[ Back to main page ]
 

Abstract

 
Abstract No.:B-G2200
Country:Canada
  
Title:RESPONSE ON PREVIOUS ANTISACCADE TRIAL AFFECTS PERFORMANCE ON SUBSEQUENT TRIAL IN BOTH AGING AND MILD ALZHEIMER’S DISEASE.
  
Authors/Affiliations:4 Liam D. Kaufman*; 1 Cori Atlin; 2 Jay Pratt; 3 Sandra E. Black;
1 Dept. of Psychology, McGill University, Montreal, QC, Canada; 2 Dept. of Psychology, University of Toronto; 3 Division of Neurology, Sunnybrook Health Science Centre; 4 Institute of Medical Science, University of Toronto, ON, Canada.
  
Content:Objectives: The antisaccade task, considered an index of dorsolateral prefrontal function, requires a participant to look away from a peripheral target, in contrast to a prosaccade task in which the person looks towards the target. While antisaccade performance in Alzheimer’s Disease (AD) has been tested in patients with moderate to severe impairment, the performance of mildly impaired patients has not been well studied, and the lack of such data may exaggerate reported performance differences between AD and healthy controls. Furthermore, it has not been determined if a response made on a trial (correct vs. incorrect) has an effect on the subsequent response in patients with AD. The objectives of the present study were twofold: 1) examine antisaccade performance in patients with mild AD, relative to controls, and 2) determine if the response (correct vs. incorrect) on previous trials affects the subsequent response.

Materials and Methods: Using a laptop-based testing paradigm, 21 patients with AD and 23 normal controls completed one block of prosaccades and two blocks of antisaccades (24 trials per block). Eye movements were captured using the laptop’s integrated web camera and later coded for correct responses. Each participant completed a mini-mental state examination (MMSE): AD patients with an MMSE less than 18 were not included in the study.

Results: MMSE scores were significantly lower in the AD group (p < 0.01), while there were no significant differences for years of education and age. There were an equal percentage of errors toward both the right and left directions in both groups on each task: therefore, left and right errors were aggregated. AD patients made significantly more errors on the prosaccade block than controls (p < 0.05), as well as significantly more errors in direction on both the antisaccade blocks (p < 0.01). Antisaccade errors and MMSE scores within the AD group (r = -0.27, p = ns) or the control group (r = 0.05, p = ns) were not correlated. In AD patients, eye movements were 2.3 times more likely to be erroneous when preceded by an incorrect response and 1.9 times more likely to be correct when preceded by a correct response. In contrast, controls were more likely to make a correct response regardless of the previous response.

Conclusions: These results are the first to suggest that patients with mild AD make significantly more errors than normal controls on the antisaccade task, and that the error rates cannot just be attributed to severity as measured by the MMSE. It is also noteworthy that in the AD group, the response on the prior trial (correct or incorrect) tended to be repeated on the current trial. The same was true for correct responses. These hitherto unexplored perseverative errors in the AD group may have resulted from a degradation of task-set within working memory.
  
Back