This thesis reportsthe results of five novel studies thatused eye movement patterns to elucidate the role of shape information content of object shape representation in human visual perception. In Experiments 1, and 2eye movements were recorded while observers either actively memorised or passively vieweddifferent setsof novel objects, and during a subsequent recognition memory task. Fixation data were contrasted against different modelsof shape analyses based on surface curvaturebounding vs. internal contour and low level image visual saliency. The results showed apreference for fixation atregions of internal local features(either concave or/and convex)during both active memorisation and passive viewingof object shape. This pattern changed during the recognition phase where there was a fixation preference towards regions containing concave surface curvature minima. It is proposed that the preference of fixation at regions of concavity reflect the operation of a depth-sensitive view interpolation process that is constrained by keypoints encoding regions of concave curvature minima. Experiments 3 and 4examined the extent to which fixation-based local shape analysis patterns are influenced by the perceptual expertise of the observer and the level of stimulus classification required by the task. Thesestudies werebased on the paradigm developed by Wong, Palmeri & Gauthier (2009) in which observers are extensively trained to categorize sets of novel objects (Ziggerins) at either a basic or subordinate level of classification. The effects of training were measured by comparing performance between a pre-and post-test sequential shape matching task that required either basic-or subordinate-level judgements. In addition, we also recorded fixation patterns during the pre-and post-tests. The resultsshowed significant effects of training on shape matchingin the post-tests. In particular, participants showed evidence of perceptual expertise inmakingbasic and subordinate-level shape classification judgements. We also found that the acquisition of perceptual expertisedid notresultin significant changes in the local spatial distributions of fixationpatterns observed. However, there was a tendency for fixations located at areas with concave curvature minima regardless of level of classification. This finding provides evidence that the preference for fixating at concave regions generalises across levels of stimulus classification in recognition tasks. The last studyexamined how eye movement patterns can be used to elucidate shape analyses strategies across tasks of object recognition and planning prehensile movement where participants wereasked to memorise and later recognise an objectand to imagine picking up an object by using their thumb and a forefinger. The results showed a significantly different fixation pattern between the recognition and motor imagery task, thus providingsupport for differential processing during shape perceptioninfluenced bytask demands.The main empirical findings in this thesis show:1) How eye movements can elucidate properties of internal mental representations of shape; 2) Consistent fixation pattern to concave areas that generalises across tasks; and 3) Different fixation patterns during recognition and motor imagery task.