Visual cognition in amnesic H.M.: Selective deficits on the What's-Wrong-Here and Hidden-Figure tasks

Abstract

Two experiments compared the visual cognition performance of amnesic H.M. and memory-normal controls matched for age, background, intelligence, and education. In Experiment 1 H.M. exhibited deficits relative to the controls in detecting “erroneous objects” in complex visual scenes—for example, a bird flying inside a fishbowl. In Experiment 2 H.M. exhibited deficits relative to the controls in standard Hidden-Figure tasks when detecting unfamiliar targets but not when detecting familiar targets—for example, circles, squares, and right-angle triangles. H.M.’s visual cognition deficits were not due to his well-known problems in explicit learning and recall, inability to comprehend or remember the instructions, general slowness, motoric difficulties, low motivation, low IQ relative to the controls, or working-memory limitations. Parallels between H.M.’s selective deficits in visual cognition, language, and memory are discussed. These parallels contradict the standard “systems theory” account of H.M.’s condition but comport with the hypothesis that H.M. has difficulty representing unfamiliar but not familiar information in visual cognition, language, and memory. Implications of our results are discussed for binding theory and the ongoing debate over what counts as “memory” versus “not-memory.”

Keywords:

• Visual cognition
• Amnesia
• Patient H.M.
• Error detection
• Hidden figures

Acknowledgments

Don MacKay gratefully acknowledges support for this research from the Samuel A. MacKay Memorial Research Fund and National Institutes of Health (NIH) Grant R01 AG 09755 (1997–2003). We thank Suzanne Corkin for generously providing the venue and opportunities to test H.M. in 1997 and 1998. We also thank Irwin Stein for recruiting participants and Ibby Ambrose, Bryan Chae, Helen Chon, Angela Kessell, and Deirdre Milligan for general research assistance. This research was summarized in MacKay and James (2000), a poster presented to the April 2000 meeting of the Cognitive Neuroscience Society, San Francisco, California.

Notes

¹The 6 new controls included 2 who were run in all but the familiar target conditions.