Within-trial contrast: The effect of probability of reinforcement in training

Cassandra D. Gipson, Holly C. Miller, Jérôme J.D. Alessandri, Thomas R. Zentall

Research output: Contribution to journalArticlepeer-review

8 Scopus citations


There is evidence that pigeons prefer conditioned reinforcers that are preceded by greater effort over those that are preceded by less effort (an effect that has been attributed to within-trial contrast). In past research the probability of reinforcement for correct choice of the conditioned reinforcer has been 100%, however, the high level of reinforcement for both alternatives in training may result in a performance ceiling when choice between those alternatives is provided on test trials. In the present study we tested this hypothesis by including a group for which the probability of reinforcement in training was only 50%. Pigeons were trained on two simultaneous discriminations, one that was preceded by a 30 peck requirement the other by a single peck requirement. On test trials, we found a significant preference for the S+ that required the greater effort in training for pigeons trained with 100% and a small but nonsignificant effect for pigeons trained with 50% reinforcement. Although the hypothesis that the within-trial contrast effect was constrained by a performance ceiling was not confirmed, we did find a reliable within-trial contrast effect with 100% reinforcement.

Original languageEnglish
Pages (from-to)126-132
Number of pages7
JournalBehavioural Processes
Issue number2
StatePublished - Oct 2009

Bibliographical note

Funding Information:
This research was supported by National Institute of Mental Health Grant 63726.


  • Partial reinforcement
  • Pigeons
  • Simultaneous discrimination
  • Within-trial contrast
  • Work ethic

ASJC Scopus subject areas

  • Animal Science and Zoology
  • Behavioral Neuroscience


Dive into the research topics of 'Within-trial contrast: The effect of probability of reinforcement in training'. Together they form a unique fingerprint.

Cite this