Kehoe E J, Horne P S, Macrae M, Horne A J
School of Psychology, University of New South Wales, Kensington, Australia.
J Exp Psychol Anim Behav Process. 1993 Jul;19(3):265-83.
Real-time models contend that a conditioned stimulus (CS) gives rise to a cascade of hypothetical stimuli that govern conditioned responses (CRs) on a moment-by-moment basis. Experiments with the rabbit nictitating membrane response successfully extended these models to external stimuli. CSs were trained in sequence with an unconditioned stimulus (CSA-CSB-US). When the CSA-CSB interval was shortened, the CR was compressed; when the CSA-CSB interval was lengthened, the CR was broadened. Peaks appeared at 2 places, namely, 1 following CSA by a period equal to its CS-US interval and another following CSB by its CS-US interval. Outside the sequence, the individual CSs evoked CRs located between their respective CS-US intervals. When, however, the 2 CSs were trained separately, the CRs were appropriate to their respective CS-US intervals when tested alone or in sequence. The results are discussed in terms of the J.E. Desmond and J.W. Moore (1988) and S. Grossberg and N.A. Schmajuk (1989) models.
实时模型认为,条件刺激(CS)会引发一系列假设刺激,这些刺激时刻控制着条件反应(CR)。对兔瞬膜反应的实验成功地将这些模型扩展到了外部刺激。CS与无条件刺激按顺序进行训练(CSA-CSB-US)。当CSA-CSB间隔缩短时,CR被压缩;当CSA-CSB间隔延长时,CR变宽。峰值出现在两个位置,即一个在CSA之后出现,间隔时间等于其CS-US间隔,另一个在CSB之后出现,间隔时间为其CS-US间隔。在序列之外,单个CS诱发的CR位于它们各自的CS-US间隔之间。然而,当两个CS分别进行训练时,单独测试或按顺序测试时,CR与其各自的CS-US间隔相符。将根据J.E. 德斯蒙德和J.W. 摩尔(1988年)以及S. 格罗斯伯格和N.A. 施马朱克(1989年)的模型对结果进行讨论。