How a Cockpit Remembers Its Speeds
\cite{Hutchins95}

Cognitive science normally takes the individual as unit of analysis. This study takes as primary unit of analysis a distributed, socio-technical system. The following analysis of memory tasks will show that the cognitive properties of such distributed system can differ dramatically from the cognitive properties of the individuals who inhabit them. What is new in this approach is the study of the way representations forms in the participant to the cognitive system and how these representations span through the system. The author shows that the same framework of the classical cognitive psychology can be applied to the cognitive system as a whole. The author then describe the cockpit system and in particular the task of remembering the speeds, from three different point of view: (a) a procedural description; (b) a cognitive description outside the pilots; (c) and finally a cognitive description inside the pilots. The author states that collaborative learning is completely different from individual learning because the cognitive processes are mixed. The function of the speed bug the engineers had in mind whilst designing the system was just a memory reinforcements for the pilots. The cognitive analysis revealed that the speed bugs translate and make redundant that information into a visual alignment esteem. This activity became, in fact, a combination of recognition, recall, pattern matching, cross modalities consistency checks, construction, and reconstruction. Technological devices introduced into the cockpit affect the flow of information.

——————————————————-
Doctorant meeting

Constructionist approach and Behavioural learning theory are on two different sides of the learning paradigm. Eventually they moved into instruction design which is based on the concept of tutoring or teacher assisted learning.
Social cognition, on the other hand is taking into account the group or the interaction among peers as the right framework (emergent communities) for CSCL. Again, SC is quite open framework. CSCL scripts move back from this openness to instruction design placing the learning process in a sort of scaffolding environment. Socio-constructivist approach is in between the constructionist and the social cognition theory (Learning by collaborating on problem solving). Theory apply at different scales. Experimental group methods do not work always in all context: we do not want to go into ethnographic methods or sociological methods because it relays on less valuable paradigm. One of the Good method is the one by Yrjo Engestrom (\cite{Engestrom93}). In measuring learning we need to find context in which things can be counted. Cognitive learning can be a way to get into the cognitive individual learning. Collaborative learning can be, at the end, assimilated to individual learning. We can use some phenomena and mechanism we observe for a singular individual to a group. Group characteristic can be original expression of the group learning but at the end they must be described as simple chunks pertaining to individual learning. A group can be described as distributed collaborative system (P. Dillenbourg): what does it mean? Social psychology can be used to describe group phenomena. What is a culture of a small group (private laughing)?
——————————————————-