The effect of the functional magnetic resonance imaging (fMRI) environment on performance in a feature binding task was investigated. A variation of Gajewski and Brockmole’s (2006) methodology was employed in which participants were asked to recall either the colour or shape of a probed object or both of these features. In regard to the feature binding aspect of the investigation the following two hypotheses were compared: 1. The Object Unit Hypothesis: participants will either remember all or none of the probed objects. 2. The Independent Stores Hypothesis: participants should be able to remember individual features such as the colour or shape of an object alone without the other feature being retained. A third hypothesis related to the effect of the fMRI environment states that the combination of environmental factors would affect performance of participants in the fMRI scanner simulator in one of two ways. Hypothesis 3a; the fMRI environment will adversely affect performance in the feature binding task. Hypothesis 3b; the fMRI environment will increase performance in the feature binding task. Results revealed that participants recalled higher numbers of correct features in the single feature conditions than in the binding conditions. Additionally, participants were affected by the fMRI environment in some conditions. When completing a single feature recall task their performance was improved, but scores were no different to the baseline condition when completing the feature binding task. These results supported the Object Unit Hypothesis in general but additional analysis revealed that independent features could also be retained in the binding condition. These results prompted the proposal of a new theory in which aspects of both the Object Unit and Independent Stores Hypotheses are compatible.