What my husband's stroke taught me about paying attention

Before Gord's stroke, I attended conferences the way most executives do — with a prepared frame, taking notes, leaving with three takeaways I mostly already knew. Then everything changed.

Before Gord's stroke, I attended conferences the way most executives do. You arrive with a prepared frame — a set of existing beliefs that you are, if you're being honest, mostly looking to refine rather than revise. You take notes on your phone. You network in the coffee break. You leave with three takeaways that you mostly already knew and will not quite remember by Friday. I had been doing this for twenty-five years across three continents. I was very efficient at it. The morning Gord had his brainstem stroke, we were at our cottage — an hour from the nearest hospital. By the time the ambulance arrived and his doctors were able to explain what had happened, I understood that the brainstem controls everything unconscious. Breathing. Swallowing. Balance. The automatic systems that run beneath thought. What happened next I will condense, because the details belong to us: three months of rehabilitation, a rearrangement of everything I had built the previous two decades around, and a specific kind of quiet that descends in hospital waiting rooms where you have no choice but to receive information exactly as it arrives. I had not listened like that since I was a child. Without a frame. Without the small, constant editorial process by which we convert incoming experience into something that fits our existing model. I was just — there. With whatever was being said. I have thought about this a great deal since. **The editorial problem** The neurologists have a term for the kind of listening most adults do: predictive processing. Your brain does not receive the world raw. It constantly generates a model of what is about to happen and updates it when reality diverges. This is efficient. It is also, in a very specific sense, the enemy of learning anything genuinely new. When you attend a lecture inside your own predictive model, you hear what you expect to hear. You update at the margins. The speaker's ideas pass through your existing framework and arrive, mildly pro