Top-down attentional modulation of electroencephalographic responses in normal aging and mild cognitive impairment

Abstract

Event Abstract Back to Event Top-down attentional modulation of electroencephalographic responses in normal aging and mild cognitive impairment Marie P. Deiber1, 2*, D. Jaques2, P. Missonnier2, G. Gold2, V. Ibañez2 and P. Giannakopoulos2, 3 1 INSERM U877, France 2 University Hospitals of Geneva, Switzerland 3 University Hospitals of Lausanne, Switzerland Background/Objectives: Top-down modulation of neural activity underlies the ability to selectively attend to relevant stimuli and ignore irrelevant stimuli [1, 2]. Such modulation depends on task instruction and could be altered when directed-attention networks are impaired, as recently described in elderly individuals with mild cognitive impairment (MCI) [3]. The objective of this study was to characterize top-down modulation of neural activity under different task instructions in healthy elderly subjects, and to study the impact of cognitive decline on top-down processing in MCI individuals, using analysis of event-related electrophysiological responses and oscillatory activities. Design/Methods: Continuous EEG was recorded using 32 surface electrodes in 28 elderly controls and 26 MCI patients matched for age and gender. Four tasks were performed in which the participants viewed a series of four sequentially presented stimuli (two faces and two letters) in a randomized order. The tasks differed in the instruction on how to process the stimuli: (1) Face (remember faces and ignore letters), (2) Letter (remember letters and ignore faces), (3) Dual (remember both faces and letters), and (4) Passive (passively view faces and letters). After a delay period, the participant reported whether the probe matched one of the previously presented, relevant stimuli (tasks 1 to 3); in Passive (task 4), the participant detected the pointing direction of an arrow. Analysis focused on occipital P1/N1, with particular attention to the occipito-temporal N170 specific for faces, and theta oscillatory activity (4-8 Hz). Results: In both groups, processing interferences occurred between faces and letters in the Dual task. Reaction time did not differ between controls and MCI, but was significantly affected by the task (Passive < Letter < Face < Dual). Performance was significantly worse in MCI than in controls in Face and Dual tasks, but similar in Letter and Passive tasks. The P1/N1 peaked earlier and the N170 was of larger amplitude for face than letter stimuli. Task instruction modulated event-related components as well as theta activity amplitude, suggesting a particular difficulty to suppress task-irrelevant face stimuli. Attentional processing of letters was additionally affected in MCI, as illustrated by reduced theta activity. Conclusions/Relevance: The results indicate a complex pattern of top-down modulation in both elderly controls and MCI patients, with a limited enhancement of task-relevant activity and a deficit in suppressing task-irrelevant activation. They also suggest the presence of an interaction between top-down selection for stimuli and bottom-up factors [1]. This interaction could be particularly strong in MCI patients, limiting the efficiency of their attentional control.