Differential affection of the visual information sub-streams in a patient with visual agnosia

Summary of the publication-based Doctoral thesis of Kirstin Lederer: Differential affection of the visual information sub-streams in a patient with visual agnosia. The two visual streams hypothesis postulates that visual information is processed along two separate but interacting cortical pathways: for purposes of object interaction, the information is computed in the fast-processing dorsal stream, while object perception is enabled by the slower ventral stream. Visual agnosia is a deficit of object recognition attributed to the damage of the ventral stream while the dorsal stream remains overall functioning. In the study at hand, a new patient case is presented. In this patient (AC), visual agnosia resulted from extensive occipito-temporal cortical lesioning after he suffered sudden cardiac arrest and a prolonged resuscitation. In stark contrast to profound visual deficits in testing, AC retained considerable proficiency in interacting with his known surrounding. For example, he was able to find his way in his home town, go shopping, and perform other activities of daily living. Similar reports have been made on other visual agnosic patients. The aim of this study was thus to further investigate this discrepancy by analyzing AC’s grasping behaviour towards everyday objects in comparison to a healthy control group. The objects had either a stereotypical color or were color-masked. Participants were instructed to either grasp-then-name or name-then-grasp the object presented. According to the two visual streams hypothesis, naming the object first would elicit processing predominantly along the ventral stream whereas immediate grasping would induce dorsal stream processing. The control group’s verbal object identification was error-free; kinematic analysis revealed a slowing effect of color-masking, irrespective of the instruction. This could be explained by an increased processing load in the healthy ventral stream when confronted with objects of atypical coloring and would be in line with pre-existing evidence that color is a significant cue for object identification and that the two streams interact when handling known objects. Intriguingly, AC identified all stereotypically colored objects but only few color-masked objects. This suggests that visual agnosic patients use their remaining color vision, presumably in combination with detail-based recognition approaches, for object identification in everyday activities. AC’s kinematic performance was overall slowed compared to the control group and even slower towards color-masked than stereotypically colored objects, indicating that AC’s impaired ventral stream was affected by color masking to a larger extent than those in the control group. Also, his maximum hand aperture was significantly larger when he had to immediately grasp color-masked objects. This could be interpreted as a safety grasp possibly due to a lack of ventral stream contribution to the visuomotor transformation under time constraint. Overall, these results indicate that visual agnosic patients’ slowed performance could constitute a compensatory behaviour and facilitate processing in their partially damaged ventral stream and hence contribute importantly to their rather proficient object interaction in their everyday environment.