letters to nature 366 NATURE | VOL 410 | 15 MARCH 2001 | www.nature.com 19. Terborgh, J. Five New World Primates (Princeton Univ. Press, Princeton, 1983). 20. Leighton, M. Modeling dietary selectivity by Bornean orangutans: evidence for integration of multiple criteria in fruit selection. Int. J. Primatol. 14, 257±313 (1993). 21. Gautier-Hion, A. et al. Fruit characters as a basis of fruit choice and seed dispersal in a tropical forest vertebrate community. Oecologia 65, 324±337 (1985). 22. Davies, A. G. & Oates, J. F. in Colobine Monkeys (eds Davies, A. G. & Oates, J. F.) 229±249 (Cambridge Univ. Press, Cambridge, 1994). 23. Wrangham, R. W., Conklin-Brittain, N. L., & Hunt, K.D. Dietary responses of chimpanzees and cercopithecines to seasonal variation in fruit abundance: I. antifeedants. Int. J. Primatol. 19, 949±970 (1998). 24. Onishi, A. et al. Dichromatism in macaque monkeys. Nature 402, 139±140 (1999). 25. Struhsaker, T. T. Ecology of an African Rainforest (Univ. Florida Press, Gainesville, 1997). 26. Lucas, P. W. et al. Fieldkit to characterize the physical, chemical, and spatial aspects of potential primate foods. Folia Primatol. 72, 11±25 (2001). 27. Osorio, D. & Vorobyev, M. Colour vision as an adaptation to frugivory in primates. Proc. R. Soc. Lond. B 263, 593±599 (1996). 28. Darvell,B.W., Lee, P. K. D., Yuen, T. D. B.,Lucas, P. W. Meas. Sci. Technol. 7, 954±962 (1996). 29. Newton-Fisher, N. E. The diet of chimpanzees in the Budongo Forest Reserve, Uganda. Afr. J. Ecol. 37, 344±354 (1999). 30. Gartlan, J. S., McKey, D. B., Waterman, P. G., Mbi, C. N. & Strusaker, T. T. A comparative study of the phytochemistry of two African rainforests. Biochem. Syst. Ecol. 8, 401±422 (1980). Acknowledgements We thank D. Osorio for help with colour registration; E. Ting, P. Y. Cheng, I. C. Bruce, R. T. Corlett, L. Ramsden, N. Yamashita and A. Walker for comments, P. Kagoro, B. Balyeganira and M. Musana for ®eld assistance in Uganda; J. Magnay, R. W. Wrangham and C. A. Chapman for logistic support in Uganda; and the Ugandan National Council for Science and Technology, Ugandan Wildlife Authority and Makerere University Biological Field Station for permission to work at Kibale. Supported by Research Grants Council of Hong Kong, National Geographic Society, Sigma Xi, Explorer's Club and Croucher Foundation of Hong Kong. Correspondence and requests for materials should be addressed to N.J.D. (e-mail: njdominy@hkusua.hku.hk). ................................................................. Suppressing unwanted memories by executive control Michael C. Anderson & Collin Green Department of Psychology, University of Oregon, Eugene, Oregon 97403-1227, USA .............................................................................................................................................. Freud proposed that unwanted memories can be forgotten by pushing them into the unconscious, a process called repression 1 . The existence of repression has remained controversial for more than a century, in part because of its strong coupling with trauma, and the ethical and practical dif®culties of studying such pro- cesses in controlled experiments. However, behavioural and neurobiological research on memory and attention shows that people have executive control processes directed at minimizing perceptual distraction 2,3 , overcoming interference during short and long-term memory tasks 3±7 and stopping strong habitual responses to stimuli 8±13 . Here we show that these mechanisms can be recruited to prevent unwanted declarative memories from entering awareness, and that this cognitive act has enduring consequences for the rejected memories. When people encounter cues that remind them of an unwanted memory and they con- sistently try to prevent awareness of it, the later recall of the rejected memory becomes more dif®cult. The forgetting increases with the number of times the memory is avoided, resists incen- tives for accurate recall and is caused by processes that suppress the memory itself. These results show that executive control processes not uniquely tied to trauma may provide a viable model for repression. Executive control processes studied in behavioural 6,9,14 and neurobiological 2,4,10±13,15±17 research on cognition may provide a mechanism for the voluntary form of repression (suppression) proposed by Freud 1 . To test this hypothesis, we adapted the go/ no-go paradigm used to study executive control over motor actions in primates 18 and humans 15±17 for use in a memory retrieval task. First, we trained subjects on 40 unrelated word pairs (for example, ordeal±roach) so that they could recall the right-hand member of each pair when provided with the left-hand member. Next, subjects performed a critical task requiring them to exert executive control over the retrieval process. On each trial of this think/no-think task, a cue from one of the pairs appeared on the computer screen. Depending on which cue appeared, subjects were told either to recall and say (think about) the associated response word (respond pairs), or not to think about the response (suppression pairs). For the latter pairs, we emphasized that subjects should not allow the associated memory to enter consciousness at all. If subjects acci- dentally responded to a suppression pair, they heard a beep signal- ling an error. To increase the need to recruit inhibitory control mechanisms, we required subjects to ®xate on the cue word for the entire time (4 s) that it appeared on the screen, discouraging perceptual avoidance and generating a constant threat that the associated memory might intrude into consciousness. Thus, sup- pression trials required the stopping of both a prepotent motor 65 75 85 95 0 1 8 16 Number of repetitions Per cent recalled Suppress Respond a 65 75 85 95 0 1 8 16 Number of repetitions Per cent recalled Suppress Respond 65 75 85 95 0 1 8 16 Number of repetitions Per cent recalled Suppress Respond 65 75 85 95 0 1 8 16 Number of repetitions Per cent recalled Suppress Respond d 65 75 85 95 0 1 8 16 Number of repetitions Per cent recalled Suppress Respond e 65 75 85 95 0 1 8 16 Number of repetitions Per cent recalled Suppress Respond f 65 75 85 95 0 1 8 16 Number of repetitions Per cent recalled Respond-same Respond-indep Supp-same Supp-indep g b c Same probe Independent probe Figure 1 Final recall for respond and suppression items as a function of the number of repetitions for the same-probe (SP) and independent-probe (IP) tests. a, b, Experiment 1; c, d, experiment 2; e, f, experiment 3; g, averaged across experiments. Note the negative slope for recall of the suppressed item, indicating increasing inhibition. Inhibition (0 vs 16 suppressions) was signi®cant (P , 0.01) in all experiments, and did not interact with type of test cue (F , 1 in all cases; analysis of variance). Inhibition was signi®cant (P , 0.05) in every SP and IP test for every experiment (a±f). © 2001 Macmillan Magazines Ltd