Explaining Reasoning Effects: A Neural Cognitive Model of Spatial Reasoning

AbstractAccording to mental model theory, spatial reasoning is based on the construction and variation of mental models representing spatial arrangements. Several effects in human spatial reasoning are known to support this theory, for example the ordering effect. Yet, reasoning effects have been observed for which the cognitive mechanisms are not entirely explained. To investigate how these effects can be attributed to neural computation, we modeled spatial reasoning in the Neural Engineering Framework. We selected three experiments to simulate tasks in a cognitive model based on an internal display. In our model, performance declines with an increase of objects which is explained by the neural drift over time. We replicated effects from the studies which we have found to be due to continuous premise integration. By modeling and simulating spatial reasoning tasks, we showed that effects reported in psychological studies can be explained by the emergent properties of neural computation.

Return to previous page