Cognitivism is a highly formalised learning theory with a solid scientific grounding on Psychology. As a departure from behaviorism, and its de-emphasizing of internal mental processes in favour of overt behavior, cognitivism sets the focus on how information is received, organized, stored, and retrieved by the mind, in the context of the learning process (Ertmer and Newby, 2013).
Rather than a formal theory, connectivism has been defined as a theoretical framework for understanding learning (Kop and Hill, 2018). With the advent of the Internet, concepts like distributed networks and hypermedia provide a suggestive metaphor to explain how informal learning occurs through communities of practice and personal networks, gaining more importance versus formal learning (Trestini, 2018).
Harvard psychologist Jerome Bruner pioneered a strong body of research that put cognition at the center of children's educational development. In contrast with the alleged mind-blindedness of behaviorists, Bruner emphasized the importance of mental processes in learning, by proposing a 3-tiered system of internal representations: enactive (action-based), iconic (image-based), and symbolic (language-based).
The work of Albert Bandura bridges the gap between behaviorism and Bruner's cognitivism. Although he acknowledges the role of environmental stimuli in shaping behavior, Bandura rejects the idea that human beings are just sounding boards passively receiving and reacting to such stimuli. In contrast, Bandura sees humans as active information processors adopting desirable behaviors that they learn through observation. What is more, the adoption of such behaviors is mediated by cognitive processes like attention, retention, reproduction and motivation (McLeod, 2016).
For Bandura, the notion of self-efficacy, understood as the expectation of one's own performance, is also an important factor determining which behaviors are adopted and how strongly (Bandura, 1977).
Connectivism as a theory of learning is an idea advanced by two separate publications that appeared in Canada in 2005. In "Connectivism: A Learning Theory for the Digital Age", professor George Siemens notices the increasingly fast pace at which traditional knowledge is updated or replaced altogether, thanks to the disruptive effect of the Internet and other distributed technologies enabling instant communication. In "An Introduction to Connective Knowledge", philosopher Stephen Downes introduces connectiveness as an important property of distributed knowledge. Grasping the interaction between separate knowledge entities becomes as important as knowing the properties of each individual entity.
As a result of both approaches, connectivism emerges as a theory built on a metaphor provided by modern information technologies: from this perspective, knowledge becomes a network made of three different types of nodes: neural, conceptual (internal) and external.
For cognitivism, the learner is an active participant in his/her own learning process. Cognitive theories focus on the acquisition of knowledge, i.e. on how the learner processes information by coding and structuring it. Learning happens when there is a discrete change between states of knowledge rather than a change in the probability of a specific response (Ertmer and Newby, 2013).
Instead of focusing on knowledge acquisition, connectivism concerns itself with learning about the relationships (connection) between different nodes of the knowledge network (which can be a peer group or an online database). The role of the teacher is to guide students to information and answer key questions, encouraging them to seek out information on their own.
Learning a modern language in a formal setting can be considered a good learning scenario for cognitivism. By mobilising the three-tiered system of internal representations developed by Bruner (enactive: language as peformative utterances; iconic: vocabulary as an abstract representation of physical phenomena; symbolic: grammar and syntax as procedures to structure real-world experiences), learning a new language solicits the following mental processes: receiving, organizing, storing, and retrieving information. However, there may be a significant difference between learning a modern language like English or French, or a defunct language like Latin, where the enactive component is less emphasized: one usually develops a passive competence to "understand" the Latin from classic texts, but we don't learn to produce new, original Latin texts.
Connectivism can be related to constructivism. A typical constructivist learning scenario where the learner is expected to build his/her own knowledge is analogous to one where the learner taps on the different nodes of a knowledge network (an MKO, the Wikipedia, etc.), as proposed by connectivists. In both cases, metacognition or the ability of "learning to learn" takes centre stage.
Sources:
Ertmer, P. A., & Newby, T. J. (2013). Behaviorism, cognitivism, constructivism: Comparing critical features from an instructional design perspective. Performance Improvement Quarterly, 26(2), 43-71.
McLeod, S. A. (2016, Febuary 05). Bandura - social learning theory. Simply Psychology. https://www.simplypsychology.org/bandura.html
Albert Bandura (1977). Self-efficacy: Toward a Unifying Theory of Behavioral Change. Psychological Review, Vol. 84, No. 2, 191-215.
Kop, R., & Hill, A. (2008). Connectivism: Learning theory of the future or vestige of the past?. The International Review of Research in Open and Distributed Learning, 9(3). https://doi.org/10.19173/irrodl.v9i3.523
Marc Trestini (2018). Modeling of Next Generation Digital Learning Environments: Complex Systems Theory. John Wiley & Sons