Investigating re-representation through categorisation

AbstractIt is not entirely clear under what conditions people are able to re-represent their knowledge of a situation without top-down influences, such as explicit hints. Categorisation paradigms can help investigate this problem. Sewell and Lewandowsky (2011) found that people can change categorisation strategies without further learning relatively quickly in response to an explicit hint. In their paradigm, the category space is designed such that the exemplars can be categorised accurately using one of two strategies. As such, the present work used and extended this paradigm in a more knowledge-rich domain, compared to the visual domain in Sewell and Lewandowsky's study. In contrast to the previous work, we did not find evidence for the change in strategy without further learning. Further, re-representation can be measured more clearly by taking away the critical cue for one strategy after participants learnt it, and measuring the rates of learning.

Return to previous page