Basis of the study

A. Previous research

A number of studies have been carried out around the world on the issue of access to digitization for students with visual impairment.

Mahanta, Abhijeet and others studied a voice assistant designed to enhance visual perception for students with visual impairment. This Android application helps persons with visual impairments move around and live more independently by capturing input data onto a smartphone in real time, for example reading a restaurant menu or helping the user to make a hotel reservation or find their personal belongings. It also recognises objects in the user’s surroundings, and uses a voice control mechanism to provide feedback, allowing the user to perform a range of tasks by voice command.[1]

Shneha Padma and others studied how AI can improve the lives of persons with partial visual impairments and alleviate difficulties in their daily lives, particularly when they encounter unexpected situations as a result of entering unfamiliar environments which can expose them to safety risks. For example, some modern AI systems can detect an approaching individual and, if that person appears to take an attack stance, alert the visually impaired person to the danger and direct him or her to take appropriate action. This study also touches on artificial emotional intelligence, also known as ‘emotion recognition’ or ‘emotion detection’ technology, which is built using AI techniques such as image processing, deep learning, voice output and voice recognition, and uses big data to store images and sound files. According to the study, major companies such as Microsoft, Facebook and Accenture are making efforts to help persons with disabilities (including those with visual impairments) to use AI[2] to improve their lives and live independently in society.

According to Samigulina and others, innovative intelligent systems should be developed to help persons with visual impairments engage in distance learning. To that end, they propose a comprehensive AI‑based approach to create an accessible learning environment and identify the intellectual, physical and psychological characteristics of digital and informational cognition and awareness, based on the cognitive approach.[3]

With regard to Morocco-based research, Mustafa Orhay and others conducted a study entitled, “Teaching at the Institutions of Cadi Ayyad University During the Covid-19 Pandemic. Stage 1 Research: Pedagogical Continuity in Distance Learning”, which examined the role of distance learning, a teaching model that emerged as the only means of ensuring pedagogical continuity, in tackling biases in the use of technology. The Centre for Pedagogical Studies, Evaluation and Research at the university carried out a study which aimed to analyze the distance learning approach, identify its strengths and weaknesses and address them by adopting appropriate administrative and educational practices. The study was accompanied by a survey which focused on the three main factors involved in pedagogical continuity, namely teachers, students and institutional measures. Questionnaires were used to collect the views of research professors and students, and interviews were conducted with institutional and management officials. The study concludes with a series of recommendations for institutional managers, universities, teachers and researchers.[4]

B.  Concepts

This study uses the concepts of ‘digitization’, ‘artificial intelligence’ and ‘students with visual impairments’ as defined below. Definitions of other concepts are provided as they occur throughout the text.

1. Digitization

According to Terry Kenney, digitization is the process of using automated software to convert a variety of different information sources (such as books, periodicals, images and sound recordings) into machine-readable format using a system of bits – the basic unit of information used in any computer-based information system. The process involves using specialized devices and technology to convert information into a set of binary digits.[5]

2. Artificial intelligence

The term ‘artificial intelligence’ (AI) was coined at the Dartmouth Summer Research Project on Artificial Intelligence in 1956, and can be defined as “the science of making machines do what humans do with a certain degree of intelligence”. Aimed at using machines to solve certain problems more effectively than humans can, AI is now used in many different fields, including human and social sciences, information technology (IT), mathematics and cognitive sciences. It is not merely a field of research, but an entire approach geared towards the ambitious goal of understanding and simulating human perception and creating cognitive processes similar to those used by humans.[6]

3. Students with visual impairments

This term refers to students who have lost their sense of sight either completely[7], in which case they read Braille, or partially, in which case they can read written material after making certain adjustments, for example by increasing text size (either automatically by zooming in, or manually using a magnifying lens). The medical definition, which is based on visual acuity, does not account for static information relating to how an individual performs tasks in a given social environment. It has therefore been necessary to come up with an educational definition of visual impairment, which focuses on how those individuals acquire language and knowledge.[8] In the context of education, a student with complete visual impairment is defined as one whose disability prevents them from learning through conventional methods, and who therefore requires educational materials, teaching methods and the general university environment to be specially adapted.[9] According to the functional definition, a person with complete visual impairment is someone whose blindness is so severe that they can only read using Braille.

This study offers a practical definition of students with visual impairment, based on the aforementioned medical, educational and functional definitions, and taking into consideration the situation at Cadi Ayyad University, which forms the field reference and organizational framework of this study, and serves as the selection pool for the sample group.

D.  Problem

Having reviewed the aforementioned studies, which highlight the challenges encountered by students with visual impairments in the context of digital learning, especially since educational institutions moved towards digital formats as a result of the Covid-19 pandemic, the main problem for researchers is to devise procedural mechanisms to enable students with visual impairments to engage in AI-assisted digital learning.

E.  Context

This study will assess how much access students with visual impairments have to digital learning and how well they are included in the education process at Cadi Ayyad University, particularly at its open access faculties in Marrakesh, such as the Faculty of Letters and Human Sciences, Faculty of Arabic Language, and Faculty of Judicial, Economic and Social Sciences. The study focuses on the efforts made by those faculties to include students with visual impairment, who constitute one of the most vulnerable social groups, with the highest rates of illiteracy and university drop-out.

Universities are key components of society and represent one of its most important institutions, since they are bound up in an integrated and structured system of social norms which are used to uphold basic values of society. They serve an integrative function, by empowering students and equipping them with a set of cognitive and technical skills that they need to take part in education, work and society.

F.  Methodology

This study was conducted based on the methodology of semi-guided interviews with students with visual impairment. The field work was carried out between 12 and 24 February 2022.

The study uses the research technique of purposive sampling, which refers to “a sample chosen intentionally and specifically because it is believed to truly represent the target population group. This is a type of non-probability sampling, whereby the researcher selects certain individuals on the basis of their specific characteristics and disregards all those who do not have those characteristics. The resulting sample thus has all the necessary features to form the appropriate research base. Purposive sampling – also known as intentional sampling – is essentially used to enable the researcher to select a sample that is fully representative of a community that has specific and known characteristics, rather than using simple random sampling methods”.[10]

The study involves 14 out of a total 30 students (both male and female) with visual impairments at Cadi Ayyad University in Marrakesh, listed as follows:

Source: Field research, February 2022.

According to well-known principles in sociological research, the required sample size depends on the type of methodological and analytical approach used, and on the process of “saturation”. According to Guest, Bunce and Johnson, saturation occurs after a maximum of 12 interviews with sample units, and the core themes begin to emerge after the sixth interview, especially in a narrow field of research with a homogenous target population. In the context of this field research, saturation was reached after 14 full interviews, at which point the information became repetitive.[11]

[1]       Mahanta, Abhijeet and others (April 2020). Application for the Visually Impaired People With Voice Assistant, International Journal of Innovative Technology and Exploring Engineering (IJITEE), ISSN: 2278-3075, vol. 9, issue 6.

[2]       Shneha, Padma, Reddy, Prathyusha and Megala, V.M. (2018). Artificial Intelligence for Vision Impaired People, International Journal of Latest Trends in Engineering and Technology, Special Issue April 2018, pp. 031–036.

[3]       Samigulina, Galina and others (2017). Innovative intelligent technology of distance learning for visually impaired people, vol. 7.

[4]       Orhay, Mustafa and others (2021). “Teaching in Cadi Ayyad University institutions during the Covid-19 pandemic. Research stage 1: pedagogical continuity in distance learning”. Education, Training and Society Journal, vol. 1, April 2021. Deposit no. PE00102021.

[5]       Nabil, Aknoush (2010). Algerian University Digital Library. PhD, library science, University of Constantine 1 (formerly University of Mentouri), p. 241.

[6]       Rouini, Dr. Abdelghani (2020), Utilisation de l’Intelligence Artificielle dans le Domaine des Sciences Sociales, Laboratory (LAADI), University of Djelfa, vol. 5, Issue 17, pp. 218–229.

[7]       Obaid, Magda al-Sayed (2016), Teaching People with Special Needs: An Introduction to Education, p. 142.

[8]       Abdel Rahim, Fathi al-Sayed and Halim Saeed Bashay (1982), The Psychology of Children with Disabilities, Kuwait Publishing House, vol. 2, p. 54.

[9]       Zeitoun, Kamal Abdel Hamid (2003), Teaching People with Special Needs, p. 297.

[10]     Al-Wez, Mohammed (2017), Methodological Techniques: An Introduction to the Sociological Approach, vol. 5, pp. 65–66.

[11]     Guest, G., Bunce, A. and Johnson, L., “How many interviews are enough? An experiment with data saturation and variability”, Field Methods, 18 1, 2006, pp. 59–82. In Ahjeij, Hassan and Fazza, Jamal (2019), Textbook on Qualitative Research in Social Sciences: Theory and Practice, first edition, pp. 65–62.