The ventral visual system

The brain processes involved in visual object recognition are described. Evidence is presented that what is computed are sparse distributed representations of objects that are invariant with respect to transforms including position, size, and even view in the ventral stream towards the inferior temporal visual cortex. Then biologically plausible unsupervised learning mechanisms that can perform this computation are described that use a synaptic modification rule what utilises a memory trace. These are compared with deep learning and other machine learning approaches that require supervision.

The dorsal visual system

The dorsal visual system computes information about where objects are in space, and their motion, and this is used for actions performed in space. This requires coordinate transforms from retinal coordinates to head based coordinates, and then in parietal cortex areas to coordinates for reaching into space, and for allocentric, world-based, spatial coordinates. Recent approaches to how these transforms are performed, with analogies to transform invariance learning in the ventral visual system, are described.

Does Metric Feedback Hinder Actions Guided by Cognition?

Providing trainees with metric feedback improves their metric distance estimations, but doing so also hinders certain actions. This paper describes a possible explanation for this hindrance. Based on that explanation, it was predicted that metric feedback should not hinder actions that are guided by cognitive processing, i.e., actions guided by the ventral visual system. To investigate this possibility, participants threw underhanded to specific metric distances during Pre and Post-Testing, e.g., throwing an object so that it came to rest 30 feet away. During the intervening Training, participants generated verbal distance estimates. Half received metric feedback. The results indicated that throws improved from Pre to Post-Test, but only when participants received metric feedback during Training. This outcome supports our hypothesis. Moreover, it suggests that trainees must know whether their distance estimation training should be applied to untrained tasks. Doing so may benefit certain tasks. Others, however, may suffer from it.