Have You Ever Seen a Robot? An Analysis of Children's Drawings Between Technology and Science Fiction

Abstract

Technologies have become an essential part of the daily life of our children. Consequently, artifacts that imply the early adoption of abstract thinking affect the imagination of children and young people in relation to the world of technology, now much more than they did in the past. With the emerging importance of robots in many aspects of our everyday lives, the goal of this study is to investigate which mental representations children have about robots. To this end, drawings from 104 children aged between 7 and 12 years old were used as a map of representations, considering the drawings as a proxy capable of evoking learned or emerging mental frameworks. The drawings were analyzed in several steps: they were first labeled using binary descriptors and then classified using clustering methods based on Hamming distances between drawings. Finally, questionnaire items covering children's perceptions about robots were analyzed for each of the resulting cluster separately to identify differences between them. The results show that there are relationships between the way children draw robots and their perception about robots' capabilities as well as their aspirations to pursue a career in science. These findings can provide meaningful insights into how to design educational robots and learning activities for children to learn with and about robots.

Keywords: Digital literacy; Drawings; Educational robotics; Primary school; Secondary school.

Fig. 1

Schematic representation of the data collection (a), classification (b), and clustering process (c)

Fig. 2

Example of a child’s robot drawing and its classification using the proposed binary descriptors

Fig. 3

The three main clusters of drawing types identified from questionnaire item Q1. Each cluster was named after robots from cinematographic origins: Wall-e cluster (a), Eve cluster (b) and HAL cluster (c). For each cluster, the centroid vector, some example drawings, and the distribution by age group are presented

Fig. 4

The two main clusters of drawing types identified from questionnaire item Q2 (reality cluster (a) and fiction cluster (b)) as well as the cluster that did not submit a drawing for this question (c). For each cluster, the centroid vector and some example drawings and the distribution by age group (ELEM, elementary school; LSEC, lower secondary school) are presented

Fig. 5

Sankey chart illustrating the relationships between the clusters from item Q1 (left) and the clusters from item Q2 (right). The labels indicate the absolute number of drawings in each cluster

Fig. 6

Answers to questionnaire items Q3 (left) and Q4 (right) by drawing clusters from Q1

Fig. 7

Answers to questionnaire items Q3 (left) and Q4 (right) by drawing clusters from Q1