Rats exposed to spaced (S) conditions of training, with intertrial intervals (ITIs) averaging 90 s, acquired autoshaping performance (contact with a retractable lever presented 10 s before the response-independent delivery of 5 food pellets) faster than rats exposed to massed (M) training, with ITIs averaging 15 s. Gradual or abrupt transitions between S and M training were followed by concomitant gradual and abrupt changes in behavior. However, S-to-M and M-to-S transitions supported different levels of adjustment. S-to-M transitions were generally less open to competition between sign and goal tracking than M-to-S transitions. These results suggest that responses that become dominant early in training regulate terminal performance after extensive training. Thus, initial massed training promotes strong goal tracking that interferes with sign tracking even under conditions that would otherwise promote sign tracking. By contrast, initial spaced training tends to promote sign tracking that can interfere with goal tracking even when ITI duration favors goal tracking. These results are discussed in the context of the theoretical issue of path dependence. Their implications for learning theory and their potential applicability to the nature of sign and goal tracking endophenotypes in relation to substance use disorders are briefly discussed.

处于间隔（S）训练条件下，平均间隔间隔（ITI）为90 s的大鼠，其获得的自塑性能（在与反应无关的5个食物颗粒递送前10 s出现与可伸缩杠杆接触）比接触成块的大鼠要快（M）训练，ITI平均为15秒。S和M训练之间逐渐或突然的过渡，随后是行为的逐渐和突然变化。但是，从S到M和从M到S的过渡支持不同的调整级别。与从M到S的过渡相比，从S到M的过渡通常对标志和目标跟踪之间的竞争不那么开放。这些结果表明，在早期训练中占主导地位的反应会在广泛训练后调节终末表现。从而，最初的大规模训练可促进强大的目标跟踪，即使在原本会促进标志跟踪的条件下，也会干扰标志跟踪。相比之下，即使ITI持续时间偏向于目标跟踪，初始的间隔训练也倾向于促进符号跟踪，这可能会干扰目标跟踪。这些结果将在路径依赖的理论问题中进行讨论。简要讨论了它们对学习理论的意义及其对与药物滥用相关的体征和目标追踪内表型性质的潜在适用性。这些结果将在路径依赖的理论问题中进行讨论。简要讨论了它们对学习理论的意义及其对与药物滥用相关的体征和目标追踪内表型性质的潜在适用性。这些结果将在路径依赖的理论问题中进行讨论。简要讨论了它们对学习理论的意义及其对与药物滥用相关的体征和目标追踪内表型性质的潜在适用性。