At the recent NTEN ResearchED meeting at York (3 May 2014), I attended a very interesting talk entitled ‘The Learning Brain: a new science of learning’ given by Jonathan Sharples of the Institute for Effective Education and Education Endowment Foundation. One aspect of his talk really caught my interest with the York Science project in mind. Jonathan presented some research on the effect of repeated testing on learning. Four groups of research participants had to learn 40 words in Swahili, all participants had no prior knowledge of the language. Initially, all students studied the list of Swahili-English word pairs (e.g. mashua –boat) and were then tested (e.g. mashua – ?). However, once a word pair was recalled correctly, it was treated differently by each of the groups, as follows:
- continued to be studied and tested throughout
- studied but no further testing
- no further study but still tested
- no further study or testing of the words.
The results for the four groups in the study are shown in the Figure below. However, before reading on decide which of the results you think match each of the four groups.
The result A is group 1 (studied and tested), B is group 3 (no study only testing), C is group 2 (study with no testing) and D is group 4 (no study and no testing). The results provide evidence that testing does not merely measure learning but contributes to the process of learning, with repeated testing enhancing learning. Jonathan advocated that in addition to Assessment for Learning and Assessment of Learning maybe we should be thinking about Assessment as Learning. The repeated recruitment of neural pathways strengthening the retrieval networks involved in the learning. Could it be that the formative assessment tasks of York Science designed to allow teachers to obtain evidence of learning may also provide an additional learning opportunity for students?
Read more in Jeffrey D. Karpicke and Henry L. Roediger (2008) The Critical Importance of Retrieval for Learning. Science 319, 966-968
The paper can be downloaded from: http://learninglab.psych.purdue.edu/publications/
Anne Scott is a member of the York Science team.