Erickson Molly A, Boudewyn Megan A, Winsler Kurt, Li Charlotte, Barch Deanna M, Carter Cameron S, Frank Michael J, Gold James M, MacDonald Angus W, Ragland John D, Silverstein Steven M, Yonelinas Andrew, Luck Steven J
Department of Psychiatry & Behavioral Neuroscience, University of Chicago, Chicago, Illinois.
Department of Psychology, University of California, Santa Cruz, California.
Biol Psychiatry Cogn Neurosci Neuroimaging. 2024 Dec;9(12):1271-1280. doi: 10.1016/j.bpsc.2024.07.022. Epub 2024 Aug 6.
People with psychosis and mood disorders experience disruptions in working memory; however, the underlying mechanism remains unknown. We focused on 2 potential mechanisms: poor attentional engagement should be associated with elevated levels of prestimulus alpha-band activity within the electroencephalogram (EEG), whereas impaired working memory encoding should be associated with reduced poststimulus alpha suppression.
We collected EEG data from 68 people with schizophrenia, 43 people with bipolar disorder with a history of psychosis, 53 people with major depressive disorder, and 90 healthy comparison subjects while they completed a spatial working memory task. We quantified attention lapsing, memory precision, and memory capacity from the behavioral responses, and we quantified alpha using traditional wavelet analysis as well as a novel approach for isolating oscillatory alpha power from aperiodic elements of the EEG signal.
We found that 1) greater prestimulus alpha power estimated using traditional wavelet analysis predicted behavioral errors; 2) poststimulus alpha suppression was reduced in the patient groups; and 3) reduced suppression was associated with a lower likelihood of memory storage. However, we also observed that the prestimulus alpha was larger among healthy control participants than patients, and single-trial analyses showed that it was the aperiodic elements of the prestimulus EEG-not oscillatory alpha-that predicted behavioral errors.
These results suggest that working memory impairments in serious mental illness primarily reflect an impairment in the poststimulus encoding processes rather than reduced attentional engagement prior to stimulus onset.
患有精神病和情绪障碍的人存在工作记忆障碍；然而，其潜在机制仍不清楚。我们关注两种潜在机制：注意力参与度差应与脑电图（EEG）中刺激前α波段活动水平升高有关，而工作记忆编码受损应与刺激后α波抑制减弱有关。
我们收集了68名精神分裂症患者、43名有精神病病史的双相情感障碍患者、53名重度抑郁症患者以及90名健康对照者在完成空间工作记忆任务时的EEG数据。我们从行为反应中量化注意力分散、记忆精度和记忆容量，并使用传统小波分析以及一种从EEG信号的非周期性成分中分离振荡α波功率的新方法来量化α波。
我们发现：1）使用传统小波分析估计的刺激前α波功率越大，预测的行为错误越多；2）患者组的刺激后α波抑制减弱；3）抑制减弱与记忆存储的可能性降低有关。然而，我们还观察到，健康对照参与者的刺激前α波比患者的更大，单次试验分析表明，预测行为错误的是刺激前EEG的非周期性成分，而非振荡α波。
这些结果表明，严重精神疾病中的工作记忆损害主要反映刺激后编码过程的损害，而非刺激开始前注意力参与度的降低。
