Learning as Computational Thinking

• In Event: Learning Under Algorithmic Conditions: Critical Perspectives on AI and Curriculum

Abstract

Objectives
-----------
The charge to teach “computational thinking” (CT) has rallied schools to include “coding”, computer programming, and computer science more generally in every level of education, with a special emphasis on introducing CT early in K-5 education. This renaissance of computer science education via CT stems from Jeanette Wing’s (2006) article for the Communications of the ACM, where she poetically rushes across the range of computer science disciplines and recasts them as essential curriculum for critical thinking. Computational thinking fills a need for effective frameworks to help us understand how experts think about computing and how these frameworks can inform curricula and educational programs. Like all curricular frameworks, however, computational thinking brings with it certain assumptions and agendas about the nature of learning. In particular, CT has been closely aligned with the technology industry, with the economic and political goals of tech giants like Microsoft, Google, Apple, and Facebook setting much of its agenda for schools. This paper examines the stated and hidden curricula of computational thinking by contrasting it with work that helps us examine the political and social dimensions of computing systems. This latter perspective will help us craft an agenda for teaching computer science with a broader purpose than the current regime’s closeness to industry.

Theoretical Framework
---------------------
This paper draws on theories emerging from software studies, platform studies, critical code studies, and related work which help us understand and study computational systems as part of complex learning ecologies. I draw inspiration from Alexander Galloway’s approach to studying the political implications of networked systems (Galloway & Thacker, 2007) and software (Galloway, 2021; Sack, 2019). I look at power dynamics in curricula from a perspective of politics and aesthetics (Rancière, 2004) and the ways these theories can be used for political action (Fuller & Weizman, 2021).

Methods & Data
--------------
The methods in this paper are (1) genealogical, which allow us to identify the historical antecedents that shaped the way we think about computing and the relationship between computers and humans and learning; (2) philosophical, which allow us to analyze computing curricula in terms of its potential impact on learning and society broadly. The data comes from key writings from the early days of AI research (e.g. Minsky, 1970), and examines where computational thinking and software studies continue and depart from these models. I look at specific curricula and standards highlighting the CT model (Code.org, n.d.; Google, n.d.; ISTE, n.d.), as well as alternative textbooks focused on teaching CS as an expressive and aesthetic mode of learning (e.g. Soon & Cox, 2020).

Findings and Scholarly Significance
-----------------------------------
The instrumental formalism of Computational Thinking falls short as a curricular paradigm to help us address the complexity of computational learning ecologies. I show how critical code studies more explicitly account for the power dynamics, political implications, and aesthetic dimension of computing systems in order to guide the curriculum that we require.

Author

• Matthew X. Curinga, Adelphi University