大鼠的因果推理能力：以味覺嫌惡制約為例

Causal inference in rats: Evidences from conditioned taste aversion

李洸楚(Kuang-Chu Li)

指導教授 ： 李季湜

國立中正大學/社會科學院/心理學系/博士(2016年)

摘要

唯有瞭解事物間的因果關係，我們才得以從解釋過去事件背後的機制，來推測將來可能發生的事情，進而決定該怎麼做，以便得到一個更美好的將來。因果推理被視為人類獨有的、藉以判斷事物間因果關係的關鍵能力。儘管過去有許多研究努力地想證明非人類的動物也具有因果歸因的能力，然而到目前為止仍然存在著許多的爭議。本研究採用味覺嫌惡制約，藉由動態共變劑量的鋰鹽注射，結合（1）制約刺激遮蔽、（2）二階段逆向制約阻斷與制約遮蔽還原、（3）後果疊加性在二階段回溯性重新評估等作業，來模擬人類受試者的因果學習實驗，大鼠對於不同味覺物與鋰鹽劑量間些微的共變差異非常敏感，並且會利用這些最近獲得的共變訊息，回溯性的改變先前已習得的味覺嫌惡制約聯結強度。這些結果難以用聯結論的模型加以解釋，卻非常符合人類利用因果歸因能力來解決二階段回溯性重新評估的實驗結果。

關鍵字

後果疊加性 ； 二階段回溯性重新評估 ； 制約遮蔽還原 ； 逆向制約阻斷 ； 共變性學習 ； 因果歸因 ； 主動學習

並列摘要

Only to understand the causal relationship between things, we were able to explain the mechanism behind the past events, so that we can speculate what might happen in the future, and further decide how to do in order to get a better future. Causal inference is the crucial ability, which is regarded as a unique competence to human by many researchers, to judge the causality between cues and outcomes. Nonetheless, numerous efforts have been given in the hope to demonstrate that non-human animals are capable of causal reasoning, though with controversial results. In this study, conditioned taste aversion with a dynamic contingent dose of LiCl was used to replicate human subjects’ causal learning experiments in the (a) overshadowing, (b) second-order backward blocking and unovershadowing, and (c) outcome additivity in second-order retrospective revaluation tasks. Rats could detect the subtle differences in the contingent relationships between the amount of LiCl and the tastants. The information which they acquired in recent times could help them back to change the strength of taste aversions learned in previously. The findings cannot be explained by any associative model, but is consistent with the second order retrospective revaluation known in human study as a type of causal reasoning.

並列關鍵字

outcome additivity ； 2nd-order retrospective revaluation ； unovershadowing ； backward blocking ； contingency learning ； causal reasoning ； active learning

參考文獻

翟本瑞(1987)。古希臘思想中的因果觀研究。鵝湖月刊, 142, 30-38.

Beckers, T., Miller, R. R., Jan, D. H., & Urushihara, K. (2006). Reasoning rats: Forward blocking in Pavlovian animal conditioning is sensitive to constraints of causal inference. Journal of Experimental Psychology: General, 135, 92-102. doi: 10.1037/0096-3445.135.1.92

Blaisdell, A. P., Sawa, K., Leising, K. J., & Waldmann, M. R. (2006). Causal reasoning in rats. Science, 311, 1020-1022. doi: 10.1126/science.1121872

Burgess, K. V., Dwyer, D. M., & Honey, R. C. (2012). Re-assessing causal accounts of learnt behavior in rats. Journal of experimental psychology. Animal behavior processes, 38, 148-156. doi: 10.1037/a0027266

Connor, P. C., Lolordo, V. M., & Trappenberg, T. P. (2014). An elemental model of retrospective revaluation without within-compound associations. Learning & behavior, 42, 22-38. doi:10.3758/s13420-013-0112-z