Use it before you lose it: greater efficacy of visual training for recovering contrast sensitivity in subacute cortical blindness

Abstract

A hallmark of chronic (>6 months) V1 stroke in humans is loss of luminance contrast sensitivity (CS) and visual discrimination abilities in the cortically blind (CB) field. Training chronic CBs on high-contrast random dot stimuli in their blind field may restore motion discriminations, but CS remains impaired.1 Recently, we showed that some subacute CBs (<3 months) retain blind field CS at a range of spatial frequencies (SFs), particularly for drifting stimuli.2 Here, we asked whether training could further improve blind field CS, and compared its efficacy in subacute (n=5) and chronic (n=5) CBs. All subjects trained at 2 non-overlapping blind field locations on static orientation and direction discrimination tasks; Gabor (1 cpd) contrast varied within session on a 3:1 staircase. After daily home training for 3–6 months, we measured changes in static (orientation discrimination) and motion (direction discrimination) CS (0.1–10 cpd) using the qCSF. Training with drifting Gabors improved motion CS across multiple SFs in both chronic and subacute CBs, but did not recover static CS. Training with static Gabors improved both static and motion qCSFs in subacutes, but failed to elicit any improvements in chronic CBs. Our results suggest greater capacity for recovery of visual discriminations in the subacute over the chronic post-stroke period, particularly regarding static stimuli. Given the potential for trans-synaptic retrograde degeneration of parvocellular units in the dorsal lateral geniculate nucleus and retina, we propose that early training may take advantage of networks that are preserved immediately after V1 lesion but deteriorate absent deliberate training.