Children have been shown to discriminate between reliable and unreliable sources from as young as 3 or 4 years old using the selective trust paradigm (Koenig, Clément & Harris, 2004). However, it is unclear whether children are discriminating between informants because they have some epistemic awareness regarding the knowledge of the informants, or because they regard the behaviour of the unreliable informants as bizarre. The current experiment manipulates source reliability in a more naturalistic way (science teacher vs. nursery child), and looks at the effects of age (6-7, 8-9, 10-11 years old) on children’s’ predictions relating to a familiar causal system cars on an inclined plane, where height, surface friction, and starting point on the slope, as well as the weight of the car, can be changed. Children are either told unintuitive information coming from a differentially reliable source, or no information, regarding the effect of weight. They are then asked to make predictions regarding how far the car travels. Children in the high reliable condition are more likely to change their predictions regarding the effect of weight following information from the more reliable source. This may occur in older children only, but more research needs to be done.