top of page

Dehumanisation of victims (Albert Bandura, Stanford University)

Bandura part1.png
  • Participant is told that there are some learners in the other room.

  • The learners will be asked to answer a series of questions. If they answer incorrectly, the participant will be instructed via the receiver to deliver some punishment through a lever.  

  • The participant gets to decide how much punishment he/she wishes to mete out. 

  • Just before the experiment begins, the receiver is "accidentally" left on, and the participant can hear the experimenters "gossiping" about the learners:

    • Scenario 1: the instructors call the learners " a real bunch of animals".​

    • Scenario 2: the instructors call the learners " a nice bunch of people".

    • Scenario 3: nothing is said



  • Even though the "learners' were actually accomplices that made the same number of errors at the same points of time in all 3 scenarios (horizontal axis), the reactions of the participants varied dramatically (vertical axis)

  • The extent of punishment directly correlated with what the examiner said before the experiment:

    • The humanised group (the bunch of nice guys) were punished less than the control group(nothing is said), who was, in turn, punished less than the dehumanised group (a bunch of animals).

  • This difference in punishment happens even when all 3 groups performed exactly the same.

  • Just one line of description is sufficient for us to formulate a story, which then affects our actions and decisions, even if this does not match reality. 

bottom of page