Learning to avoid windows may explain variation in species risk of window strikes. Recent research has found that <40% of window strikes resulted in a fatal outcome (http://birdswindows.biology.ualberta.ca/our-research/). It would be very interesting to know if the survivors of these window crashes learned what had happened to them. In a proximate sense, a bird that survives a window crash needs to formulate a causal relationship between the event (window crash) and the feature (window) that caused the event, and then live the rest of its life without being fooled again by a window.
The ‘learning to avoid windows hypothesis’ assumes that post-strike cognition would have to function at a high enough level to enable learning. Window collisions cause a range of effects in birds that depend on the intensity of the collision. I have seen many birds hit windows and immediately fly away and behave ‘normally’. In other cases, post-collision survivors fall to the ground and display a range of symptoms, such as inability to move, shaking/shivering, vomiting/defecating, abnormal breathing patterns, loss of balance, and temporary loss of consciousness. In humans, we know that these symptoms are associated with concussions and are correlated with memory problems, issues with taste and smell, confusion, and fatigue (www.mayoclinic.com). Indeed, the cause of death in birds after window collisions is mainly due to massive brain contusions and bleeding, and not injury to the bony features of the head and neck as is often assumed (see Veltri and Klem, 2005, http://www.muhlenberg.edu/main/academics/biology/faculty/klem/aco/documents/FieldJournaltowers-windows2005.pdf). Thus, if a bird survives a window collision, then the opportunity for learning may be hampered by the negative cognitive effects of acute and chronic brain trauma.