Do we have distinct systems for immediate and delayed actions? A selective review on the role of visual memory in...

The perception-action model with its assumptions of distinct visual pathways for perception and visuomotor control has been highly influential but also contentious. The controversy largely focused on the evidence from studies on perceptual illusions and this scientific field has been reviewed quite a few times in recent years. In contrast another aspect of the model, namely the role of visual memory in action control, received comparatively little attention. With respect to visual memory the perception-action model proposes that only the perceptual or ventral stream can maintain a sustained representation of the visual world while the visuomotor system or dorsal stream has to rely on currently available visual information. Consequently, visual information from the dorsal system cannot guide actions that are based on memorized visual information. We call this feature of the perception-action model: the dorsal amnesia hypothesis. There are at least two reasons for why this hypothesis is of special relevance. Firstly, it provides a particularly clear criterion to distinguish between functions of the ventral and dorsal stream. Secondly, this hypothesis led to some unexpected discoveries which provided particularly compelling evidence in favour of the model. In this review, we will revisit all relevant empirical areas, ranging from physiological examinations and neuropsychological studies to behavioural experiments in neurologically intact participants. Based on this review, we conclude that the dorsal amnesia hypothesis is in our view no longer tenable.