to overcome the invisible wall separating imagination from everyday life, blurring the difference between the real “real” and the real “unreal”. The former is in the here-and-now, the latter exists on screens, books and locked in people’s imaginations. It is not about a space for experimenting with how things are now, making them better or different, but about other possibilities altogether (Dunne & Raby, 2013).

Interviews

Seven interviews were selected from case study_ MDes staff and students at QCA University__ for semi-instructed interviews. All interviews were kept anonymous. Four staff and three students have been interviewees for this research. There were three interviewees in the age range of 32-38 years, two in the range of 25-31, and one each in the range of 45-58 and 59 & above. Of the total, six interviewees were female, and one was male. Four out of seven interviewees live in Brisbane, two in Gold Coast. All the interviewed staff and one the student have online teaching experience since covid happened, one each student has less than one year and from her bachelor.

Persona

Ideation

The ideation process was sparked by a review of the accessible data accessible from various sources. The data were used to create different scenarios for recording students’ lack of comprehension during a live lecture. If we brainstorm and study a variety of ideas, we have a better chance of discovering the optimum design approach (rather than just one). As a result, when all of the ideas converged on recognising the state of students in an online classroom using modern technology. Researching existing solutions and constructing a mood board aided in the brainstorming of innovative solution ideas that might be related with the desired target audience (Harley, 2017, para. 11)

Auto Assistant Tutor – Students’ side

While a remote lecture is taking place, the sensor captures the status of each student. The auto tutor will begin delivering hints based on the current lecture if the Emotional Detector module senses that a student has been puzzled for a period of time. If the auto tutor detects frustration, it suggests halting the current lecture and reporting to the instructor whatever part of the presentation is difficult for the student to understand based on the data received from that sensor. Accepting or refusing the auto tutor’s invitation to pause the present lesson is up to the learner. If the learner accepts, the auto tutor matches topics based on the student’s talents and learning style using topic selection, intervention modules, and curriculum scripts. It also alerts the human instructor that the student is continuing the lecture with the auto tutor and will return as soon as their status is reverted to positive confusion and the current lecture level is reached (the auto tutor listens in on the human tutor session). If the student remains frustrated after the auto tutor intervention, and the auto tutor discovers that the student is also bored, the auto tutor immediately alerts the human tutor, who then performs the affective intervention and assists the student.

Auto Assistant Tutor – Teacher’ side

The auto assistant tutor works as an assistance to help the online lecturer for the duration of the class (See Figure 26) by supporting them in knowing how the students are feeling during the lecture. The tutor is informed to undertake a human intervention if the majority of the class remains confused for an extended period of time. It also notifies the teacher, based on sensor data, which part of the lecture the students found the most confusing.

User Testing

The auto assistant tutor works as an assistance to help the online lecturer for the duration of the class by supporting them in knowing how the students are feeling during the lecture. The tutor is informed to undertake a human intervention if the majority of the class remains confused for an extended period of time. It also notifies the teacher, based on sensor data, which part of the lecture the students found the most confusing.

Iteration

To overcome these weaknesses and limitations, new ideas were developed around how the online classroom could look like. At this stage of brainstorming regarding the feedback of the initial idea and primary research, physical white boards were picked up to obtain a smooth flow of thinking. After this stage, results were transformed to Miro board to communicate the idea in a better way. In this brainstorming, it started from where this advanced technology can help if it existed. For branding, this advanced technology was called Ahura (Persian name means ‘wise lord’) and it is described in details in the branding section. The four areas of the Ahura mentorship program were respectively described,: psychology, education, children and finance. Each of them has different prices and Ahura offers a free trial as well.