HOLZ W C, AZRIN N H, ULRICH R E
J Exp Anal Behav. 1963 Jan;6(1):115-22. doi: 10.1901/jeab.1963.6-115.
The responses of pigeons were maintained by a DRL schedule of food reinforcement. With this schedule, responses were reinforced only when a fixed period of time elapsed without an intervening response. Punishment of all responses reduced the frequency of these responses as a direct function of the punishment intensity. As a consequence of the increased temporal spacing of responses, more reinforcements resulted during punishment. Under progressively higher intensities of punishment, the reinforcement frequency increased to a maximum value and then decreased at the highest intensities. The increased frequency of reinforcement which resulted during punishment did not counteract the suppressive effect of punishment, nor did it lead to a low response rate after punishment was removed. Punishment did not reduce the inter-response time distribution uniformly, but rather especially reduced the number of short inter-response times. Even at the low punishment intensities, the number of short inter-response times was considerably reduced. After punishment was discontinued, performance recovered almost completely after a compensatory burst. The number as well as the temporal pattern of responses returned to normal.
鸽子的反应由食物强化的DRL(Differential Reinforcement of Low Rates,低反应率差异强化)程序维持。在这个程序中,只有当经过一段固定时间且没有中间反应时,反应才会得到强化。对所有反应进行惩罚会降低这些反应的频率,且惩罚频率是惩罚强度的直接函数。由于反应的时间间隔增加,在惩罚期间会产生更多强化。在逐渐增加的惩罚强度下,强化频率增加到最大值,然后在最高强度时下降。惩罚期间产生的强化频率增加并没有抵消惩罚的抑制作用,也没有导致惩罚撤销后反应率降低。惩罚并没有均匀地减少反应间隔时间分布,而是特别减少了短反应间隔时间的数量。即使在低惩罚强度下,短反应间隔时间的数量也大幅减少。惩罚停止后,经过一次补偿性爆发,表现几乎完全恢复。反应的数量以及时间模式都恢复正常。