Moxon Karen A, Gerhardt Greg A, Adler Lawrence E
Drexel University, School of Biomedical Engineering, Science and Health Systems, 3141 Chestnut St., Philadelphia, PA 19104-2875, USA.
Biol Cybern. 2003 Apr;88(4):265-75. doi: 10.1007/s00422-002-0372-8.
We modeled the neuronal circuits that may underlie a sensory-processing deficit associated with schizophrenia. Schizophrenic patients have small P50 auditory-evoked responses to click stimuli compared to normal subjects. The P50 auditory-evoked response is a positive waveform recorded in the EEG approximately 50 ms after the auditory click stimulus. In addition to relatively small amplitudes, schizophrenic patients do not gate or suppress the P50 auditory-evoked response to the second of two paired-click stimuli spaced 0.5 s apart. Neuropleptic medication, which decreases dopaminergic neuronal transmission, increases the amplitude of the P50 auditory-evoked response but does not improve gating. Normal subjects have large P50 auditory-evoked responses to click stimuli when compared to unmedicated schizophrenic patients, and they gate their response to paired click stimuli or have smaller P50 auditory-evoked response amplitudes to the second of two click stimuli spaced 0.5 s apart. Schizophrenic patients do not gate and have similar response amplitudes to both clicks. We hypothesized that the small amplitudes of unmedicated schizophrenic subjects were due to a state of occlusion whereby excessive background noise in local circuits reduced the ability of cells to respond synchronously to sensory input, thereby reducing the amplitude of the P50 waveform in the EEG. Because the P50 auditory-evoked potential amplitudes increased with neuroleptic medication, which reduces dopaminergic neuronal transmission, we hypothesized a role for dopamine in modulating the signal-to-noise (S/N) in the local circuits responsible for sensory gating. To test the hypothesis that modulation of the S/N ratio reduces sensory gating, we developed a model of the effects of dopaminergic neuronal transmission that modulates the S/N in neuronal circuits. The model uses the biologically relevant computer model of the CA3 region of the hippocampus developed in the companion paper [Moxon et al. (2003) Biol Cybern, this volume]. Modified Hebb cell assemblies represented the response of the network to the click stimulus. The results of our model showed that excessive dopaminergic input impaired the ability of cells to respond synchronously to sensory input, which reduced the amplitudes of the P50 evoked responses.
我们对可能构成与精神分裂症相关的感觉处理缺陷基础的神经回路进行了建模。与正常受试者相比，精神分裂症患者对点击刺激的P50听觉诱发电位反应较小。P50听觉诱发电位是在听觉点击刺激后约50毫秒在脑电图中记录到的一个正向波形。除了幅度相对较小外，精神分裂症患者不会对间隔0.5秒的两个配对点击刺激中的第二个刺激进行门控或抑制P50听觉诱发电位反应。抗精神病药物可减少多巴胺能神经元传递，增加P50听觉诱发电位的幅度，但不能改善门控功能。与未服用药物的精神分裂症患者相比，正常受试者对点击刺激有较大的P50听觉诱发电位反应，并且他们对配对点击刺激的反应进行门控，或者对间隔0.5秒的两个点击刺激中的第二个刺激有较小的P50听觉诱发电位反应幅度。精神分裂症患者不会进行门控，并且对两个点击的反应幅度相似。我们假设未服用药物的精神分裂症受试者的小幅度是由于一种阻塞状态，即局部回路中过多的背景噪声降低了细胞对感觉输入同步反应的能力，从而降低了脑电图中P50波形的幅度。由于P50听觉诱发电位幅度随着减少多巴胺能神经元传递的抗精神病药物而增加，我们假设多巴胺在调节负责感觉门控的局部回路中的信噪比（S/N）方面发挥作用。为了检验调节信噪比可减少感觉门控的假设，我们开发了一个多巴胺能神经元传递效应的模型，该模型可调节神经回路中的信噪比。该模型使用了在配套论文[莫克森等人（2003年），《生物控制论》，本期]中开发的海马体CA3区的生物学相关计算机模型。经过修改的赫布细胞集合代表了网络对点击刺激的反应。我们模型的结果表明，过多的多巴胺能输入损害了细胞对感觉输入同步反应的能力，这降低了P50诱发反应的幅度。
