Floor-Plan Drawing After Right-Hemisphere Stroke
You are an occupational therapist evaluating a 58-year-old man three months after a right-hemisphere stroke. He lives alone and navigates his own apartment without difficulty, but he got lost twice in the new outpatient clinic. To assess visual memory you show him a simple six-room floor plan for 20 seconds, then remove it and ask him to draw it from memory. He draws four rooms, places the bathroom on the wrong side, omits the corridor, and leaves the left third of the page blank. When you show him the plan again and ask him to point to each room, he identifies all six correctly. On the Corsi block-tapping task his forward span is 4 and backward span is 3. His wife reports he "remembers faces but not where he parked." You need to decide whether this is a visual memory storage problem, a visuospatial working memory problem, or an encoding artifact from neglect, and what to assess next.
1. The drawing is poor but recognition is perfect. What does this dissociation tell you about storage versus retrieval, and what alternative explanation must be excluded?
Read the worked answer
Intact recognition with poor free reconstruction argues against a storage deficit: the layout was encoded well enough to support accurate recognition. The failure is at retrieval or strategic reconstruction, or in the visuospatial working memory needed to hold and manipulate the plan while drawing. However, the blank left third and the misplaced bathroom raise left hemispatial neglect as an encoding artifact: if he did not attend to left-sided elements, the "memory" error is really an attentional one. Before calling it visual memory, test copying with the model present, line bisection, and a landmark task. If he omits the same left-side rooms when copying, the drawing is not a clean memory measure. Also check constructional praxia: a drawing problem can mimic a memory problem.
2. Design a next-session assessment that separates neglect, constructional difficulty, and visual memory. Be specific about conditions and scoring.
Read the worked answer
Use a 2x2 design: copy versus memory, symmetrical versus asymmetrical layout. In the copy condition the model stays visible; in the memory condition it is removed after 20 seconds. Score left and right omissions separately, room identity, and room location. Add a recognition-from-array test with four alternative layouts to measure storage without drawing. Add Corsi forward and backward and a spatial span with interference such as backwards tapping to index visuospatial working memory. Administer line bisection and a landmark task to quantify neglect. If left omissions occur in copy and memory equally, neglect dominates; if they occur only in memory, encoding or retrieval of left space is implicated. If recognition is normal but reconstruction fails across layouts, target strategic retrieval and working memory rather than visual storage.
3. What functional recommendation follows, and what outcome would you monitor?
Read the worked answer
Recommend externalized spatial support rather than memory drills: take a photo of the clinic route, use numbered landmarks at decision points, and have him verbalize "turn left at the blue desk" at each junction. For parking, pair the floor and bay number with a distinctive visual anchor and a phone photo. Monitor transfer: number of wrong turns per visit, time to locate the therapy room, and whether he self-corrects after a wrong turn. Also monitor left-side awareness in everyday tasks such as bumping the left side or missing items on the left of a tray. If neglect is confirmed, treat that directly; if working memory load is the bottleneck, reduce route complexity and provide a map. Reassess drawing after four weeks with the same layout so the comparison is meaningful.