Encyclopedia of Animal Cognition and Behavior

Living Edition
| Editors: Jennifer Vonk, Todd Shackelford


  • Miguel A. VadilloEmail author
  • Itxaso Barberia
Living reference work entry
DOI: https://doi.org/10.1007/978-3-319-47829-6_1236-1


Human Predictive Learning Associative Learning Theory Unconditioned Stimulus (US) Conditioned Stimulus (CS) Close Contiguity 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


In the associative learning literature, blocking is typically defined as a deficit in responding to a conditioned stimulus (CS) that has always been paired with an unconditioned stimulus (US) in the presence of another CS previously established as a reliable predictor of the same US.


Until the late 1960s, it was generally assumed that the mere temporal and spatial contiguity between a CS and a US was a necessary and sufficient condition for conditioning to take place. The discovery of blocking by Leon Kamin (1968) was the first of a series of findings that defied this view, stimulating the development of modern learning theories. His seminal experiments on blocking relied in the experimental design summarized in Table 1. As can be seen, in this experimental design, subjects in both groups experienced the same number of pairings between a light (L) and an electric shock. However, conditioned responding (CR) to the light was significantly stronger in the control group than in the experimental group. In other words, close contiguity between a conditioned and an unconditioned stimulus (L and the shock, in this case) did not suffice to produce strong conditioning.
Table 1

Design summary of Kamin’s (1968) blocking experiments


Phase 1

Phase 2




16 N – shock

8 LN – shock


Weak CR


8 LN – shock


Strong CR

Note: N and L refer to a white noise and to the turning on of a house light, respectively; the unconditioned stimulus is a 1-milliampere electric shock; CR refers to conditioned responding

According to Kamin (1968), this effect reveals that pairings of a CS and a US only give rise to conditioning if the US is unexpected. In the experimental group, when L is paired with the shock, this event produces little or no surprise, because the shock is readily predicted by a previously conditioned white noise (N). This is not the case in the control group, where the shock is completely unanticipated the first time L is presented. The idea that surprise drives learning was soon implemented in formal theories of associative learning. For instance, in the influential model of classical conditioning developed by Rescorla and Wagner (1972), surprise is represented as a prediction error, i.e., as the arithmetic difference between the “expected US” and the “experienced US.” In this model, the amount of learning supported by a conditioning trial is directly proportional to this difference.

Some models developed over the following years attempted to explain blocking by postulating different mechanisms. For instance, according to Mackintosh (1975), blocking can be explained by assuming that attention is biased toward predictive stimuli. In the case of the experimental group in Table 1, when L and N are presented together for the first time, N is the most reliable predictor of the US and subjects will learn to pay more attention to it, leaving few attentional resources to process (and learn about) L. In a radically different vein, Miller and Matzel (1988) proposed that blocking is not due to a deficit in learning itself but to late processes related to the production of the conditioned response. In any case, as shown by these examples, offering a successful account of blocking soon became a touchstone for any theory of associative learning.

Although the blocking effect was initially explored in the area of animal conditioning, during the 1980s, several studies showed that it could also be detected in human predictive learning experiments (Dickinson et al. 1984). This led naturally to the conclusion that the mechanisms underlying different forms of human learning must be similar to the ones responsible for classical and operant conditioning in nonhuman animals. The following decades of research in human predictive learning were largely driven by the impetus provided by associative learning theories and animal conditioning research.

More recently, it has been suggested that blocking might not be the product of an associative learning process but rather the result of a simple deductive inference (De Houwer et al. 2002). If two conditioned stimuli predict a US, then presenting both of them together should predict and even more intense US. In a blocking design, however, the intensity of the US is exactly the same when only one CS is presented (N) and when two CSs are presented (L and N). Therefore, it can be concluded that one of the CSs (L) is not a predictor of the US. Consistent with this view, blocking seems to be sensitive to participants’ assumptions about the additivity of CSs (i.e., whether two CSs should predict a stronger US or not) or the maximality of the US (i.e., whether the intensity of the US is not changing because it is already happening at its maximal level). Manipulating these assumptions does not only affect blocking in human participants but also in nonhuman animals (Beckers et al. 2006). These findings have raised concerns about the validity of the whole associative learning framework, leading some authors to suggest that blocking and other learning effects are better understood in terms of propositional inference processes (Mitchell et al. 2009; but see Shanks 2010).



  1. Beckers, T., Miller, R. R., De Houwer, J., & 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.CrossRefGoogle Scholar
  2. De Houwer, J., Beckers, T., & Glautier, S. (2002). Outcome and cue properties modulate blocking. Quarterly Journal of Experimental Psychology, 55, 965–985.CrossRefPubMedGoogle Scholar
  3. Dickinson, A., Shanks, D., & Evenden, J. (1984). Judgement of act-outcome contingency: The role of selective attribution. Quarterly Journal of Experimental Psychology, 36, 29–50.CrossRefGoogle Scholar
  4. Kamin, L. J. (1968). “Attention-like” processes in classical conditioning. In M. R. Jones (Ed.), Miami Symposium on the prediction of behavior, 1967: Aversive stimulation (pp. 9–31). Coral Gables: University of Miami Press.Google Scholar
  5. Mackintosh, N. J. (1975). A theory of attention: Variations in the associability of stimuli with reinforcement. Psychological Review, 82, 276–298.CrossRefGoogle Scholar
  6. Miller, R. R., & Matzel, L. D. (1988). The comparator hypothesis: A response rule for the expression of associations. In G. H. Bower (Ed.), The psychology of learning and motivation (Vol. 22, pp. 51–92). San Diego: Academic.Google Scholar
  7. Mitchell, C. J., De Houwer, J., & Lovibond, P. F. (2009). The propositional nature of human associative learning. Behavioral and Brain Sciences, 32, 183–198.CrossRefPubMedGoogle Scholar
  8. Rescorla, R. A., & Wagner, A. R. (1972). A theory of Pavlovian conditioning: Variations in the effectiveness of reinforcement and nonreinforcement. In A. H. Black & W. F. Prokasy (Eds.), Classical conditioning II: Current research and theory (pp. 64–99). New York: Appleton-Century-Crofts.Google Scholar
  9. Shanks, D. R. (2010). Learning: From association to cognition. Annual Review of Psychology, 61, 273–301.CrossRefPubMedGoogle Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  1. 1.Departamento de Psicología BásicaUniversidad Autónoma de MadridMadridSpain
  2. 2.Department of Cognition, Development and Educational PsychologyUniversitat de BarcelonaBarcelonaSpain

Section editors and affiliations

  • Oskar Pineno
    • 1
  1. 1.Hofstra UniversityLong IslandUSA